eLife Assessment

This important study shows that retinal bipolar cell subtype-specific differences in the size of synaptic ribbon-associated vesicle pools contribute to the transient versus sustained kinetics of the responses of retinal ganglion cells. The data are extensive and compelling. This work will be of broad interest to researchers working on synaptic transmission, retinal signal processing, and sensory neurobiology.

Reviewer #1 (Public review):

Summary:

In the retina, parallel processing of cone photoreceptor output under bright light conditions dissects critical features of our visual environment, and fundamental to visual function. Cone photoreceptor signals are sampled by several types of bipolar cells and passed onto the ganglion cells. At the output of retinal processing, retinal ganglion cells send about 40 different codes of the visual scene to the brain for further processing. In this study, the authors focus on whether subtype-specific differences in the size of synaptic ribbon-associated vesicle pools of bipolar cells contribute to different retinal ganglion cell (RGC) responses.

Specifically, inputs to ON alpha RGCs producing transient versus sustained kinetics (ON-S vs. ON-T, respectively) are compared. The authors first demonstrate that ON-S vs. ON-T RGCs are readily identifiable in a whole mount preparation and respond differently to both static and to a spatially uniform, randomly fluctuating (Gaussian noise) light stimulus. Liner-nonlinear (LN) models were used to estimate the transformation between visual input and excitatory synaptic input for each RGCs; these models suggested the presence of transient versus sustained kinetics already in the excitatory inputs to ON-T and ON-S RGCs.

Indeed, the authors show that (glutamatergic) excitatory inputs to ON-S vs. ON-T RGCs are of distinct kinetics. The subtypes of bipolar cells providing input to ON-S are known (i.e., type 6 and 7), but the source of excitatory bipolar inputs to ON-T RGCs needed to be determined. In a tedious process, it is elegantly shown here that ON-T RGCs receive most of their excitatory inputs from type 5 and 6 bipolars. Interestingly, the temporal properties of light-evoked responses of type 5, 6 and 7 bipolars recorded from the somas were indistinguishable and rather sustained, suggesting that the origin of transient kinetics of excitatory inputs to ON-T RGCs suggested by the LN model might be found in the processing of visual signals at the bipolar cell axon terminal. Blocking GABA- or glycinergic inhibitory inputs did not alter the light-evoked excitatory input kinetics to ON-T and ON-S RGCs. Two-photon glutamate sensor imaging revealed significantly faster kinetics of light-evoked glutamate signals at ON-T versus ON-S RGCs, and that differences in glutamate release from presynaptic bipolar cells are retained without amacrine feedback to bipolar cells. Detailed EM analysis of bipolar cell ribbon synapses onto ON-T and ON-S RGCs revealed fewer ribbon-associated vesicles at ON-T synapses, that is consistent with stronger paired-flash depression of light-evoked excitatory currents in ON-T RGCS versus ON-S RGCs. This study suggests that bipolar subtype-specific differences in the size of synaptic ribbon-associated vesicle pools contributes to transient versus sustained kinetics in RGCs.

Strengths:

The use of multiple, state-of-the-art tools and approaches to address the kinetics of bipolar to ganglion cell synapse in an identified circuit.

Reviewer #2 (Public review):

Summary:

Goal of the study. The authors tried to pinpoint the origins of transient and sustained responses measured at retinal ganglion cells (rgcs), which is the output layer of the retina. Response characteristics of rgcs are used to group them into different types. The diversity of rgc types represents the ability of the retina to transform visual inputs into distinct output channels. They find that the physical dimensions of bipolar cell's synaptic ribbons (specialized release sites/active zones) vary across the different types of cone on-bpcs, in ways that they argue could facilitate transient or sustained release. This diversity of release output is what they argue underlies the differences in on-rgcs response characteristics, and ultimately represents a mechanism for creating parallel cone-driven channels.

Strengths:

The major strengths of the study are the anatomical approaches employed and the use of the "glutamate sniffer" to assay synaptic glutamate levels. The outline of the study is elegant and reflects the strengths of the authors.

Comments on revised version:

The authors have addressed my comments either through new experiments and/or with additional citations.

Explanation of the studies significance. I think the study provides a solid set of data, acquired through exceptional methodologies, and delivers a compelling hypothesis. This is an exceptionally talented group of systems level thinkers and experimentalists, who are now pointing to smaller scale biophysical principles of synaptic transmission.

Reviewer #3 (Public review):

Summary:

Different types of retinal ganglion cell (RGC) have different temporal properties - most prominently a distinction between sustained vs. transient responses to contrast. This has been well established in multiple species, including mouse. In general, RGCs with dendrites that stratify close to the ganglion cell layer (GCL) are sustained; whereas those that stratify near the middle of the inner plexiform layer (IPL) are transient. This difference in RGC spiking responses aligns with similar differences in excitatory synaptic currents as well as with differences in glutamate release in the respective layers - shown previously and here, with a glutamate sensor (iGluSnFR) expressed in the RGCs of interest. Differences in glutamate release were not explained by differences in the distinct presynaptic bipolar cells' voltage responses, which were quite similar to one another. Rather, the difference in transient vs. sustained responses seems to emerge at the bipolar cell axon terminals in the form of glutamate release. This difference in the temporal pattern of glutamate release was correlated with differences in the size of synaptic ribbons (larger in the bipolar cells with more sustained responses), which also correlated with a greater number of vesicles in the vicinity of the larger ribbons.

The main conclusion of the study relates to a correlation (because it is difficult to manipulate ribbon size or vesicle density experimentally): the bipolar cells with increased ribbon size/vesicle number would have a greater possibility of sustained release, which would be reflected in the postsynaptic RGC synaptic currents and RGC firing rates. This model proposes a mechanism for temporal channels that is independent of synaptic inhibition. Indeed, some experiments in the paper suggest that inhibition cannot explain the transient nature of glutamate release onto one of the RGC types. Still, it is surprising that such a diverse set of inhibitory interneurons in the retina would not play some role in diversifying the temporal properties of RGC responses.

Strengths:

(1) The study uses a systematic approach to evaluating temporal properties of retinal ganglion cell (RGC) spiking outputs, excitatory synaptic inputs, presynaptic voltage responses, and presynaptic glutamate release. The combination of these experiments demonstrates an important step in the conversion from voltage to glutamate release in shaping response dynamics in RGCs.

(2) The study uses a combination of electrophysiology, two-photon imaging and scanning block face EM to build a quantitative and coherent story about specific retinal circuits and their functional properties.

Weaknesses:

(1) There were some interesting aspects of the study that were not completely resolved, and resolving some of these issues may go beyond the current study. For example, it was interesting that different extracellular media (Ames medium vs. ACSF) generated different degrees of transient vs. sustained responses in RGCs, but it was unclear how these media might have impacted ion channels at different levels of the circuit that could explain the effects on temporal tuning.

(2) It was surprising that inhibition played such a small role in generating temporal tuning. The authors explored this further in the revision, which supported the original claim that inhibition plays a minor role in glutamate release dynamics from the bipolar cells under study.

Author response:

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

Reviewer #1 (Public Review): 

Summary: 

In the retina, parallel processing of cone photoreceptor output under bright light conditions dissects critical features of our visual environment and is fundamental to visual function. Cone photoreceptor signals are sampled by several types of bipolar cells and passed onto the ganglion cells. At the output of retinal processing, retinal ganglion cells send about 40 different codes of the visual scene to the brain for further processing. In this study, the authors focus on whether subtype-specific differences in the size of synaptic ribbon-associated vesicle pools of bipolar cells contribute to different retinal ganglion cell (RGC) responses. Specifically, inputs to ON alpha RGCs producing transient versus sustained kinetics (ON-S vs. ON-T, respectively) are compared. The authors first demonstrate that ON-S vs. ON-T RGCs are readily identifiable in a whole mount preparation and respond differently to both static and to a spatially uniform, randomly fluctuating (Gaussian noise) light stimulus. Liner-nonlinear (LN) models were used to estimate the transformation between visual input and excitatory synaptic input for each RGCs; these models suggested the presence of transient versus sustained kinetics already in the excitatory inputs to ON-T and ON-S RGCs. Indeed, the authors show that (glutamatergic) excitatory inputs to ON-S vs. ON-T RGCs are of distinct kinetics. The subtypes of bipolar cells providing input to ON-S are known (i.e., type 6 and 7), but the source of excitatory bipolar inputs to ON-T RGCs needed to be determined. In a tedious process, it is elegantly shown here that ON-T RGCs receive most of their excitatory inputs from type 5 and 6 bipolars. Interestingly, the temporal properties of light-evoked responses of type 5, 6, and 7 bipolars recorded from the somas were indistinguishable and rather sustained, suggesting that the origin of transient kinetics of excitatory inputs to ON-T RGCs suggested by the LN model might be found in the processing of visual signals at the bipolar cell axon terminal. Blocking GABA- or glycinergic inhibitory inputs did not alter the light-evoked excitatory input kinetics to ON-T and ON-S RGCs. Twophoton glutamate sensor imaging revealed significantly faster kinetics of light-evoked glutamate signals at ON-T versus ON-S RGCs. Detailed EM analysis of bipolar cell ribbon synapses onto ON-T and ON-S RGCs revealed fewer ribbon-associated vesicles at ON-T synapses, which is consistent with stronger paired-flash depression of lightevoked excitatory currents in ON-T RGCS versus ON-S RGCs. This study suggests that bipolar subtype-specific differences in the size of synaptic ribbon-associated vesicle pools contribute to transient versus sustained kinetics in RGCs. 

Strengths: 

The use of multiple, state-of-the-art tools and approaches to address the kinetics of bipolar to ganglion cell synapse in an identified circuit. 

Weaknesses: 

For the most part, the data in the paper support the conclusions, and the authors were careful to try to address questions in multiple ways. Two-photon glutamate sensor imaging experiment showing that blocking GABA- and glycinergic inhibition does not change the kinetics of light-evoked glutamate signals at ON-T RGCs would strengthen the conclusion that bipolar subtype-specific differences in the size of synaptic ribbon-associated vesicle pools contribute to transient versus sustained kinetics in RGCs. 

Thank you for this suggestion. We have revised the text throughout to be careful not to imply that amacrine cells have no role in shaping EPSCs and spike output, but instead that the transience of the On-T responses persists without amacrine cells (see for example lines 91, 450-453, 514-518, 696-714). We have also added additional iGluSnFR experiments to the paper to further test this conclusion (new Figure 7). The new data shows that the transience of glutamate release from the On-T cells is retained when 1) spiking amacrine cell activity is suppressed by blocking voltage-gated Na⁺ channels with TTX or 2) all amacrine cell activity is suppressed by blocking AMPA receptors with NBQX. This does provide nice additional evidence that amacrine cells are not necessary for the sustained/transient distinction.

Reviewer #2 (Public Review): 

Summary: 

Goal of the study. The authors tried to pinpoint the origins of transient and sustained responses measured at retinal ganglion cells (rgcs), which is the output layer of the retina. Response characteristics of rgcs are used to group them into different types. The diversity of rgc types represents the ability of the retina to transform visual inputs into distinct output channels. They find that the physical dimensions of bipolar cell's synaptic ribbons (specialized release sites/active zones) vary across the different types of cone on-bpcs, in ways that they argue could facilitate transient or sustained release. This diversity of release output is what they argue underlies the differences in on-rgcs response characteristics, and ultimately represents a mechanism for creating parallel cone-driven channels. 

Strengths: 

The major strengths of the study are the anatomical approaches employed and the use of the "glutamate sniffer" to assay synaptic glutamate levels. The outline of the study is elegant and reflects the strengths of the authors. 

Weaknesses: 

The major weakness is that the ambitious outline is not matched with a complete set of results, and the set of physiological protocols is disjointed, not sufficient to bridge the systems-level question with the presynaptic release question. 

Thank you for this comment as it provides an opportunity (here and in the paper) for us to clarify our main goal. We wanted to link the well-established distinction between transient and sustained retinal responses to anatomy. This required locating where this difference arises within the circuitry – which we show to be at least largely the bipolar output synapse – and then examining the structure of this synapse in detail. While we would certainly be interested in connecting our results to a biophysical description of the synapse, that was not the primary focus of our study and was not something we could add without substantial additional work.  

Major comments on the results and suggestions. 

The ribbon model of release has been explored for decades and needs to be further adapted to systems-level work. The study under consideration by Kuo et al. takes on this task. Unfortunately, the experimental design does not permit a level of control over presynaptic/bpc behavior that is comparable to earlier studies, nor do they manipulate release in ways that test the ribbon model (i.e., paired recordings or Ribeye-ko). Furthermore, the data needs additional evaluation, and the presentation and interpretations should draw on published biophysical and molecular studies. 

As described above, our goal was to test several possible explanations for the difference between transient and sustained responses in OnT and OnS ganglion cells: (1) differences in the light responses of the bipolar cells that convey photoreceptor signals to the relevant ganglion cells; (2) shaping of bipolar transmitter release by presynaptic inhibition; (3) shaping of ganglion cell responses by postsynaptic inhibition or spike generation; (4) differences in feedforward bipolar synapses. We were surprised to find that the feedforward bipolar synapses play a central role in this difference, and your comment nicely prompts us to relate this to the large literature on biophysical studies of release from ribbon synapses. We have made substantial revisions in the text to do this. This includes anticipating the importance of feedforward synaptic properties in the abstract and introduction (lines 36-37 and 61-64), pointers in the results (lines 539-548), and several new paragraphs in the discussion (starting on lines 751, 773 and 787). By showing that the transient/sustained differences originates largely at feedforward bipolar synapses, we set the stage for future work that shows how biophysical properties of the synapse shape physiological signals that traverse it.

To build a ribbon-centric context, consider recent literature that supports the assertion that ribbons play a role in forming AZ release sites and facilitating exocytosis. Reference Ribeye-ko studies. For example, ribbonless bpcs show an 80% reduction in release (Maxeiner et al EMBO J 2016), the ribbonless retina exhibits signaling deficits at the output layer (Okawa et al ...Rieke, ..Wong Nat Comm 2019), and ribbonless rods show an 80% reduction the readily releasable pool (RRP) of SVs (Grabner Moser, elife 2021). In addition, the authors could refer to whole-cell membrane capacitance studies on mammalian rods, cones, and bpcs, because the size of the RRP of SVs scales with the dimensions and numbers of ribbons (total ribbon footprint). For comparison, bipolars see the review by Wan and Heidelberger 2011. For a comparison of mammalian rods and cones, see, rods: Grabner and Moser (2021 eLife), Mueller.. Regus Leidig et al. (2019; J Neurosci) and cones Grabner ...DeVries (Nat Comm 2023). A comparison of cell types shows that the extent of release is (1) proportional to the total size of the ribbon footprint, and (2) less release is witnessed when ribbons are deleted (also see photo ablation studies by Snellman.... And Mehta..Zenisek, Nat Neurosci and Neuron).

Thank you for these pointers into the literature.  We have included much of this work in the revised Discussion (see three paragraphs starting on line 751). The revised text focuses on the evidence that larger and more numerous ribbons lead to increased release. The direct evidence from previous work for this relationship supports our (indirect) conclusions in the current paper about the role of ribbon size and associated vesicle pools in transient vs sustained responses.  

Ribbon morphology may change in an activity-dependent manner. The rod ribbon AZ has been reported to lengthen in the dark (Dembla et al 2020), and deletion of the ribbon shortens the length of the AZ (defined by Cav1,4 or RIM2); in addition, the Ribeye-ko AZs fail to change in size with light and dark conditioning. Furthermore, EM studies on rod and cone AZs in light and dark argue that the number of SVs at the base of the ribbon increases in the dark, when PRs are depolarized (see Figure 10, Babai et al 2016 JNeurosci). Lastly, using goldfish Mb1 on-bipolars, Hull et al (2006, J Neurophysio) correlated an increase in release efficiency with an increase in ribbon numbers, which accompanied daylight. >> When release activity is high, ribbon AZ length increases (Dembla, rods), the number of docked SVs increases (Babai, rods cones), and the number of ribbons increases (Hull, diurnal Mb1s). 

We have extensively revised the discussion section to include more discussion of ribbons, particularly emphasizing evidence supporting the general argument that larger ribbons support higher release rates. We focused on studies that provided direct links between release rates and ribbon size or number of ribbon-associated vesicles.  This includes studies that pair electrophysiology and anatomy and those that measure the consequences of ablating ribbons,

The results under review, Kuo et al., were attained with SBF-SEM, which has the benefit of addressing large-volume questions as required here, yet it achieves lower spatial resolution than what is attained with TEM tomography and FIB-EM. Ideally, the EM description would include SV size, and the density of ribbon-tethered SVs that are docked at the plasma membrane, because this is where the SVs fuse (additional non-ribbon release sites may also exist? Mehta ... Singer 2014 J Neurosci). Studies by Graydon et al 2011 and 2014 (both in J Neurosci), and Jean ... Moser et al 2018 (eLife) are good examples of quantitative estimates of SVs docking sites at ribbons. SBF-SEM does not allow for an assessment of SVs within 5 nm of the PM, but if the authors can identify the number of SVs that appear within the limit of resolution (10 to 15 nm) from the PM, then this data would be useful. Also, what dimension(s) of the large ribbons make them larger? Typically, ribbons are fixed in height (at least in the outer retina, 200 to 250 nm), but their length varies and the number ribbons per terminal varies. Is the larger ribbon size observed in type 6 bpcs do to longer ribbons, or taller ribbons? A longer ribbon likely has more docked SVs. An additional possibility is that more SVs are about the ribbon-PM footprint, either more densely packed and/or expanding laterally (see definitions in Jean....Moser, elife 2018). 

We have included an additional analysis of ribbon surface area from our 3D SBFSEM reconstructions. As with the volume measurements included in the original submission, ribbon surface areas are distinct between type 5i and type 6 bipolar cells (Fig. S10A), ON-T RGCs on average receive input from ribbons with smaller surface area than ON-S RGCs (Fig. S10B), and ribbon surface area predicts the number of adjacent vesicles across bipolar cell types (Fig. S10C).  We agree that a higher resolution view of presynaptic structures would be very helpful, but the resolution of our SBF-SEM data is limited (e.g. each pixel is 40 nm on a side).  This resolution does not allow us to distinguish between vesicles at vs near the membrane. 

In our observations, both length and height of the ribbons showed variability across individual bipolar cells. And ribbons in type 6 bipolar cells tended to be either longer and/or taller compared to those in type 5 cells. We agree that a longer ribbon may accommodate more docked SVs. A more definitive analysis would benefit from higher-resolution, isotropic 3D reconstructions of ribbons, which would allow more precise shape analysis and ,together with a detailed assessment of docked SVs at the ribbons.

The ribbon literature given above makes the argument that ribbons increase exocytotic output, and morphological studies suggest that release activity enhances 1) ribbon length (Dembla) and 2) the density of SVs near the PM (Babai). These findings could lead one to propose that type 6 bpcs (inputs to On-sustained) are more active than type 5i (feed into On-transient). Here Kuo et al. show that the bpcs have similar Vm (measured from the soma) in response to light stimulation. Does Vm predict release? Not entirely as the authors acknowledge, because: Cav channel properties, SV availability, and negative feedback are all downstream of bpc Vm. The only experiment performed to test downstream factors focused on negative feedback from amacrines. The data presented in Figures 5C-F led me to conclude the opposite of what the authors concluded. My impression is that the T-ON rgc exhibits strong disinhibition when GABA-blockers are applied (the initial phase is greatly increased in amplitude and broadened with the drug), which contrasts with the S-On rgc responses that show a change in the amplitude of the initial phase but not its width (taus would be nice). Here and in many places the authors refer to changes in release kinetics, without implementing a useful description of kinetics. For instance, take the cumulative current (charge) in Figure 5C and fit the control and drug traces to arrive at taus, and their respective amplitudes, and use these values to describe kinetic phases. One final point, the summary in Figure 5D has a p: 0.06, very close to the cutoff for significance, which begs for more than an n = 5. Given that previous studies have shown that bpc output is shaped by immediate msec GABA feedback, in ways that influence kinetic phases of release (..Mb1 bipolars, see Vigh et al 2005 Neuron), more attention to this matter is needed before the authors rule out feedback inhibition in favor of ribbon size. If by chance, type 5i bpcs are under uniquely strong feedback inhibition, then ribbon size may result from less activity, not less output resulting from smaller ribbons.

The text surrounding Figure 5 led to some confusion, and we have revised that text and the figure for clarity.  First, the data in that figure is entirely from On-T cells (the upper and lower panels show block of GABA and glycine receptors separately).  Second, the observation that we make there is that block of inhibitory receptors increases the transience of the On-T excitatory input, rather than decreasing it as would be expected if the transience is created by presynaptic inhibition. We have added additional data and that increase in transience is now significant. Inhibitory block does substantially increase the amplitude of the postsynaptic response, and a likely origin of this change in response is inhibitory feedback to the bipolar synaptic terminal. We now indicate this in the text on page 13, lines 438-453. 

The key result of this figure for our purposes here is that the transience of the excitatory input to the OffT cell remains with inhibitory input blocked. We have clarified throughout the text that our results indicate that inhibitory feedback is not necessary for the difference between transient release into On-T and sustained release onto On-S. This does not mean that inhibitory feedback does not shape the responses in other ways or contribute to the transient/sustained difference - just that for the specific stimuli we use that difference is retained without presynaptic inhibition. We have also added citations to past work showing that activity of amacrine cells can modulate bipolar transmitter release. 

Whether strong feedback inhibition limits activity and therefore limits ribbon size in an activity-dependent way is an intriguing possibility. Indeed, addressing why ribbons are larger in type 6 bipolar cells vs. other bipolar types will be an interesting avenue of further study. However, it would be surprising if ribbon sizes changed during the acute pharmacological block conditions (~10-15 minutes) we employed in our study. Our point here is that there is an interesting correlation between presynaptic ribbon size and the kinetics of glutamate release. We do not think that the two possibilities stated in the last sentence (“…ribbon size may result from less activity, not less output resulting from smaller ribbons”) are mutually exclusive.

We have not further quantified the response kinetics in the experiments of Figure 5 as the large changes induced by the pharmacology (especially GABA receptor block) make it unclear how to interpret quantitative differences.  In other places we have quantified kinetics through the STA or specified that our focus was more qualitative (i.e. transient vs sustained kinetics). 

As mentioned above, the behavior of Cav channels is important here. This is difficult to address with voltage clamps from the soma, especially in the Vm range relevant to this study. Given that it has previously been modeled that the rod bpc to AII pathway adapts to prolonged depolarization of rbcs through downregulating Cav channel-mediated Ca²⁺ influx (Grimes ....Rieke 2014 Neuron), it seems important for Kou et al to test if there is a difference in Cav regulation between type 6 and 5i bpcs. Ca²⁺  imaging with a GCaMP strategy (Baden....Lagnado Current Biology, 2011) or filling the presynapse with Ca dyes (see inner hair cells: Ozcete and Moser, EMBO J 2020) would allow for the correlation of [Ca]intra with GluSnf signals (both local readouts).

This is a good suggestion but is outside the scope of our current paper. Our focus was on the circuit origin of the difference in response of the OnT and OnS responses rather than the specific biophysical mechanism.  We are of course interested in the mechanism, but the additional experiments needed to pin that down would need to be a part of future experiments. The work here represents an important step in that direction as it greatly reduces the number of possible locations and mechanisms for the sustained/transient difference and hence serves to focus any future mechanistic investigations.

Stimulation protocol and presentation of Glutamate Sniffer data in Figure 6. In all of your figures where you state steady st as a % of pk amplitude, please indicate in the figure where you estimate steady state. Alternatively, if you take the cumulative dF/F signal, then you can fit the different kinetic phases. From the appearance of the data, the Sustained Glu signals look like square waves (Figure 6B ROI1-4), without a transient at onset, which is not predicted in your ribbon model that assumes different kinetic phases (1. depletion of docked SVs, and 2. refilling and repriming). The Transient responses (Figure 6B ROI5-8) are transient and more compatible with a depressing ribbon scheme. If you take the cumulative, for all of the On-S and compare it to all of the On-T responses, my guess is the cumulative dF/F will be 10 to 20 larger for the S-On. Would you conclude that bpc inputs to On-S (type 6) release 20fold more SVs per 4 seconds on a per ribbon basis, and does the surface area of the type 6 bpcs account for this difference? From Figures 8B and D, the volume of the ribbon is ~2 fold greater for type 6 vs 5i, but the Surface Area (both faces of ribbon) is more relevant to your model that claims ribbon size is the pivotal factor. If making cumulative traces, and comparisons on an absolute scale is unfounded, then we need to know how to compare different observations. The classic ribbon models always have a conversion factor such as the capacitance of an SV or q size that is used to derive SV numbers from total dCm or Qcontent. See Kim ....et al von Gersdorff, 2023, Cell Reports. Why not use the Gaussian noise stimulus in Fig 6 as in Figure 1 and 2? 

For iGluSnFR recordings, steady-state responses were measured from the mean fluorescence over the last 1 sec of the light step (2 sec duration) response. We have included this information in the figure caption and in the Methods. 

There is a good deal of variability in the iGluSnR responses from one ROI to another, and the ROIs shown in the original submission had a less prominent transient component than many other ROIs. We have replaced this figure with another that is more representative of the average behavior across ROIs. The full range of behavior is captured in Figure 6C; it is clear across ROIs that glutamate release near ON-S dendrites shows both sustained and transient components. The new experiments in which we block amacrine cell activity also include a few more example ROIs from ON-S cells, and those also show both transient and sustained components.

Your suggestion to integrate the iGluSnFR signals to compare to our structural analysis of ribbons is interesting. However, we are hesitant to make a quantitative comparison between the two without further experiments to validate how the iGluSnFR signals we measure relate to release of single vesicles. For example, a quantitative measure of release based on the iGluSnR experiments would require accounting for possible differences in the expression of the indicator - which could differ both in overall level and/or location relative to release sites. 

This comment and one above highlight the importance of measures of ribbon surface area, which we now provide (Figure S10).

Figure 7. What is the recovery time for mammalian cones derived from ribbon-based models? There are estimates from membrane capacitance studies. Ground squirrel cones take 0.7 to 1 sec to recover the ultrafast, primed pool of SVs when probed with a paired-pulse protocol (Grabner ...DeVries 2016, Neuron). Their off-bpcs take anywhere from under 0.2 sec to a second to recover, which is a combination of many synaptic factors (Grabner ...DeVries Nat Comm 2023). Rod On bpcs take over a second (Singer Diamond 2006, reviewed Wan and Heidelberger 2011). In Figure 7B, the recovery time is ~150 ms for the responses measured at rgcs. This brief recovery time is incompatible with existing ribbon models of release. Whole-cell membrane capacitance measurements would be helpful here.

Thanks for drawing our attention to this issue. Indeed, we see a relatively rapid recovery in the paired-flash experiments. We now discuss this recovery time in the context of past measurements of recovery of responses in cones and bipolar cells (paragraph starting on line 773). There are many factors that could contribute to the relatively rapid recovery we observe - including synaptic factors such as those highlighted by Grabner et al., (2016) either at the cone-to-bipolar synapses or the bipolar-to-RGC synapses. We are certainly interested in a more detailed understanding of this issue, but the additional experiments are outside the scope of this paper.  

Experimental Suggestion: Add GABA blockers and see if type 5i bpc responds with more release (GluSniff) and prolonged [Ca2+] intra (GCaMP). Compare this to type 6 bpc behavior with GABA/gly blockers. This will rule in or out whether feedback inhibition is involved. 

Figure 7 in the revised manuscript includes two new experiments examining glutamate release (without the simultaneous measurement of bipolar cell intracellular calcium) while blocking (1) all/most amacrine cell-mediated inhibition via inclusion of NBQX in the bath solution, and (2) blocking spiking amacrine cells via inclusion of TTX in the bath solution. The transient vs sustained difference in light-evoked glutamate release around ON-T and ON-S RGC dendrites remained with amacrine activity suppressed. These new results are consistent with the anatomical and pharmacological data that were included in the initial submission of the manuscript (Fig. 5) that indicate presynaptic inhibition does not have a major role in shaping release kinetics at these synapses. 

Reviewer #3 (Public Review): 

Summary: 

Different types of retinal ganglion cell (RGC) have different temporal properties - most prominently a distinction between sustained vs. transient responses to contrast. This has been well established in multiple species, including mice. In general, RGCs with dendrites that stratify close to the ganglion cell layer (GCL) are sustained; whereas those that stratify near the middle of the inner plexiform layer (IPL) are transient. This difference in RGC spiking responses aligns with similar differences in excitatory synaptic currents as well as with differences in glutamate release in the respective layers - shown previously and here, with a glutamate sensor (iGluSnFR) expressed in the RGCs of interest. Differences in glutamate release were not explained by differences in the distinct presynaptic bipolar cells' voltage responses, which were quite similar to one another. Rather, the difference in transient vs. sustained responses seems to emerge at the bipolar cell axon terminals in the form of glutamate release. This difference in the temporal pattern of glutamate release was correlated with differences in the size of synaptic ribbons (larger in the bipolar cells with more sustained responses), which also correlated with a greater number of vesicles in the vicinity of the larger ribbons. 

The main conclusion of the study relates to a correlation (because it is difficult to manipulate ribbon size or vesicle density experimentally): the bipolar cells with increased ribbon size/vesicle number would have a greater possibility of sustained release, which would be reflected in the postsynaptic RGC synaptic currents and RGC firing rates. This model proposes a mechanism for temporal channels that is independent of synaptic inhibition. Indeed, some experiments in the paper suggest that inhibition cannot explain the transient nature of glutamate release onto one of the RGC types. Still, it is surprising that such a diverse set of inhibitory interneurons in the retina would not play some role in diversifying the temporal properties of RGC responses. 

Strengths: 

(1) The study uses a systematic approach to evaluating temporal properties of retinal ganglion cell (RGC) spiking outputs, excitatory synaptic inputs, presynaptic voltage responses, and presynaptic glutamate release. The combination of these experiments demonstrates an important step in the conversion from voltage to glutamate release in shaping response dynamics in RGCs. 

(2) The study uses a combination of electrophysiology, two-photon imaging, and scanning block-face EM to build a quantitative and coherent story about specific retinal circuits and their functional properties. 

Weaknesses: 

(1) There were some interesting aspects of the study that were not completely resolved, and resolving some of these issues may go beyond the current study. For example, it was interesting that different extracellular media (Ames medium vs. ACSF) generated different degrees of transient vs. sustained responses in RGCs, but it was unclear how these media might have impacted ion channels at different levels of the circuit that could explain the effects on temporal tuning.

We do not have an explanation for the quantitative differences in response kinetics we observed in Ames’ medium vs. ACSF. There are modest differences in calcium and magnesium concentration and a larger difference in potassium (2.5 mM in ACSF vs 3.6 mM in Ames). It would be interesting to test which of these (or other) differences accounts for the difference in response kinetics.

(2) It was surprising that inhibition played such a small role in generating temporal tuning. At the same time, there were some gaps in the investigation of inhibition (e.g., IPSCs were not measured in either of the RGC types; pharmacology was used to investigate responses only in the transient RGCs).

We were also surprised at this result. We have included additional data on inhibition in the revised manuscript. Figure S3 shows light-evoked IPSC data from both RGC types (Fig. S3) and Fig. 7 shows additional iGluSnFR measurements around both ON-T and ON-S RGC dendrites with inhibition blocked via bath application of NBQX (Fig. 7) and separately with inhibition from spiking amacrine cells blocked with TTX. These experiments provide additional evidence for the small role of inhibition. We attempted to measure the kinetics of excitatory input to ON-S cells with inhibition blocked, but we found that the excitatory input showed strong spontaneous oscillations under these conditions and the light responses were changed so drastically that we did not feel we could make a clear comparison with control conditions.

(3) There could be additional discussion and references to the literature describing several topics, including: temporal dynamics of glutamate release at different levels of the IPL; previous evidence that release sites from a single presynaptic neuron can differ in their temporal properties depending on the postsynaptic target; previous investigations of the role of inhibition in temporal tuning within retinal circuitry. 

Thanks, we have included more discussion and references to the relevant literature as you have suggested in the recommendations to authors.

Reviewer #1 (Recommendations For The Authors): 

The presented raw data of the pharmacological experiments show that SR95531 and TPMPA robustly increased both the amplitude and duration of the transient component of the light step-evoked excitatory currents, with slight, if any enhancement of the sustained component in ON-T RGCs Figure 5C. Statistical analysis of the population data (n=5) with Wilcoxon signed rank test yielded no significant difference (ln 363). However, reanalyzing the data extracted from the graph (Figure 5D) revealed that the difference between the paired observations is normally distributed (Shapiro-Wilk normality test, P=0.48) allowing parametric statistics to be used, which provides higher statistical power. Accordingly, reanalyzing the presented data with paired Student's t-test data revealed significant differences (P=0.01) in the steady-state amplitude normalized to that of the peak, recorded in the presence of SR95531 and TPMPA. In other words, based on the (rough) analysis of the presented pharmacology data GABAergic feedback inhibition significantly contributes to shaping the transient portion of the light-evoked excitatory currents in ON-T RGCs, by making it more transient. I believe a similar analysis based on the actual data is necessary, and the results should be communicated either way. However, if warranted, two-photon glutamate sensor imaging experiments showing that blocking GABA- and glycinergic inhibition does not change the kinetics of light-evoked glutamate signals at ON-T RGCs should also be performed, as these would be critical in drawing a conclusion regarding the effect of feedback inhibition on glutamate release from bipolar cells.

Thanks for this feedback. We have added another cell to the data set in Fig. 5D. With this addition, SR95531/TPMPA application significantly increases the response transience of excitatory currents measured in ON-T RGCs compared to control. This enhanced transience in GABA_A/C receptor blockers is due to an increase in the amplitude of the initial peak component of the response (control peak amplitude: -833.7±103.3 pA; SR95531+TPMPA peak amplitude: 2023±372.7pA; p=0.03, Wilcoxon signed rank test), with no change to the later sustained component (control plateau amplitude: -200.7±14.71pA; SR95531+TPMPA plateau amplitude: -290.9±43.69pA; p=0.15, Wilcoxon signed rank test).

We should clarify that this result indicates that GABAergic inhibition makes the excitatory inputs to ON-T RGCs less transient. Block of GABA receptors increased transience, thus intact GABAergic transmission appears to limit the initial peak of the response and therefore make excitatory currents more sustained. We unfortunately were not able to examine whether sustained excitatory currents in ON-S RGCs would become more transient using the same approach. In our hands, bath application of SR95531+TPMPA led to the generation of large-amplitude (>1nA) oscillatory bursts of excitatory input that developed within 5 minutes and persisted for the duration of the incubation (up to ~30 min) in drugs. Further, presentation of light steps tended to induce variable amplitude responses, likely dependent on the presence of spontaneous bursts; when large amplitude responses were evoked, these typically oscillated for several seconds after the step.

To examine a potential role for presynaptic inhibition in transient vs. sustained bipolar cell output, we therefore chose to eliminate amacrine cell-mediated inhibition by bath application of the AMPA/kainate receptor antagonist NBQX in additional iGluSnFR measurements. This manipulation should leave ON bipolar cell responses intact while eliminating most amacrine cell-mediated responses (and OFF bipolar cell driven responses). In separate experiments, we also eliminated inhibition from spiking amacrine cells by bath application of TTX. As shown in new Fig. 7, sustained and transient responses persisted in distal versus proximal RGC dendrites, respectively. Compared to SR95531/TPMPA, bath application of NBQX was not associated with spontaneous bursts of glutamate release around ON-S dendrites. These results show that amacrine cell-mediated inhibition is not required for either sustained or transient glutamate release from bipolar cells that provide input to ON-S and ON-T RGCs.

Small points: 

(1) The legend of Figure 1 (D) refers to shaded areas to show {plus minus} SEM, but no shade is visible (at least in my printout).

The SEM shading is there in Fig. 1D but is mostly obscured by the mean lines for the respective RGC types. We have added this to the figure caption.

(2) I found the reported Vrest for the ON bipolar cells somewhat depolarized. Perhaps due to the uncompensated junction potentials? 

These measurements are indeed not corrected for the liquid junction potential (which is approximately -10.8 mV between K-gluconate internal and Ames’ solution). We did not apply this correction since the appropriate value is not clear in perforated patch recordings as the intracellular chloride concentration is unknown (and can differ from that in the pipette solution). We have clarified this in the results text where we describe the Vrest values (lines 335-338).

(3) It is Wilcoxon signed rank test, not Wilcoxan. 

Thanks for catching this. This has been corrected in the revised manuscript.

Reviewer #2 (Recommendations For The Authors): 

Some amacrines express vesicular Glut-3 transporter and are reported to release glutamate (Marshak, Vis Neurosci 2016). Are Amacrine vGlut3 signals postsynaptic (within ~0.5 um) to cone bpc ribbons?

We did not characterize VgluT3-expressing amacrine cells in our SEM datasets. A recent study by Friedrichson et al. (Nat. Comm. 2024; PMID 38580652) using 3D SEM reconstructions found that Vglut3-amacrines are postsynaptic to both type 5i and type 6 bipolar cells, as well as other type 5/xbc bipolar cells (and receive >50% of their input from type 3a OFF bipolar cells).

How far apart are the postsynaptic targets from the ribbon release sites? The ribbons at type 5i bpc/On-T input appear separated from the dendrites of On-T rgcs (Figure 8C). At least further away than the type 6 bpc ribbons are from On-S rgc dendrites (Figure 8C). Distance may create a thresholding phenomenon, whereby only multivesicular bouts at the onset of depolarization are able to elevate synaptic Glu to levels needed to activate On-T GluRs. See Grabner et al Nat Comm 2023 for such scenarios in the outer retina.

This is an intriguing possibility, but we should point out that the presynaptic ribbons in Fig. 9C (former Fig. 8C) are similar distances (within the resolution of our reconstructions) from the ON-T and ON-S dendrites. We have increased the brightness of the dendrite segments for both RGC types in the resubmission figure; note that ON-T RGCs have spine-like protrusions that may not have been as apparent in the previously submitted version of our manuscript.

In Figures 1 and 2, Sustained responses look like the derivative of Transient responses, minus the negative going inflection. In addition, the sustained responses appear to have a lower threshold of activation than the transient On rgcs, because there are more bouts of action potentials (and membrane depol in V-clamp) with earlier onset in sustained than transients traces. It would be great if the GLuSniff data captured these differences. Take cumulative dF/F and see what the onset time is, or an initial tau if possible.

This is a good suggestion. However, we are reluctant to make detailed quantitative comparisons such as this without further validation of how the kinetics of the iGluSnFR signals relate to kinetics of glutamate release.  A specific concern is that differences in the location and amount of iGluSnFR expression could impact any such comparisons.

A recent study by Kim et al von Gersdorff (Cell Reports, 2023) presents interesting phases of release in response to light flashes, measured from AIIs, and complementary results from pairs of rbcs-AIIs. The findings highlight the complexity of SV pools under well-controlled experiments. Could their results be explained as variations in rbc ribbon size through development, and possibly between rbcs or within an rbc? 

This certainly seems possible and would be consistent with the dependence of release on ribbon size that our results support.  It would be interesting to see if there are clear anatomical correlates of that change in release properties.  

Figure 5 is a pivotal point in the study, but my review has identified numerous weaknesses. The feedback inhibition onto bipolar cell terminals is likely to sculpt glutamate release, and the results do not convincingly rule out this possibility. The suggestions for improvements range from the data needing to be reanalyzed with regard to statistical tests, and/or adding a few more data points (n = 5) before concluding a p: 0.06 is insignificant. 

We have added an additional recording to this data set. With n= 6 cells, there is now a statistically significant difference between ON-T RGC excitatory currents measured in control conditions versus during GABA_A/C receptor blockade. Please note that all the recordings shown in Figure 5C-F are from ON-T RGCs (the two panels show separately block of GABergic and glycinergic receptors). We did not make it sufficiently clear that the original trend (now statistically significant) is opposite of that expected if presynaptic GABAergic inhibition contributes to response transience in ON-T RGCs.  What we see is that excitatory synaptic inputs to ON-T RGCs become more transient (rather than mpre sustained) during GABA_A/C receptor blockade. We have revised the text in that section to make this point more clearly.

We have also included new data from iGluSnFR measurements showing that bath application of NBQX does not affect light step-evoked glutamate release kinetics at proximal (sustained) or distal (transient) RGC dendrites (control: steady-state amp. as % of peak amp. 13 ± 10; mean ± S.D.; n = 189 ROIs/4 FOVs for ON-T dendrites vs 40 ± 12; mean ± S.D.; n = 287 ROIs/8 FOVs for ON-S dendrites; NBQX: 6 ± 3; mean ± S.D.; n = 112 ROIs/1 FOV for ON-T dendrites vs 23 ± 9; mean ± S.D.; n = 97 ROIs/2 FOVs for ON-S dendrites; *p<0.001). By blocking glutamate receptors on amacrine cells, NBQX (AMPA/KAR antagonist) eliminates all/most amacrine cell-mediated signaling in the retina and should therefore abolish presynaptic inhibitory input to bipolar cell terminals across the IPL. Taken together, our results indicate that presynaptic inhibition does not play a critical role in establishing transient versus sustained kinetics for the stimulus conditions we employed in our study.

There is a need to cite more recent literature on bipolar cell ribbons (e.g. see Wakeham et al., Front. Cell. Neurosci., 2023), in order to support experimental design and interpretation of the results. The authors should discuss their Ribeye-KO data from Okawa et al 2019 Nat Comm, Figure 7, in the context of their new iGluSnFR results. 

Thank you for prompting us on this issue. We have expanded the discussion regarding ribbons and included more citations to the ribbon literature. That is largely in the three paragraphs starting on line 727.

One point deserves emphasis because it is central to the authors' ribbon model but not consistent with their data. The ribbon model as they put it, and as commonly stated, holds that a transient phase of release at the onset of depolarization indicates the depletion of the primed SVs, and the subsequent slower rate of release (steady state release in the authors' terms) reflects recruiting, priming, and release of new SVs. The On-transient dendrite GluSnf responses agree with this multiphasic process, but the sustained responses show only an elevation in glutamate without a pronounced initial peak, creating a square-wave-shaped response (Figure 6B). This does not agree with the simple ribbon-based release model. I would expect the signals from the T- and S-on dendrites to have a comparable initial phase, while the sustained phase should be greater in amplitude for the S-on dendrites. More discussion may clarify possible mechanisms.

Thanks for pointing this out. The example iGluSnFR traces we originally included in the manuscript were not entirely representative in that they did not show much initial transient phase. Note there is a distribution of steady-state amplitudes for proximal dendrites in Fig. 6C; the examples are from ROIs from the upper end of the distribution. In the new Figure 7, we have included some additional examples that show both a clear transient and sustained component. The summary data in Figure 6C shows the distribution of sustained/transient ratios across ROIs.  

Reviewer #3 (Recommendations For The Authors): 

(1) It would be interesting to understand the differences in IPSCs in the two RGC types. Perhaps they are small in both types, which would explain their apparent lack of impact on temporal tuning. The authors may already have these data.

We did make measurements of noise-evoked IPSCs (as well as EPSCs) in a subset of ON-T and ON-S recordings. We have now included this data as Figure S3. There are slight differences in the kinetics of inhibition between RGC types (Fig. S3C) and there is a trend towards stronger inhibition (relative to excitation) in ON-T RGCs compared to ON-S RGCs (Fig. S3E), although there is not a statistically significant difference. In both cases excitatory synaptic currents are as large or larger than inhibitory currents, and this does not include the difference in driving force near spike threshold which will favor excitatory input by a factor of 2-3.  Hence our data suggests that postsynaptic inhibition does not play a major role in generating the differential temporal spiking responses of ON-T and ON-S RGCs. However, additional experiments examining the relative contribution of excitation and inhibition to spiking output in these RGCs would be needed to reach a firm conclusion.

The pharmacological experiments in which we blocked inhibition (Fig. 5C-F, new Fig. 7) were designed to test the effect of presynaptic inhibition on bipolar cell output (voltage-clamp isolation of excitatory currents in Fig. 5; iGluSnFR measurements of glutamate release in Fig. 7). We do not mean to suggest that postsynaptic inhibition does not have any role in shaping the spiking behavior of these RGC types, but that transient vs. sustained kinetics are already present in the bipolar cell output and that presynaptic inhibition of bipolar cell terminals does not appear to account for this difference.  We have revised the text throughout to be clearer on this point.

(2) It could be convincing to show transient/sustained differences between RGC types in dim light, where the response would depend on the rod bipolar/AII circuit. In this case, any difference in temporal properties would presumably be explained by differences that localize to the cone bipolar cell axon terminals. Indeed, is that the result in Figure 1B? This seems to be a dim stimulus presented on darkness, which may be driven through the rod bipolar pathway. The authors could then discuss the interpretation of this data in terms of the rod bipolar circuit. 

Yes, Figure 1B is a dim light step (~30R*/rod/s) presented from darkness and the distinction between cells is clear down at still lower light levels that more effectively isolate signaling through the rod bipolar pathway. Thanks for making this point that observation of distinct temporal responses under scotopic conditions where signals suggests these differences must arise at and/or downstream of cone bipolar cell output. We have included additional text (lines 361-365) in the results describing bipolar cell responses that raise this point.

(3) Glutamate release was already measured across the full IPL depth by Borghuis et al. (2013) and Franke et al. (2017). It would be appropriate to better motivate the current study based on these existing measurements.

We have clarified that these important studies provided important motivation for measuring excitatory synaptic input to ON-T vs. ON-S RGCs (lines 165-169).   

(4) Line 212/213. It would be appropriate to add to the list of papers showing the different stratification of transient vs. sustained responses: Borghuis et al. (2013) and Beaudoin et al. (2019).

Thank you - these references have been added.  

(5) Line 635-638. It would be useful to discuss papers by Pottackal et al. (2020, 2021), which suggested that a single presynaptic cell (starburst) can signal with different temporal properties depending on the postsynaptic target (other starburst vs. DSGCs). The mechanism was not completely resolved (i.e., it was not explained by differences in presynaptic Ca channels at the two synapse types), but it at least shows that neurotransmitter release can show different filtering depending on the postsynaptic target from the same presynaptic neuron. (This could also be at play for the type 6 bipolar cell inputs to ON-S vs. ON-T RGCs in the present study.)

We have added a reference to Pottackal et al 2021 in this section.

(6) Line 714. Should describe the procedure for embedding the tissue in agarose. 

We have added more detail regarding agarose embedding for preparation of retinal slices in the methods.

(7) Line 775. Need a better description of the virus (not the construct), what serotype? Provide the Addgene number if available. 

This has been added to the methods.

(8) Line 808. Was the SD for the gaussian really 50%? That would cut off a lot of the distribution, i.e., it would get clipped at 0. 

Yes, the SD for Gaussian noise was 50%. This high contrast stimulus was used in part to achieve measurable signals from bipolar cells. You are correct that some of the distribution was clipped at 0 (it was also clipped at twice the mean to make sure that the distribution remained symmetrical). The clipping was accounted for during our LN analyses.

(9) The paper should discuss Swygart et al. (2024) results showing different spatial surround properties of neighboring synapses from a type 6 bipolar cell. Based on this result, it would seem very likely that amacrine cells could play a role in shaping the temporal processing of bipolar cell glutamate release as well. Indeed, spatial and temporal processing will not be completely independent in a typical experiment. For example, with the spot stimulus used in the present study, bipolar cells within the center versus the edge of the spot will have different balances of center/surround activation, which could potentially influence their temporal processing.

We have included discussion of results from Swygart et al 2024 in the section of the Discussion in which we point out differences in surround inhibition between ON-S and ON-T RGCs (lines 710-714). We agree that spatial and temporal processing are not completely independent. Our results with SR95531/TPMPA indicate ON-T RGCs receive stronger GABAergic surround inhibition than ON-S RGCs (Fig. S8). However, our results in Fig. 5C-D show GABAergic surround inhibition makes ON-T excitation more sustained rather than more transient. So even though bipolar cells presynaptic to ON-T RGCs receive stronger surround inhibition (Fig. S8), this inhibition does not establish the transient kinetics of glutamate release from these bipolar cells (in fact, it works to make release more sustained). Additional iGluSnFR experiments where we used NBQX to block all/most amacrine cell-mediated responses also suggest presynaptic inhibition does not have an important role in establishing differential glutamate release kinetics onto ON-S vs. ON-T RGC dendrites (Fig. 7).

(10) Cui et al. 2016 described ON-S Alpha as having a divisive suppression mechanism that explained the temporal properties of white-noise response better than a standard LN model. Do the authors think the divisive suppression reflects a property of the excitatory synapses independent of inhibition?

This is an interesting question, but one for which we don’t have a good answer for now. As mentioned in some of the above responses and as we have tried to clarify in the manuscript, we do not mean to imply that there is no role for presynaptic inhibition in modulating bipolar cell output, including for the divisive suppression described by Cui et al. Rather, our point is that the distinction between transient and sustained excitatory input to ON-T and ON-S RGCs does not require presynaptic inhibition and is more likely an intrinsic property of the bipolar cell synapses. 

(11) Do the authors mean to imply that the pool size at bipolar cell ribbon synapses could depend on the use of Ames vs. ACSF? 

For now, we do not have a good answer as to why there are quantitative differences in response kinetics between Ames and ACSF. We have not done any experiments to investigate whether ribbon sizes or ribbon pools are different in the different solutions.

(12) More generally, different mean luminance levels could drive different levels of baseline glutamate release, which could alter the available pool of vesicles at bipolar cell ribbon synapses. Can we explain varying degrees of transient/sustained in the same cell at different levels of mean luminance based on this mechanism (e.g., Grimes et al., 2014)?

Yes, the emergence of a transient component of excitatory input to ON-S RGCs at ~100 R*/rod/s versus at scotopic levels (0.5 R*/rod/s) in Grimes et al. (2014) could be due to differences in the number of releasable vesicles (due to different type 6 bipolar cell axon terminal membrane potentials and hence differences in spontaneous release rates) at the different light levels.

We should note that although ON-T and ON-S RGCs exhibit some changes in transient/sustained kinetics across different light levels, the relative differences between these RGC types are preserved across light levels. We have included a statement about this in the text (lines 361-367).

(13) Figure 1. Have the authors considered performing the LN analysis of the firing responses, to compare the degree of rectification between the two RGC types?

This is a good suggestions. From an LN analysis of spiking responses, we do not observe a clear difference between the static nonlinearity component of the model for ON-T and ON-S RGCs. Both RGC types are strongly rectified under our experimental conditions.  

(14) Figure 5. Do the authors have the pharmacology data for the ON-S cells? There are examples of sustained EPSCs in amacrine cells that become more transient after blocking inhibition, which at least suggests that inhibition can play some role in the transient/sustained nature of glutamate release (Park et al., 2015, Figure 3). Perhaps ON-S cells likewise become more transient with inhibition blocked. 

(The colored symbols in A were not visible in a printout. It would be useful to indicate the cell type (ON-T) in C and E). 

As described above in the response to reviewer 1’s recommendation for authors, we were not able to use SR95531/TPMPA for recordings from ON-S RGCs. Bath application of these drugs led to oscillatory bursts of excitatory input to ON-S RGCs. However, the lack of effect of bath-applied NBQX on the kinetics of glutamate release around either ON-T or ON-S RGC dendrites (new Fig. 7) suggests that presynaptic inhibition does not contribute to generating sustained excitation to ON-S RGCs (or transient excitation to ON-T RGCs).  

We have corrected Fig. 5A to include the referenced colored symbols and have also edited Fig 5C and E to clarify that measurements in Fig. 5C-F are from ON-T RGCs.

(15) Figure 6 legend. Should be Kcng4-Cre, not KCNG-Cre. Also, it should make clear that this is cre-dependent expression of iGluSnFR. For C, were the statistics based on the number of FOVs? 

Thanks for catching this, we have corrected Figure 6 legend. The methods section includes a description of how we achieved iGluSnFR expression on alpha RGC dendrites via a cre-dependent viral strategy in Kcng4-Cre mice.  We have also clarified that the statistics are based on ROIs in Figure 6C.

(16) Figure 7, Flashes were apparently 400% contrast on a dim background. What was the background? Is there a rod component to the response in this case? 

In Figure 7 (now Figure 8), the same background (~3300 R*/rod/s; 2000 P*/Scone/s) was used as in the Gaussian noise and step response experiments. At this light level, the response should be primarily be mediated by cones.

(17) Figure S1. The colors here differ from those in previous figures (Here, ON-T, magenta; ON-S, cyan). Is something mislabeled? 

Thanks for catching this. We mistakenly swapped the labels in the legend for Fig. S1. The figure colors were correct, but we have corrected the legend in the revised manuscript.

(18) Figure S2. For the LN model for RGC synaptic currents, the ON-S are more rectified than some previous recordings (Cui et al., 2016). Is this perhaps explained by different light levels?

We aren’t sure why ON-S excitatory currents are more strongly rectified in our recordings compared to Cui et al., 2016. Cui et al. used an ~20-fold higher background light intensity (~40,000 P*/cone/s vs. ~2000 P*/cone/s in our study), so different light levels may be a factor (although we should point out that rectification increases in these RGCs between scotopic to low photopic light levels (see Grimes et al., 2014 and Kuo et al., 2016).

(19) The study is apparently comparing PV1 and PV2 described in Farrow et al. (2013; see Supplementary information for stratification analysis), which should be cited.

Thanks, we have corrected this oversight in the revised manuscript. We now cite Farrow et al and mention the connection to PV1 and PV2 in the first paragraph of Results (lines 104-108).

eLife Assessment

This important study reports the conservation of sperm-egg envelope binding by demonstrating successful recognition of the micropyle in fish eggs by mouse sperm. The evidence supporting the conclusions drawn is convincing. This study will be of interest to reproductive biologists and clinicians studying the biology of fertilization and fertility.

Reviewer #1 (Public review):

Summary:

The paper is well written and investigates the cross-species insemination of fish eggs with mouse sperm. and I have a few major and minor comments.

Strengths:

The experiments are well executed and could provide valuable insights into the complex mechanisms of fertilization in both species. I found the information presented to be very interesting,

Weaknesses:

The rationale of some of the experiments, in particular those using CatSper KO sperm is, in my view.

Author response:

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

Reviewer #1 (Public review):

Summary:

The paper is well written and investigates the cross-species insemination of fish eggs with mouse sperm. I have a few major and minor comments.

Strengths:

The experiments are well executed and could provide valuable insights into the complex mechanisms of fertilization in both species. I found the information presented to be very interesting,

Thank you.

Weaknesses:

The rationale of some of the experiments is not well defined.

Thank you. In the revised manuscript, we have clarified and expanded the rationale behind each experiment to better highlight the specific questions being addressed and how each approach contributes to our overall investigation. These clarifications have been integrated throughout the Results and Discussion sections. We provide detailed rationale in our point-by-point responses to both reviewers, outlining how each experimental design was motivated by prior findings, hypotheses, or specific gaps in knowledge. We hope these revisions make the experimental logic and progression better defined and more compelling.

Major Comments:

(1) Figure 5

I do not understand the rationale for performing experiments using CatSper-null sperm and CD9-null oocytes. It is well established that CatSper-null sperm are unable to penetrate the zona pellucida (ZP), so the relevance of this approach is unclear.

We thank the reviewer for this comment. This experiment was conducted as the basis to then evaluate the contributions of progressive and hyperactivated motility to the ability of mouse sperm to locate and traverse the zebrafish micropyle. In earlier experiments (Figures 1 and 3), we assessed whether sperm-micropyle interaction was robust by comparing it to binding to the mouse zona pellucida and testing whether both interactions persisted after washing, which is standard approach to distinguish specific binding from non-specific adherence (Avella et al., 2014; Baibakov et al., 2012). Thus, we extended this analysis to CatSper1^Null sperm; CatSper1^Null sperm were still capable of binding the zona pellucida comparably to heterozygous controls, though they were unable to cross the zona of Cd9^Null eggs. These observations served as a validation step for the use of CatSper1^Null sperm for downstream micropyle interaction assays. Thus, we proceeded to test whether hyperactivated motility, absent in CatSper1^Null sperm, is required for locating and crossing the micropyle.

It is indeed well established that CatSper1^Null sperm are unable to penetrate the zona pellucida, and previous studies have typically used the absence of fertilized eggs as a readout. However, failed fertilization may result from multiple factors, including impaired sperm motility, reduced capacity to bind the zona pellucida, or an inability to penetrate it. To our knowledge, no study has quantitatively assessed the number of CatSper-deficient sperm that successfully bind, cross the zona and reach the perivitelline space. To address this, we first used normal oocytes for sperm binding and Cd9^Null oocytes (Le Naour et al., 2000), which allow direct quantification of sperm accumulation in the perivitelline space. We have 7included a detailed explanation in the Results to clarify this point, lines 352-365 and 376-369.

(2) Micropyle penetration and sperm motility

CatSper-null sperm are reportedly unable to cross the micropyle, but this could be due to their reduced motility rather than a lack of hyperactivation per se. Were these experiments conducted using capacitated or non-capacitated spermatozoa? What was the observed motility of CatSper-null sperm during these assays? Clarifying these conditions is essential to avoid drawing incorrect conclusions from the results.

Thank you for raising these points. Under our IVF conditions, qualitative observations confirmed that CatSper1^Null sperm displayed progressive motility, maintained sufficient progressive motility during the first hour post-insemination and exhibited zona binding efficiency comparable to that of CatSper1^Het controls (Figure 5A and B). This is consistent with previous reports showing that within the first 90 minutes of sperm incubation in media, approximately 20% of CatSper1^Null sperm preserve motility (Qi et al., 2007). Given previous studies indicating that 15–35% of sperm undergo hyperactivation within 90 minutes (Goodson et al., 2011), and considering that 100,000 progressively motile sperm were used for insemination, we estimate that approximately 3,000 hyperactivated CatSper1^Null sperm were present in the cross-species insemination dish (mouse sperm x zebrafish eggs). Based on these numbers, we would have expected at least some sperm to locate the micropyle if hyperactivation were not required for its detection and entry. Nevertheless, CatSper1^Null sperm were detected in proximity to the micropyle canal, its opening, or within the inter-chorion space (ICS). These observations support the conclusion that the inability ofCatSper1^Null sperm to locate and enter the micropyle is attributable to their failure to hyperactivate. Also, all sperm used in these assays were exposed to identical capacitating conditions (HTF/HSA, 37 °C, 5% CO2). We now clarify this in the Methods, line 624, and we added more rationale under the Results, lines 361-365 and in the Discussion, lines 470-483.

(3) Rheotaxis and micropyle navigation

Previous studies have shown that CatSper-null sperm fail to undergo rheotaxis. Could this defect be related to their inability to locate and penetrate the micropyle? Exploring a potential shared mechanism could be informative.

Thank you for raising this interesting point. Indeed, homozygous mutant mice lacking expression of a different component of the CatSper channel, CatSperz, show reduced rheotactic efficiency and severe subfertility (Chung et al., 2017). We cannot exclude that complete lack of CatSper as shown in CatSper1^Null mice could lead to reduced rheotactic efficiency, hence we include this interpretation in the Discussion (lines 484-486).

(4) Lines 61-74

This paragraph omits important information regarding acrosomal exocytosis, which occurs prior to sperm-egg fusion. Including this detail would strengthen the discussion.

Thank you. We have revised the text in the discussion to describe the process of acrosome exocytosis, and its relevance for fertilization (lines 504-518).

Reviewer #2 (Public review):

Summary:

Garibova et al. investigated the conservation of sperm recognition and interaction with the egg envelope in two groups of distantly related animals: mammals (mouse) and fish (zebrafish). Previous work and key physiological differences between these two animal groups strongly suggest that mouse sperm would be incapable of interaction with the zebrafish egg envelope (chorion) and its constituent proteins, though homologous to the mammalian zona pellucida (ZP). Indeed, the authors showed that mouse sperm do not bind recombinant zebrafish ZP proteins nor the intact chorion. Surprisingly, however, mouse sperm are able to locate and bind to the zebrafish micropyle, a specialized canal within the chorion that serves as the egg's entry point for sperm. This study suggests that sperm attraction to the egg might be highly conserved from fish to mammals and depends on the presence of a still unknown glycosylated protein within the micropyle. The authors further demonstrate that mouse sperm are able to enter the micropyle and accumulate within the intrachorionic space, potentially through a CatSper-dependent mechanism.

Strengths:

The authors convincingly demonstrate that mouse sperm do not bind zebrafish ZP proteins or the chorion. Furthermore, they make the interesting observation that mouse sperm are able to locate and enter the zebrafish micropyle in an MP-dependent manner, which is quite unexpected given the large evolutionary distance between these species, the many physiological differences between mouse and zebrafish gametes, and the largely different modes of both fertilization and reproduction in these species. This may indicate that the sperm chemoattractant in the egg is conserved between mammals and fish; however, whether zebrafish sperm are attracted to mouse eggs was not tested.

Thank you. We performed an additional experiment with fish sperm used to inseminate ovulated mouse eggs, and results are reported in lines 183-187 and in Supplementary Figure 2.

Weaknesses:

The key weakness of this study lies in the rationale behind the overall investigation. In mammals, the zona pellucida (ZP) has been implicated in binding sperm in a taxon-specific manner, such that human sperm are incapable of binding the mouse ZP. Indeed, work by the corresponding author showed that this specificity is mediated by the N-terminal region of the ZP protein ZP2 (Avella et al., 2014). The N-termini of human and mouse ZP2 share 48% identity, which is higher than the overall identity between mouse and zebrafish ZP2, with the latter ortholog entirely lacking the N-terminal domain that is essential for sperm binding to the ZP. Given this known specificity for mouse vs. human sperm-ZP binding, it does not follow that mouse sperm would bind ZP proteins from not only a species that is much more distantly related, but also one that is not even a mammal, the zebrafish. Furthermore, the fish chorion does not play a role in sperm binding at all, while the mammalian ZP can bind sperm at any location. On the contrary, the zebrafish chorion prevents polyspermy by limiting sperm entry to the single micropyle.

We thank the reviewer for this detailed comment. In this study, our goal was precisely that one of validating the hypothesis that mouse sperm would not bind either recombinant fish ZP proteins or the chorion; in addition, we found it important to examine the observation that mouse sperm could detect the micropyle. We further elaborated this rationale in the Introduction (lines 93-100).

In addition, though able to provide some information regarding the broad conservation of sperm-egg interaction mechanisms, the biological relevance of these findings is difficult to describe. Fish and mammals are not only two very distinct and distantly related animal groups but also employ opposite modes of fertilization and reproduction (external vs. internal, oviparous vs viviparous). Fish gametes interact in a very different environment compared to mammals and lack many typically mammalian features of fertilization (e.g., sperm capacitation, presence of an acrosome, interaction with the female reproductive tract), making it difficult to make any physiologically relevant claims from this study. While this study may indicate conserved mechanisms of sperm attraction to the egg, the identity of the molecular players involved is not investigated. With this knowledge, the reader is forced to question the motivation behind much of the study.

We thank the reviewer for their perspective, and we appreciate the opportunity to further elaborate on our rationale. As outlined in our Results and Discussion sections, a growing body of evidence supports the presence of conserved molecular players and signaling pathways involved in gamete interaction across species with diverse reproductive strategies. While zebrafish and mice do differ in their fertilization environments and modes of reproduction, these differences may not necessarily exclude the possibility of conserved molecular mechanisms underlying gamete interaction. For example, the CatSper calcium channel, which plays a key role in regulating sperm motility and hyperactivation, is conserved across a broad range of taxa—from echinoderms such as sea urchins (external fertilizers)(Seifert et al., 2015) to mammals, including mice and humans (internal fertilizers)(Lishko and Mannowetz, 2018). Moreover, sperm from some fish species possess acrosomes that undergo exocytosis prior to fertilization while sperm cross the micropyle (Psenicka et al., 2010). Also, in ovoviviparous species with internal fertilization, such as the black rockfish, sperm do undergo molecular changes while in the female reproductive tract—including immunomodulatory adaptations, glycocalyx remodeling, and interactions with ovarian cells—enabling the sperm with a longer-term survival and a selective persistence that ensures only the fittest sperm can successfully fertilize eggs (Li et al., 2024). As per the mammalian capacitation, it is broadly defined as the process during which sperm undergo hyperactivation (Yanagimachi, 1970), and acquire the ability to undergo the acrosome exocytosis, making the sperm competent for gamete fusion and fertilization (Bhakta et al., 2019; Puga Molina et al., 2018; Yanagimachi, 1957; Yanagimachi et al., 2017). Of note, acrosome exocytosis or changes in sperm motility are not exclusive to internal fertilizers. For example, as we cite in our manuscript (and as just stated above), acrosome exocytosis has been described to occur as sturgeon sperm cross the micropyle (Psenicka et al., 2010). As per changes in flagellar motility, investigations in the Pacific herring (Clupea sp.) demonstrated that sperm remain nearly immotile upon release into seawater and only initiate motility when approaching the micropyle region of the egg (Yanagimachi, 1957; Yanagimachi et al., 2017). In other fish, including bitterling and zebrafish, further enhancement in sperm motility is observed as sperm approach the micropyle area (Suzuki, 1958; Yanagimachi et al., 2017). These studies suggest that functional equivalents of capacitation may exist across taxa.

We interpret the observation that mouse sperm can locate and enter the micropyle as suggesting that underlying guidance mechanisms may be more broadly conserved across distant species than previously recognized. We have now elaborated on these points in the revised Discussion (lines 531-552), and we hope the motivation behind our study is now more clearly articulated.

During fertilization in fish, the sperm enters the micropyle and subsequently, the egg, as it is simultaneously activated by exposure to water. During egg activation, the chorion lifts as it separates from the egg and fills with water. This mechanism prevents supernumerary sperm from entering the egg after the successfully fertilizing sperm has bound and fused. In this study, the authors show that mouse sperm enter the micropyle and accumulate in the intrachorionic space. Whether any sperm successfully entered the egg is not addressed, and the status of egg activation is not reported.

We appreciate the reviewer’s detailed comments and the opportunity to elaborate on this important aspect for our cross-insemination assay. We interpret the reviewer’s reference to “sperm entering the egg” as pertaining to sperm adhesion to the oocyte plasma membrane followed by fusion with the egg cell, two separate steps regulated by different molecular players for sperm-egg plasma membrane adhesion (Bianchi et al., 2014; Fujihara et al., 2021; Herberg et al., 2018; Inoue et al., 2005) and for fusion. It is important to note that proteins mediating gamete fusion are still unidentified in fish and mammals (Bianchi and Wright, 2020; Deneke and Pauli, 2021).

In our cross-species insemination experiments, zebrafish oocytes were maintained in Hank’s solution to limit spontaneous activation; however, as the reviewer correctly notes, activation likely occurred upon exposure to HTF. While this model does not recapitulate full fertilization events, it serves as a platform to explore whether mammalian sperm can detect (within the scope of our study) and respond (future studies) to putative evolutionarily conserved signals, such as those guiding fish sperm toward the micropyle.

While investigating cross-species sperm–oocyte fusion was not within the scope of this study and would require a distinct set of experimental approaches, we believe this question is an important one. However, we do not expect our platform to be informative for evaluating sperm adhesion to the fish oolemma or for enabling cross-species gamete fusion. In our assays focused on sperm-micropyle interaction, Hoechst staining of nuclei of transgenically-tagged acrosome sperm revealed no evidence of sperm adhesion to or fusion with the fish egg membrane (Figure 4D). Also, molecular incompatibilities may further prevent this interaction: in zebrafish, the Ly6/uPAR family protein Bouncer is expressed exclusively in the egg and is necessary for sperm–egg membrane adhesion (Herberg et al., 2018). Recent studies in zebrafish and mice have shown that a conserved trimeric complex composed of Izumo1, Spaca6, and Tmem81 on the sperm surface is required for mediating adhesion to the oocyte membrane by interacting with the mammalian oocyte receptor Izumo1R (also known as JUNO) or the zebrafish oocyte receptor Bouncer (Deneke et al., 2024). One would hypothesize that for mouse sperm to adhere to the zebrafish egg membrane, the mouse Izumo1-Spaca6-Tmem81 complex would need to establish binding with Bouncer. To explore this possibility, we performed AlphaFold2-Multimer structural predictions and docking analyses to mimic an interaction between mouse Izumo1-Spaca6-Tmem81 and zebrafish Bouncer, using mouse Izumo1-Spaca6-Tmem81 and Juno or zebrafish Izumo1-Spaca6-Tmem81 and Bouncer as positive controls. We observed low binding affinity between zebrafish Bouncer and the mouse trimeric complex (Izumo1, Spaca6, and Tmem81), as indicated by low ipTM scores and high predicted aligned error (PAE) values. These findings suggest that the mouse complex is unlikely to form an interaction with Bouncer (now shown in Suppl. Figure 7). These predictions were consistent with our observations that no sperm were found adhering or fusing to the egg cell. We describe methods and results in the supplementary files (Supporting Info, lines 53-66) and in the result sections (lines 335-339).

In Supplementary Videos 3-4, the egg shown has been activated for some time, as evident by the separation of yolk and cytoplasm, yet the chorion is only partially expanded (likely due to mouse IVF conditions). How multiple sperm were able to enter the micropyle but presumably not the egg is not addressed, yet this suggests that the zebrafish mechanism of blocking polyspermy (fertilization by multiple sperm) is not effective for mouse sperm or is rendered ineffective due to mouse IVF conditions. The authors do not discuss these observations in the context of either species' physiological process of fertilization, highlighting the lack of biological context in interpreting the results.

Thank you for raising this important point. One model for mammalian gamete recognition at the zona supports the notion that mouse sperm can penetrate extracellular matrices as long as sperm can bind to them, and binding is dependent on the cleavage status of ZP2. Zonae surrounding unfertilized mouse eggs present uncleaved ZP2 and these zonae support sperm binding. After gamete fusion, the cortical granules release ovastacin which cleaves ZP2 at the N-terminus, and consequently, zonae presenting cleaved ZP2 no longer support sperm binding. This mechanism acts as block to zona binding and prevents further crossing (Bhakta et al., 2019). Indeed, fertilized mouse eggs or 2-cell embryos surrounded by a zona containing uncleaved ZP2 support de novo sperm binding, and supernumerary sperm cross the zona and accumulate in the perivitelline space, unable to fuse with the fertilized oocyte plasma membrane or blastomere cells (Baibakov et al., 2012, 2007; Burkart et al., 2012; Gahlay et al., 2010). Thus, because under our experimental conditions, mouse sperm could interact with the micropyle opening, we interpret these findings to suggest that once interaction occurs at the micropyle opening, mouse sperm are capable of crossing it, even under conditions where the micropyle may be detached from the oocyte due to oocyte activation. Therefore, our data indicates that mouse sperm may be able to bypass the mechanism of zebrafish oocytes blocking multiple sperm to pass through the micropyle, even after oocyte activation. This point has now been incorporated into the revised Discussion (lines 425-441).

The authors further show that the zebrafish micropyle does not trigger the acrosome reaction in mouse sperm. Whether the acrosome reacts is not correlated with a sperm's ability to cross the micropyle opening, as both acrosome-intact and acrosome-reacted sperm were observed within the intrachorionic space. While the acrosome reaction is a key event during mammalian fertilization and is required for sperm to fertilize the egg, zebrafish sperm do not contain an acrosome. Thus, these results are particularly difficult to interpret biologically, bringing into question whether this observation has biological relevance or is a byproduct of egg activation/chorion lifting that indirectly draws sperm into the chorion.

We thank the reviewer for raising this point and we appreciate the opportunity to elaborate on the biological relevance of this experiment. Our motivation to assess acrosome status in mouse sperm following entry into the zebrafish micropyle stemmed from the following biological considerations.  In fish species such as the sturgeon, sperm present an acrosome and undergo acrosome exocytosis while passing through the micropyle, before gamete fusion (Alavi et al., 2012; Psenicka et al., 2010). By contrast, zebrafish sperm lack an acrosome, raising the hypothesis that the zebrafish micropyle may not be able to trigger acrosome exocytosis. However, this possibility has not been experimentally tested. We therefore considered it important to investigate whether passage through the zebrafish micropyle induces acrosome exocytosis in mouse sperm. We have revised the Discussion to better clarify the rationale behind the experiment as well as the interpretation of the findings (lines 504-518). As per the chorion lifting indirectly drawing sperm into the chorion, we have not observed this phenomenon.

The final experiments regarding CatSper1's role in mediating mouse sperm entry into the micropyle/chorion are not convincing. As no molecular interactions are described or perturbed, the reader cannot be sure whether the sperm's failure to enter is due to signaling via CatSper1 or whether the overall failure to undergo hyperactivation limits sperm motility such that the mutant sperm can no longer find and enter the zebrafish micropyle. Indeed, in Figure 5E, no CatSper1 mutant sperm are visible near any part of the egg, suggesting that overall motility is impaired, and this is not a phenotype specific to interactions with the micropyle.

We appreciate the comment and the opportunity to further elaborate on the rationale of this experiment. While our data demonstrates a lack ofCatSper1^Null sperm accumulation within the micropyle and ICS, we appreciate that this may be interpreted as the result of general motility defects, rather than a specific failure in undergoing hyperactivation and micropyle recognition. CatSper1^Null  sperm are known to lack hyperactivated motility and exhibit a progressive loss of forward motility over time. After 90 minutes, only ~20% of CatSper1^Nulll sperm remain motile, compared to over 70% in fertile sperm (Qi et al., 2007). Of note, under our IVF conditions, CatSper1^Null sperm retained sufficient progressive motility during the first hour post-insemination to bind the zona pellucida with comparable efficiency to CatSper1^Het controls. Based on prior reports indicating that 15–35% of sperm exhibit hyperactivation by 90 minutes (Goodson et al., 2011), and considering that we inseminated with 100,000 progressively motile sperm, we estimate that approximately 3,000 hyperactivated CatSper1^Null sperm were present in the dish. Yet, none were observed near the micropyle canal, its opening, or within the ICS. This led us to conclude that failure to hyperactivate underlies the inability of CatSper1^Null sperm to reach and traverse the micropyle. Also, we appreciate that identifying the molecular components of the micropyle would allow direct testing of whether the CatSper channel is activated in response to micropyle-associated signals. Indeed, no targeted perturbation of molecular interaction regulating micropyle recognition was performed in this study, as the molecular identity of the zebrafish micropyle guidance cue remains unknown. Efforts to identify and characterize this factor are ongoing in our lab and lie outside the scope of the current work. Therefore, throughout the manuscript, we have clarified that it is the failure to undergo hyperactivation, rather than the absence of CatSper per se, that limits the ability of sperm to locate and traverse the micropyle. The rationale for the experiment, the interpretation of our findings, and relevant future directions have been further elaborated in the revised Abstract, Impact Statement and Discussion (lines 40-41; 46-47; 343-365; 376-379; 389-399; 470-486).

Reviewer #1 (Recommendations for the authors):

Minor Comments

(1) Figure numbering

There appear to be inconsistencies in the figure references. For example, what is referred to as Figure 3F in the text is actually Figure 4F. Please review and correct all figure labels for accuracy.

We thank the reviewer for pointing this out. We have carefully reviewed the manuscript and corrected all figure references throughout the text. Also, for better flow and coherence, we have moved the paragraph describing the videos to the end of the Results section titled "Mouse sperm recognize the micropylar region of fish oocytes." Previously, the callout of panels in Figure 3 was out of order (3A, 3B, 3E, 3C, 3D), and this reorganization also helps maintain logical progression through the figure panels.

(2) Figure 5 terminology:

The term "normal" sperm should be replaced with "CatSper heterozygous (Het)" sperm to avoid confusion and improve precision.

We thank the reviewer for this helpful suggestion. We have revised the terminology in Figure 5 and throughout the manuscript, replacing “normal” sperm with “CatSper1 heterozygous (Het)”

Reviewer #2 (Recommendations for the authors):

In addition to my comments in the public review, I would encourage the authors to consider the following suggestions:

The authors show that mouse sperm can find and enter the fish micropyle, and that this depends on the presence of MP. To better assess sperm binding to the micropyle region, the number of sperm binding to the micropyle vs. non-micropyle chorion should be clearly quantified, as well as the percentage of sperm that enter the micropyle compared to the total used for insemination. The authors state several times throughout the text that a "subpopulation" of mouse sperm finds and enters the micropyle, but it would be more precise and informative to give a percentage.

We thank the reviewer for this suggestion. We have now reported also the number of sperm bound to the other regions of the chorion (away; lines 231-233), as well as the percentage of sperm that entered the micropyle relative to the total number used for insemination (lines 276-279).

To ensure that all sperm are inside the chorion, the egg should be removed from the insemination dish, washed thoroughly, and then the chorion should be torn open to definitively show that the sperm were indeed inside.

We thank the reviewer for these excellent suggestions. As per ensuring that the sperm are inside the ICS, (as shown now in Figures 4A, F, G , Supplementary Figure 6 and Supplementary Movies 3–5), the inseminated oocytes were thoroughly washed prior to imaging to ensure that only sperm located inside the chorion were visualized (as described in the Methods, lines 646-648). In addition, to confirm the spatial localization of sperm within the ICS, we are now including additional TEM images showing sperm in the ICS (Figure 4G, right panel). Also, we generated orthogonal views using ZEN Lite software (Zeiss, Germany) from a z-stack encompassing the full volume of the chorion, ICS, and oocyte (added in the supplementary materials, as Supplementary Figure 6). These views display three focal planes: the surface of the WGA-stained chorion, the middle of the ICS, and the oocyte plasma membrane. Sperm nuclei stained with Hoechst are clearly visible below the chorion surface and above the oocyte plasma membrane, confirming their localization within the ICS. Additionally, in a separate set of experiments, as recommended by this reviewer, we mechanically disrupted the chorion and consistently detected sperm within the ICS. This procedure, however, was technically challenging: upon disruption, the chorion often collapsed onto the oocyte, and during the extraction process, sperm were sometimes displaced. As a result, it was not always possible to determine with complete confidence whether the sperm had originally been located inside or outside the chorion. However, we hope that the additional TEM and confocal images (Figure 4G and Supplementary Figure 6) offer further support for the localization of sperm within the ICS.

I would further suggest that they examine the micropyle opening after the entry of multiple sperm, as well as the dynamics of egg activation during insemination with mouse sperm.

Thank you. We now include one additional TEM image capturing the full structure of a micropyle that was traversed by multiple mouse sperm (shown in Figure 4G, left panel).

At what point does the micropyle detach from the egg surface? Live imaging of this process with a confocal microscope would be very informative.

During live imaging, the interval between placing the oocyte in the imaging dish, replacement of Hank’s solution with HTF and the addition of sperm, followed by the initiation of video acquisition, is approximately 2 to 3 min. By this time, the ICS is already apparent (Supplementary Video 2), although the micropyle appears to remain adherent to the egg cell. Partial detachment of the micropyle from the egg cell begins around 6–7 minutes after imaging starts and continues progressively over time. We provide time-lapse imaging frames to show the micropyle detachment under mouse IVF conditions (Supplementary Figure 5).

Along the same lines, sperm should be doubly labeled with an acrosome-independent marker, i.e., a live DNA stain or MitoTracker. Then the authors could track if any sperm are actually able to enter the egg itself, which would be highly unlikely but an important detail to confirm.

Thank you for pointing this out. In our assays designed to study sperm–micropyle interactions, Hoechst staining of nuclei in transgenically labeled acrosome sperm showed no indication of sperm adhesion to, or fusion with, the zebrafish egg cell (Figure 4D).

Line 242, 282: The text should refer to Figure 4, not 3. Please make sure all figure references correspond to the correct figure and panel.

Thank you for bringing this to our attention. We have carefully reviewed the manuscript and corrected the reference to Figure 4, along with all other figure and panel citations to ensure they accurately correspond to the correct content. Also, to improve the overall flow, we relocated the paragraph describing the videos to the end of the Results section titled "Mouse sperm recognize the micropylar region of fish oocytes". This change also helped correct the sequence of figure panel references, which were previously cited out of order (i.e., 3A, 3B, 3E, 3C, 3D).

Line 244: The authors quantify sperm that are "away" from the micropyle, but this is not clearly defined. This should be given as a set radius or distance from the center (e.g., in microns). If the sperm are still motile, can this be accurately measured?

We thank the reviewer for this valuable suggestion. We have now defined “away from the micropyle” as a distance greater than 160 µm from the center of the micropyle. This measurement was determined using confocal z-stack projections of fixed samples. These details have been added to the revised Methods section (lines 670-674).

To strengthen the conclusion that the sperm chemoattractant is indeed conserved from fish to mammals, the authors could show that zebrafish sperm are also able to find/approach mouse eggs. Even more compelling would be to show the same is true for other species combinations. As it stands, the choice of comparing mouse and zebrafish does not seem scientifically motivated but rather due to their availability.

We thank the reviewer for this important suggestion. To test whether zebrafish sperm are capable of binding to the mammalian zona pellucida, we conducted the suggested experiment: ovulated, cumulus-free mouse oocytes were placed in water and incubated with zebrafish sperm. We did not observe any zebrafish sperm bound to the mouse zona pellucida, consistent with the hypothesis that zebrafish sperm do not recognize or interact with mammalian zonae or ZP proteins. This has now been added in the Results (lines 183-187) and shown in Supplementary Figure 2. We interpret these findings as in cross-species insemination assays, reciprocity in sperm-egg interaction is not always observed. For example, while human sperm bind only to human zonae and not to mouse zonae, mouse sperm are able to bind both mouse and human zonae (Avella et al., 2014; Baibakov et al., 2012; Bedford, 1977). This asymmetry may reflect species-specific adaptations in sperm-egg recognition. We have now added this point to the revised Discussion to clarify the rationale and context of our approach (lines 416-423).

As per the choice of experimental models, while we agree that testing additional species combinations would broaden the scope of the findings, the choice to compare mouse and zebrafish was not solely based on availability. Rather, it was motivated by the opportunity to examine sperm guidance across two evolutionary distant vertebrates. This contrast allows us to seek for potential conservation of structural or molecular cues involved in gamete interaction. Additionally, both zebrafish and mouse offer extensive gene editing, blotting and imaging reagents, which are particularly valuable should future studies aim to identify and functionally disrupt genes encoding micropyle-associated proteins and their putative orthologs in mammals.

For the CatSper experiment, I would suggest that the authors repeat this experiment with another mouse sperm mutant that is known to have reduced/altered motility. With the current data, I do not believe the failure to find/enter the micropyle is necessarily CatSper-specific. Because we do not know what the sperm interacts with in the micropyle or what the MP interacts with on the sperm, the signaling pathway cannot be tested, making other controls necessary for these results to be meaningful.

Thank you for highlighting this important point. A wide range of mouse models with sperm motility defects exhibit subfertility or infertility due to structural abnormalities in the axoneme or midpiece rigidity. (Miyata et al., 2024). These defects often result in impaired progressive motility, failure to reach the zona pellucida, or inability to bind or penetrate it. In contrast, we could test and validate that CatSper1^Null sperm display preserved early progressive motility but fail to transition into hyperactivated motility, making them particularly well suited for specifically assessing the role of hyperactivation in sperm navigation toward and entry into the micropyle. Taken together, these points, along with those discussed in our response to the public review, led us to conclude that the CatSper1^Null model provides the most biologically relevant context currently available to assess the role of hyperactivation in guiding sperm to the micropyle.

The authors could greatly strengthen the discussion by addressing the key points I raised in the public review, particularly in terms of interpreting these results in the context of each species' physiological mode of fertilization.

We thank the reviewer for this important recommendation. We have carefully revised the Discussion to address the key points raised in the public review, particularly by framing our findings within the context of the distinct physiological modes of fertilization in each species, as indicated n our answers to the public review. We hope these additions have strengthened the manuscript as suggested.

eLife Assessment

The article presents important findings on the impact of climate change on odonates, integrating phenological and range shifts to broaden our understanding of biodiversity change. The study leverages extensive natural history data, offering a convincing analysis of temporal trends in phenology and range limit and their potential drivers.

Reviewer #1 (Public review):

Summary:

This study evaluates whether species can shift geographically, temporally, or both ways in response to climate change. It also teases out the relative importance of geographic context, temperature variability, and functional traits in predicting the shifts. The study system is large occurrence datasets for dragonflies and damselflies split between two time periods and two continents. Results indicate that more species exhibited both shifts than one or the other (or neither), and that geographic context and temperature variability were more influential than traits. The results have implications for future analyses (e.g. incorporating habitat availability) and for choosing winner and loser species under climate change. The results also seem to support climate vulnerability assessments for species that rely on geographic range size and geospatial climate data layers rather than more detailed information (like demographic rates, abundances, or traits) that may not be so readily available. The methodology would be useful for other taxa and study regions with strong participatory ("citizen") science and extensive occurrence data.

Strengths:

This is an organized and well written paper that builds on a popular topic and moves it forward. It has the right idea and approach, and the results are useful answers to the predictions and for conservation planning (i.e. identifying climate winners and losers). There is technical proficiency and analytical rigor driven by an understanding of the data and its limitations.

Reviewer #2 (Public review):

Summary:

This paper explores a highly interesting question regarding how species migration success relates to phenology shifts, and it finds a positive relationship. The findings are significant, and the strength of the evidence is solid. However, there are substantial issues with the writing, presentation, and analyses that need to be addressed. First, I disagree with the conclusion that species that don't migrate are "losers" - some species might not migrate simply because they have broad climatic niches and are less sensitive to climate change. Second, the results concerning species' southern range limits could provide valuable insights. These could be used to assess whether sampling bias has influenced the results. If species are truly migrating, we should observe northward shifts in their southern range limits. However, if this is an artifact of increased sampling over time, we would expect broader distributions both north and south. Finally, Figure 1 is missed panel B, which needs to be addressed.

Comments on revised version:

The revision has substantially improved the paper.

Reviewer #3 (Public review):

Summary:

In their article "Range geography and temperature variability explain cross-continental convergence in range and phenology shifts in a model insect taxon" the authors rigorously investigate the spatial and temporal trends in the occurrence of odonate species and their potential drivers. Specifically, they examine whether species shift their geographic ranges poleward or alter their phenology to cope with changing conditions. Leveraging opportunistic observations of European and North American odonates, they find that species showing significant range shifts also exhibited shifts to earlier emergence. Considering a broad range of potential predictors, their results reveal that geographical factors, but not functional traits, are associated with these shifts.

Strengths:

The article addresses an important topic in ecology and conservation that is particularly timely in the face of reports of substantial insects declines in North America and Europe over the past decades. Through data integration the authors leverage the rich natural history record for odonates, broadening the taxonomic scope of analyses of temporal trends in phenology and distribution. The combination of phenological and range shifts in one framework presents an elegant way to reconcile previous findings and informs about the drivers of biodiversity loss.

Weaknesses:

To better understand whether species shifting both their ranges and phenology are more successful, or as stated here are 'clear winners', and hence whether those that do neither are more vulnerable would require integrating population trends alongside the discussed response. The ~10% species that have not shifted their distribution or phenology might have not declined in abundance, if they have rapidly adapted to local changes in climatic conditions (i.e. they might show a plastic response). These species might be the real 'winners', while species that have recently shifted their ranges or phenology may eventually reach hard limits. The authors are discussing this limitation but might want to adapt their wording, given the potential for misinterpretation. The finding that species with more northern ranges showed lesser northward shifts would speak to the fact that some species have already reached such a geographical range limit.

Achievements and impact:

The results support broad differences in the response of odonate species to climate change, and the prediction that range geography and temperature seasonality are more important predictors of these changes than functional traits. Simultaneously addressing range and phenological shifts highlights that most species exhibit coupled responses but also identifies a significant portion of species that do not respond in these ways that are of critical conservation concern. These results are important for improving forecasts of species' responses to climate change and identifying species of particularly conservation concern. Although not exhaustive regarding abundance trends, the study presents an important step towards a general framework for investigating the drivers of multifaceted species responses.

Author response:

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

Reviewer #1 (Public review):

Sumary:

This study evaluates whether species can shift geographically, temporally, or both ways in response to climate change. It also teases out the relative importance of geographic context, temperature variability, and functional traits in predicting the shifts. The study system is large occurrence datasets for dragonflies and damselflies split between two time periods and two continents. Results indicate that more species exhibited both shifts than one or the other or neither, and that geographic context and temp variability were more influential than traits. The results have implications for future analyses (e.g. incorporating habitat availability) and for choosing winner and loser species under climate change. The methodology would be useful for other taxa and study regions with strong community/citizen science and extensive occurrence data.

We thank Reviewer 1 for their time and expertise in reviewing our study. The suggestions are very helpful and will improve the quality of our manuscript.

Strengths:

This is an organized and well-written paper that builds on a popular topic and moves it forward. It has the right idea and approach, and the results are useful answers to the predictions and for conservation planning (i.e. identifying climate winners and losers). There is technical proficiency and analytical rigor driven by an understanding of the data and its limitations.

We thank Reviewer 1 for this assessment.

Weaknesses:

(1) The habitat classifications (Table S3) are often wrong. "Both" is overused. In North America, for example, Anax junius, Cordulia shurtleffii, Epitheca cynosura, Erythemis simplicicollis, Libellula pulchella, Pachydiplax longipennis, Pantala flavescens, Perithemis tenera, Ischnura posita, the Lestes species, and several Enallagma species are not lotic breeding. These species rarely occur let alone successfully reproduce at lotic sites. Other species are arguably "both", like Rhionaeschna multicolor which is mostly lentic. Not saying this would have altered the conclusions, but it may have exacerbated the weak trait effects.

We thank the reviewer for their expertise on this topic. We obtained these habitat classifications from field guides and trait databases, and reviewed our primary sources to clarify the trait classifications. We reclassified the species according to the expertise of this reviewer and perform our analysis again; please see details below.

(2) The conservative spatial resolution (100 x 100 km) limits the analysis to wide- ranging and generalist species. There's no rationale given, so not sure if this was by design or necessity, but it limits the number of analyzable species and potentially changes the inference.

It is really helpful to have the opportunity to contextualize study design decisions like this one, and we thank the reviewer for the query. Sampling intensity is always a meaningful issue in research conducted at this scale, and we addressed it head-on in this work.

Very small quadrats covering massive geographical areas will be critically and increasingly afflicted by sampling weaknesses, as well as creating a potentially large problem with pseudoreplication. There is no simple solution to this problem. It would be possible to create interpolated predictions of species’ distributions using Species Distribution Models, Joint Species Distribution Models, or various kinds of Occupancy Models. None of these approaches then leads to analyses that rely on directly observed patterns. Instead, they are extrapolations, and those extrapolations typically fail when tested, although they have still been tested (for example, papers by Lee-Yaw demonstrate that it is rare for SDMs to predict things well; occupancy models often perform less well than SDMs and do not capture how things change over time - Briscoe et al. 2021, Global Change Biology). The result of employing such techniques would certainly be to make all conclusions speculative, rather than directly observable. 

Rather than employing extrapolative models, we relied on transparent techniques that are used successfully in the core macroecology literature that address spatial variation in sampling explicitly and simply. Moreover, we constructed extensive null models that show that range and phenology changes, respectively, are contrary to expectations that arise from sampling difference. 100km quadrats make for a reasonable “middle-ground” in terms of the effects of sampling, and we added a reference to the methods section to clarify this (see details below).

(3) The objective includes a prediction about generalists vs specialists (L99-103) yet there is no further mention of this dichotomy in the abstract, methods, results, or discussion.

Thank you for pointing this out - it is an editing error that should have been resolved prior to submission. We replaced the terms specialist and generalist with specific predictions based on traits (see details below).

(4) Key references were overlooked or dismissed, like in the new edition of Dragonflies & Damselflies model organisms book, especially chapters 24 and 27.

We thank Reviewer 1 for making us aware of this excellent reference. We have reviewed the text and include it as a reference, in addition to other references recommended by Reviewer 1 and other reviewers (see details below).

Reviewer #2 (Public review):

Summary:

This paper explores a highly interesting question regarding how species migration success relates to phenology shifts, and it finds a positive relationship. The findings are significant, and the strength of the evidence is solid. However, there are substantial issues with the writing, presentation, and analyses that need to be addressed. First, I disagree with the conclusion that species that don't migrate are "losers" - some species might not migrate simply because they have broad climatic niches and are less sensitive to climate change. Second, the results concerning species' southern range limits could provide valuable insights. These could be used to assess whether sampling bias has influenced the results. If species are truly migrating, we should observe northward shifts in their southern range limits. However, if this is an artifact of increased sampling over time, we would expect broader distributions both north and south. Finally, Figure 1 is missed panel B, which needs to be addressed.

We thank Reviewer 2 for their time and expertise in reviewing our study.

It is possible that some species with broad niches may not need to migrate, although in general failing to move with climate change is considered an indicator of “climate debt”, signaling that a species may be of concern for conservation (ex. Duchenne et al. 2021, Ecology Letters). We revised the discussion to acknowledge potential differences in outcomes (please see details below).

We used null models to test whether our results regarding range shifts were robust, and if they varied due to increased sampling over time. We found that observed northern range limit shifts are not consistent with expectations derived from changes in sampling intensity (Figure S1, S2). 

We thank Reviewer 2 for pointing out this error in Figure 1. This conceptual figure was a challenge to construct, as it must illustrate how phenology and range shifts can occur simultaneously or uniquely to enable a hypothetic odonate to track its thermal niche over time. In a previous version of the figure, we had a second panel and we failed to remove the reference to that panel when we simplified the figure. We have updated the figure and figure caption (please see details below).

Reviewer #3 (Public review):

Summary:

In their article "Range geographies, not functional traits, explain convergent range and phenology shifts under climate change," the authors rigorously investigate the temporal shifts in odonate species and their potential predictors. Specifically, they examine whether species shift their geographic ranges poleward or alter their phenology to avoid extreme conditions. Leveraging opportunistic observations of European and North American odonates, they find that species showing significant range shifts also exhibited earlier phenological shifts. Considering a broad range of potential predictors, their results reveal that geographical factors, but not functional traits, are associated with these shifts.

We thank Reviewer 3 for their expertise and the time they spent reviewing our study. Their suggestions are very helpful and will improve the quality of our manuscript.

Strengths:

The article addresses an important topic in ecology and conservation that is particularly timely in the face of reports of substantial insect declines in North America and Europe over the past decades. Through data integration the authors leverage the rich natural history record for odonates, broadening the taxonomic scope of analyses of temporal trends in phenology and distribution to this taxon. The combination of phenological and range shifts in one framework presents an elegant way to reconcile previous findings improving our understanding of the drivers of biodiversity loss.

We thank Reviewer 3 for this assessment.

Weaknesses:

The introduction and discussion of the article would benefit from a stronger contextualization of recent studies on biological responses to climate change and the underpinning mechanism.

The presentation of the results (particularly in figures) should be improved to address the integrative character of the work and help readers extract the main results. While the writing of the article is generally good, particularly the captions and results contain many inconsistencies and lack important detail. With the multitude of the relationships that were tested (the influence of traits) the article needs more coherence.

We thank Reviewer 3 for these suggestions. We revised the introduction and discussion to better contextualize species’ responses to climate change and the mechanisms behind them (see details below). We carefully reviewed all figures and captions, and made changes to improve the clarity of the text and the presentation of results (see details below).

Reviewer #1 (Recommendations for the authors):

Comment:

(1) Following weakness #1 in the public review, the authors should review the habitat classifications, consult with an odonatologist, and reclassify many species from Both to Lentic and redo the analysis.

Thank you for pointing out this disagreement among expert habitat classifications that we cited and other literature. We reclassified species’ habitat preferences based on classifications by Hof et al., a source that was consistent with your suggestions, and identified additional species as Lentic that our other references had identified as Both. We performed our analysis with this new dataset and, as you suspected, our results did not change qualitatively: species habitat preferences did not predict their range shifts.

Hof, Christian, Martin Brändle, and Roland Brandl. "Lentic odonates have larger and more northern ranges than lotic species." Journal of Biogeography 33.1 (2006): 63-70.

Comment:

(2) Following weakness #2, would it be worthwhile or interesting to analyze a smaller ranging group (e.g. cut the quad size in half, 50 x 50 km) to bring in more species and potentially change the inference? Or is the paper too tightly constructed to allow this, even as a secondary piece?

Thank you for this comment, as it highlights an important consideration for macroecological analyses, and the importance of balancing multiple factors for determining quadrat size. Issues exist with identifying drivers of range boundaries among species with narrow ranges when they are analyzed separately from wide-ranging species, and examining larger quadrats can actually help clarify drivers (Szabo, Algar, and Kerr 2009). The smaller quadrats are, the higher the likelihood that the species is actually there but was never observed, or that the quadrat only covers unsuitable habitat and the species is absent from the entire (or almost entire) quadrat. Too many absences creates issues with violating model assumptions, and creates noise that makes it difficult to identify drivers of species’ range and phenology shifts.

Moreover, we constructed extensive null models that show that range and phenology changes, respectively, are contrary to expectations that arise from sampling difference. 100km quadrats make for a reasonable “middle-ground”, and we have included a brief explanation of this in the text: “We assigned species presences to 100×100 km quadrats, a scale that is large enough to maintain adequate sampling intensity but still relevant to conservation and policy (Soroye et al., 2020), to identify the best sampled species.”  (Lines 170-172).

Szabo, Nora D., Adam C. Algar, and Jeremy T. Kerr. "Reconciling topographic and climatic effects on widespread and range‐restricted species richness." Global Ecology and Biogeography 18.6 (2009): 735-744.

Comment:

(3) Following weakness #3, are specialists the ones that "failed to shift" (L18)? If so please specify. The prediction about generalists vs specialists needs to be removed or incorporated in other parts of the paper.

Thank you for pointing this out, we intended to suggest that species with more generalist habitat requirements might be better able to shift, but ultimately found that traits did not predict species’ shifts. We corrected our prediction regarding habitat generalists as follows: “We predicted that species able to use both lentic and lotic habitats would shift their phenologies and geographies more than those able to use just one habitat type, as generalists outperform specialists as climate and land uses change (Ball-Damerow et al., 2015, 2014; Hassall and Thompson, 2008; Powney et al., 2015; Rapacciuolo et al., 2017).” (Lines 128-132).

Comment:

(4) Following weakness #4, cite Pinkert et al at lines 70-73 and Rocha-Ortega et al at lines 73-77 along with https://doi.org/10.1098/rspb.2019.2645. Add Sandall et al https:// doi.org/10.1111/jbi.14457 to L69 references.

Thank you for the excellent reference suggestions, we have added them as suggested (Lines 80, 86, 77).

Comment:

Other comments/suggestions:

(1) Title: consider adding temp variability 'Range geography and temperature variability, not functional traits,...'.

Thank you for this suggestion, we have added temperature variability to the title: “Range geography and temperature variability explain cross-continental convergence in range and phenology shifts in a model insect taxon”.

Comment:

(2) L125: is (northern) Mexico included in North America?

Yes, we did include observations from Northern Mexico, and have specified this in the text: “We retained ~1,100,000 records from Canada, the United States, and Northern Mexico, comprising 76 species (Figure 2).” (Lines 174-176).

Comment:

(3) L128: I'd label this section 'Temperature variability' rather than 'Climate data'.

Thank you, we agree that this is a more appropriate title for this section, and have replaced ‘Climate data’ with ‘Temperature variability’ (Line 185).

Comment:

(4) Table 2: why are there no estimates for the traits?

We apologise, this information should have been included in the main body of the manuscript, but was only explained in the Table 2 caption. We have added the following explanation: “Non-significant variables, specifically all functional traits, were excluded from the final models.”. (Line 312-323).

Comment:

(5) Figure 2: need to identify the A-D panels.

We apologise for this error and have clarified the differences between panels in the figure caption:

“Figure 2: Richness of 76 odonate species sampled in North America and Europe in the historic period (1980-2002; panes A and C) and the recent period (2008-2018; panes B and D). Species richness per 100 × 100 km quadrat is shown in panes A and B, while panes C and D show species richness per 200 × 200 km quadrat. Dark red indicates high species richness, while light pink indicates low species richness.” (Lines 1002-1006).

Comment:

(6) L163-173: I am not familiar with this analysis but it sounds interesting and promising, I am not sure if this can be clarified further. Why the -25 to 25, and -30 to 30, doesn't the -35 to 35 cover these? And what is meant by "include only phenology shifts that could be biologically meaningful", that larger shifts would not be meaningful or tied to climate change?

We used different cutoffs for phenology shifts to inspect for outliers that were likely to be errors, potentially do to insufficient sampling to calculate phenology. We clarified in the text as follows:

“We retained emergence estimates between March 1st and September 1st, as well as species and quadrats that showed a difference in emergence phenology of -25 to 25 days, -30 to 30 days, or -35 to 35 days between both time periods, to include only phenology shifts that could be biologically meaningful to environmental climate change (i.e. exclude errors).” (Lines 169-173).

Comment:

(7) L193-200: I agree but would make a distinction between ecological vs functional traits, as other studies view geographic traits as ecological manifestations of functional biology, e.g. https://doi.org/10.1016/j.biocon.2019.07.001 and https://doi.org/10.1016/ j.biocon.2023.110098.

Thank you for this suggestion, and for making us aware of the thinking around range geographies as ecological traits. We have specified throughout the manuscript that the ‘traits’ we are considering are ‘functional traits’, changed the methods subsection title to “Range geographies and functional traits” (Line 252), and added a brief discussion of ecological traits: “Geographic range and associated climatic characteristics are often considered ecological traits, as they are consequences of functional traits and their interactions with geographic features (Bried and Rocha-Ortega, 2023; Chichorro et al., 2019).” (Lines 256-259).

Comment:

(8) L203: What's the rationale for egg-laying habitat as "biologically relevant to spatial and temporal responses to climate change"? That one's not as obvious as the others and needs a sentence more. Also, I am wondering why other traits were not considered here, like color lightness and voltinism. And why not wing size instead of body size, or better yet the two combined (wing loading) as a proxy for dispersal ability?

We agree that our rationale for using this trait should be better explained, and we have included the following explanation: “Egg laying habitat was assigned according to whether species use exophytic egg-laying habitat (i.e. eggs laid in water or on land, relatively larger in number), or endophytic egg-laying habitat (i.e. eggs laid inside plants, usually fewer in number); species using exophytic habitats are associated with greater northward range limit shifts (Angert et al., 2011).” (Lines 271-275).

We considered traits that have been found to be important for range and phenology shifts among odonates, as well as being key traits for expectations for species responses to climate change. Flight duration and body size are correlated with dispersal ability (Powney et al. 2015). Body size is also correlated with competitive ability (Powney et al. 2015), potentially making it an important predictor of a species’ ability to establish and maintain populations in expanding range areas. Traits correlated with range shifts also include breeding habitat type (Powney et al. 2015; Bowler et al. 2021) and egg laying habitat (Angert et al. 2011). Ideally, we would have used dispersal data from mark/release/recapture studies, but it was not available for many of the species included in this study. After finding that none of the functional traits we included were related to range shifts, there was no reason to believe that a further investigation of traits would be meaningful.

Angert AL, Crozier LG, Rissler LJ, Gilman SE, Tewksbury JJ, Chunco AJ. 2011. Do species’ traits predict recent shifts at expanding range edges? Ecology Letters 14:677–689. doi:10.1111/j.1461-0248.2011.01620.x

Bowler DE, Eichenberg D, Conze K-J, Suhling F, Baumann K, Benken T, Bönsel A, Bittner T, Drews A, Günther A, Isaac NJB, Petzold F, Seyring M, Spengler T, Trockur B, Willigalla C, Bruelheide H, Jansen F, Bonn A. 2021. Winners and losers over 35 years of dragonfly and damselfly distributional change in Germany.Diversity and Distributions 27:1353–1366. doi:10.1111/ddi.13274

Powney GD, Cham SSA, Smallshire D, Isaac NJB. 2015. Trait correlates of distribution trends in the Odonata ofBritain and Ireland. PeerJ 3:e1410. doi:10.7717/peerj.1410

Comment:

(9) L210: I count at least 5 migratory species in table S3, so although maybe not enough to analyze it's misleading to say "nearly all" were non-migratory, revise to "most" or "vast majority".

Thank you for pointing this out, we have made the suggested correction (Line 277).

Comment:

(10) L252-254: save this for the Discussion and write a more generalized statement for results to avoid citations in the results.

Thank you for this suggestion, we have moved this to the discussion (Lines 517-527).

Comment:

(11) Figures S5 & S6: these are pretty important, I'd consider elevating them to the main document as one figure with two panels.

Thank you for this suggestion, we agree these figures should be elevated to the main text, and have made them into a panel figure (Figure 4).

Comment:

(12) L305-307: great point and recommendation!

Thank you very much for this positive feedback!

Comment:

(13) L335-336: another place to cite https://doi.org/10.1098/rspb.2019.2645 which includes a thermal sensitivity index and would add an odonate citation behind the statement.

Thank you for this excellent suggestion, we have added this citation (line 480). (Rocha-Ortega et al. 2020)

Comment:

(14) L352-353: again see also https://doi.org/10.1098/rspb.2019.2645.

Thank you for highlighting this reference, we have added it to Line 505 as suggested.

Comment:

(15) L355: revise "populations that coexist" to "species that co-occur" (big difference between population and species levels and between coexistence and co-occurrence).

Thank you very much for pointing this out, we have made the suggested change (Line 507).

Comment:

(16) L359-365: are the winners and losers depicted in Figures S5 & S6? If so reference the figure (which I suggest combining and promoting to the main text), if not create a table listing the analyzed species and their winner/loser status.

We agree that this is an excellent place to bring up Figures S5 and S6 from the supplemental. We have moved them to the main document as one figure and referenced it at line 510.

Reviewer #2 (Recommendations for the authors):

Comment:

(1) Line 53-55: The claim that "These relationships generalize poorly taxonomically and geographically" is valid, but the study only tests Odonata on two continents.

Thank you for this comment – the word ‘generalize’ may imply that our study tries to find a general pattern across many groups. We have changed the language to: “However, these relationships are inconsistent across taxa and regions, and cross-continental tests have not been attempted (Angert et al., 2011; Buckley and Kingsolver, 2012; Estrada et al., 2016; MacLean and Beissinger, 2017).” (Lines 57-59).

Comment:

(2) Line 58-59: Is this statement only true for Odonata? It does not seem to hold for plants, for example.

Thank you for this comment – this statement references a meta-analysis of multiple animal and plant taxa, but the evidence for the importance of range location comes from animal taxa. We have specified that we are referring to animal species to clarify (Line 60).

Comment:

(3) Line 87-91: This section is difficult to understand and needs clarification.

We have clarified this section as follows: “While warm-adapted species with more equatorial distributions could expand their ranges poleward following warming (Devictor et al., 2008), they could also increase in abundance in this new range area relative to species that historically occupied those areas and are less heat-tolerant (Powney et al., 2015).” (Lines 95-121).

Comment:

(4) Line 99-100: Please define "generalist" and "specialist" more clearly here (e.g., based on climate niche?).

Thank you for pointing this out, we intended to suggest that species with more generalist habitat requirements might be better able to shift, but ultimately found that traits did not predict species’ shifts. We corrected our prediction regarding habitat generalists as follows: “We predicted that species able to use both lentic and lotic habitats would shift their phenologies and geographies more than those able to use just one habitat type, as generalists outperform specialists as climate and land uses change (Ball-Damerow et al., 2015, 2014; Hassall and Thompson, 2008; Powney et al., 2015; Rapacciuolo et al., 2017).” (Lines 128-132).

Comment:

(5) Line 122: Replace the English letter "X" in "100x100 km" with the correct mathematical symbol.

We have made the suggested replacement throughout the manuscript.

Comment:

(6) Line 148: To address sampling effects, you could check the paper: https://onlinelibrary.wiley.com/doi/full/10.1111/gcb.15524. Additionally, maximum and minimum values are sensitive to extreme data points, so using 95% percentiles might be more robust.

Thank you for sharing this paper, as it offers a valuable perspective on the study of species’ ranges. While our dataset is substantially composed of observations from adult sampling protocols, unlike the suggested paper which compares adults and juveniles, this is an interesting alternative approach.

For our purposes it is meaningful to include outliers, as otherwise we may have missed individuals at the leading edge of range expansions. Our intent here was to detect range limits, as opposed to finding the central tendency of species distributions. This approach is widely accepted in the macroecology literature (i.e. Devictor et al., 2012, 2008; Kerr et al. 2015).

We have included the following discussion of our approach in the methods section:

“We followed widely accepted methods to determine species range boundaries (Devictor et al., 2012, 2008; Kerr et al., 2015), although other methods exist that are appropriate for different data types and research questions i.e. (Ni and Vellend, 2021). We assigned species presences to 100×100 km quadrats, a scale that is large enough to maintain adequate sampling intensity but still relevant to conservation and policy (Soroye et al., 2020), to identify the best sampled species.” (Lines 168-173).

Kerr JT, Pindar A, Galpern P, Packer L, Potts SG, Roberts SM, Rasmont P, Schweiger O, Colla SR, Richardson LL,Wagner DL, Gall LF, Sikes DS, Pantoja A. 2015. Climate change impacts on bumblebees converge across continents. Science 349:177–180. doi:10.1126/science.aaa7031

Soroye P, Newbold T, Kerr J. 2020. Climate change contributes to widespread declines among bumble bees across continents. Science 367:685–688. doi:10.1126/science.aax8591

Devictor V, Julliard R, Couvet D, Jiguet F. 2008. Birds are tracking climate warming, but not fast enough.Proceedings of the Royal Society B: Biological Sciences 275:2743–2748. doi:10.1098/rspb.2008.0878

Devictor V, van Swaay C, Brereton T, Brotons L, Chamberlain D, Heliölä J, Herrando S, Julliard R, Kuussaari M,Lindström Å, Reif J, Roy DB, Schweiger O, Settele J, Stefanescu C, Van Strien A, Van Turnhout C,

Vermouzek Z, WallisDeVries M, Wynhoff I, Jiguet F. 2012. Differences in the climatic debts of birds and butterflies at a continental scale. Nature Clim Change 2:121–124. doi:10.1038/nclimate1347

Comment:

(7) Line 195: The species' climate niche should also be considered a product of evolution.

Thank you for this suggestion. To address this comment and a comment from another reviewer, we changed the text to the following: “Geographic range and associated climatic characteristics are often considered ecological traits, as they are consequences of functional traits and their interactions with geographic features (Bried and Rocha-Ortega, 2023; Chichorro et al., 2019).” (Lines 256-259).

Comment:

(8) Line 244: This speculative statement belongs in the Discussion section.

Thank you for this suggestion, we have moved this statement to the discussion (Lines 451-453).

Comment:

(9) Line 252-254: The projection of Coenagrion mercuriale's range contraction is not part of your results and should be clarified or removed.

Following this suggestion and a similar suggestion from another reviewer, we moved this text to the discussion (Line 517-527).

Comment:

(10) Line 314-316: If the species can tolerate warmer temperatures better, why would they migrate?

We apologize for the confusion, and we have reworded the section as follows: “Emerging mean conditions in areas adjacent to the ranges of southern species may offer opportunities for range expansions of these relative climate specialists, which can then tolerate climate warming in areas of range expansion better than more cool-adapted historical occupants (Day et al., 2018).” (Lines 445-448).

Comment:

(11) Line 334-335: Species' tolerance to temperature likely depends on their traits, which were not tested in this study. This should be noted.

We agree, and we have removed the wording “rather than traits” from this sentence (Line 479).

Reviewer #3 (Recommendations for the authors):

Comment:

(1) Title: The title is too general not specifying that your results are on odonates only, but also stressing the implicit role of climate change to a degree the tests do not support.

Following this comment and a suggestion from another reviewer we changed the title to the following: “Range geography and temperature variability explain cross-continental convergence in range and phenology shifts in a model insect taxon”. We wanted to emphasize our use of Odonates as a model species that we used to ask broad questions, while being more specific about the climatic variable that we examined (temperature variability).

Comment:

(2) L32: consider including Novella-Fernandez et al. 2023 (NatCommun) which addresses this topic in Odonates.

Thank you for suggesting this very interesting paper, we have added it as a citation (Line 31-32).

Comment:

(3) L35: consider including Grewe et al. 2013 (GEB) and Engelhardt et al. 2022(GCB).

Thank you for these excellent suggestions, we have added the citations (Line 35).

Comment:

(4) L47: rather write 'result from' instead of 'driven by'.

We agree this is a better characterization and have corrected the wording (Line 48-49).

Comment:

(5) L49-52: There has been a recent study on this topic for birds (Neate-Clegg et al., 2024 NEE). However, specifying this to insects would make it not less relevant. This review for odonates might be helpful in this regard (Pinkert et al.. 2022, Chapter: "Odonata as focal taxa for biological responses to climate change" IN Dragonflies & Damselflies: Córdoba-Aguilar et al. (2022) Model Organisms for Ecological and Evolutionary Research.

Thank you for again suggesting excellent references, we have added them to line 52-53, as well as adding the Pinkert citation to lines 61 and 82.

Comment:

(6) L53-66: Combine into one paragraph about drivers. With traits first and the environment second. The natural land cover perspective may be too complicated in this context. Consider focusing on generalities of the impact of changes within species' ranges.

As suggested we have combined these into one paragraph about drivers (Line 59).

Comment:

(7) L67-69: The book from before would be a much stronger reference for this claim. Kalkmann et al (2018) do not address the emphasis of global change research in insects on bees and butterflies. Also, I would highlight that most of the current work is at a national scale, rather than cross-continental.

Thank you for this suggestion, we have added the suggested reference and included that “…recently assembled databases of odonate observations provide a rare opportunity to investigate species’ spatiotemporal responses at larger taxonomic and spatial scales, particularly as most work has been done at national scales.” (Lines 75-77).

Comment:

(8) L68: consider rephrasing this part to '..provide a rare opportunity to investigate spatiotemporal biotic responses at larger taxonomic and spatial scales'

We appreciate this suggestion and really like the wording. We have changed the phrase to read as follows: “While global change research on insects often emphasizes butterfly and bee taxa, recently assembled databases of odonate observations provide a rare opportunity to investigate species’ spatiotemporal responses at larger taxonomic and spatial scales, particularly as most work has been done at national scales.” (Lines 74-77).

Comment:

(9) L69: This characteristic is not unique to odonates and would hamper drawing general conclusions. Honestly, I think the detailed and comprehensive data on them is the selling point.

Thank you for this suggestion, we have edited the sentence to emphasize their use as an indicator species: “Due to their use of aquatic and terrestrial habitat across life different stages, dragonflies and damselflies are also considered indicator species for both terrestrial and aquatic insect responses to changing climates (Hassall, 2015; Pinkert et al., 2022; Šigutová et al., 2025), giving the study of these species broad relevance for conservation.” (Lines 78-81)

Comment:

(10) L73: Indicator for what? The first part of the sentence would suggest lesser surrogacy for responses of other taxa. Reconsider this statement. They are well- established indicators for habitat intactness and freshwater biodiversity. Darwell et al. suggested their diversity can serve as a surrogate for the diversity of both terrestrial and aquatic taxa.

Thank you for this suggestion, we have edited the sentence to emphasize their use as an indicator species: “Due to their use of aquatic and terrestrial habitat across life different stages, dragonflies and damselflies are also considered indicator species for both terrestrial and aquatic insect responses to changing climates (Hassall, 2015; Pinkert et al., 2022; Šigutová et al., 2025), giving the study of these species broad relevance for conservation.” (Lines 78-81)

Comment:

(11) L76: Fritz et al., is a study on mammals, not odonates.

Thank you for pointing out this error, the reference has been removed (Line 84-85).

Comment:

(12) L84: Lotic habitats are generally better connected than lentic ones. Lentic species are considered to have a greater propensity for dispersal DUE to the lower inherent spatiotemporal stability (implying lower connectivity) compared to lotic habitats.

Thank you for your comment, we have rewritten this section as follows: “For example, differences in habitat connectivity and dispersal ability may constrain range shifts for lentic species (those species that breed in slow moving water like lakes or ponds) and lotic species (those living in fast moving-water) in different ways (Kalkman et al., 2018). More southerly lentic species may expand their range boundaries more than lotic species, as species accustomed to ephemeral lentic habitats better dispersers (Grewe et al., 2013), yet lotic species have also been found to expand their ranges more often than lentic species, potentially due to the loss of lentic habitat in some areas (Bowler et al., 2021).” (Lines 88-95).

Comment:

(13) L90: I would be cautious with this interpretation. If only part of the range is considered (here a country in the northern Hemisphere) southern species are moving more of their range into and northern species more of their range out of the study area in response to warming (implying northward shifts).

We have clarified this section as follows: “While warm-adapted species with more equatorial distributions could expand their ranges poleward following warming (Devictor et al., 2008), they could also increase in abundance in this new range area relative to species that historically occupied those areas and are less heat-tolerant (Powney et al., 2015).” (Lines 95-121)

Comment:

(14) L117: Odonata Central contains many county centroids as occurrence records. These could be an issue for your use case. I may have overlooked the steps you took to address this, but I think this requires at least more detail and possibly further removal/checks using for instance CoordinateCleaner. The functions implemented in this package allow you to filter records based on political units to avoid exactly this source of error.

Thank you for this suggestion, we weren’t aware of this issue with Odonata Central. We used the CoordinaterCleaner tool in R to filter all odonate records that we used in our analyses. Less than 1% of observations in our dataset were identified as having potential problems by the tool, so we would not expect this to affect our inferences. However, in future we will employ this tool when using similar datasets.

Comment:

(15) L119: Please add a brief explanation of why this was necessary. I am ok with something along the lines in the supplement.

We moved this information from the supplemental to the main text as follows: “If a species was found on both continents, we only retained observations from the continent that was the most densely sampled. If we merged data for one species found on both continents, we could not perform a cross-continental comparison. However, if the same species on different continents was treated as different species, this would lead to uninterpretable outcomes (and the creation of pseudo-replication) in the context of phylogenetic analyses. In addition, species found on both continents did not have sufficient data to meet criteria for the phenology analysis.” (Lines 161-167).

Comment:

(16) L132: This is the letters 'X' or 'x' are not multiplier symbols! Please change to the math symbol (×), everywhere.

Thank you for pointing out this error, we have made the correction throughout the manuscript.

Comment:

(17) L133: add 'main' before 'flight period'

Thank you for this suggestion, we have made the change. (Line 190)

Comment:

(18) L135: I suggest using the coefficient of variation, as it is controlled for the mean. Otherwise, what you see is partly the signature of temperature and not of its variation. For me, it's very difficult to understand what this variation of the variation means and at least needs more explanation.

Thank you very much for this suggestion, we agree that using the coefficient of variation is a better fit for the question that we’re asking. We re-ran out analyses with the coefficient of variation as the measure of climate variability: all the results reported in the manuscript are now updated for that analysis (Line 377, Table 2), and we have also updated the methods section (Line 191). The results are qualitatively the same to our previous analysis, but we agree that they are now easier to interpret.            

Comment:

(19) L155: Please adequately reference all R packages (state the name, and a reference for them including the authors' names, title, and version).

Thank you for pointing out this omission, we have added reference information for the glm function in base R (Line 298) and ensured all other packages are properly referenced.

Comment:

(20) L207: Mention the literature sources here (again).

We agree that they should be referenced here again, and we have done so (Lines 267-268).

Comment:

(21) L209: You could use the number of grid cells as a proxy for range size.

Following this excellent suggestion, we re-analysed our data using range size, calculated as the number of quadrats occupied by a species in the historical time period, as a predictor. Range size was not significant in our models, but we believe this is the best way to analyze our data, and so have updated our methods (Lines 261-263) and results (375-378).

Comment:

(22) L218: It would be preferable to say 'species-level' instead of 'by-species'.

Thank you for this suggestion, we agree that this is clearer and made the change (Line 298).

Comment:

(23) L219-220: this is unclear. Please rephrase.

We have clarified as follows: “We used both species-level frequentist (GLM; glm function in R) and Bayesian (Markov Chain Monte Carlo generalized linear mixed model, MCMCglmm; Hadfield, 2010) models to improve the robustness of the results.” (Lines 298-300).

Comment:

(24) L224: At least for Europe there is a molecular phylogeny available, which you should preferably use (Pinkert et al. 2018, Ecography). Otherwise, I am ok with using what is available

We apologize that the nature of the phylogeny that we used was not clear; the phylogeny that we used was built similarly to that in Pinkert et al. 2018, Ecography. It created a molecular phylogeny with a morphological/taxonomic tree as the backbone tree, so that species could only move within their named genera or families. We clarified this in the manuscript as follows:

“We used the molecular phylogenetic tree published by the Odonate Phenotypic Database (Waller et al., 2019), which used a morphological and taxonomic phylogeny as the backbone tree, allowing species to move within their named genera or families according to molecular evidence (Waller and Svensson, 2017).” (Lines 302-305).

Comment:

(25) L233: You said so earlier (1st sentence of this paragraph).

Thank you for pointing this out, we removed the repetitive sentence (Line 323).

Comment:

(26) L236-238: To me, it makes more sense to test this prior to fitting the phylogenetic models.

MCMC-GLMM is considerably less familiar to most researchers than general linear models or there derivatives/descendants, such as PGLS. We report models both with and without phylogenetic relationships included for the sake of transparency, and we are happy to acknowledge that no interpretation here changes substantially relative to these decisions. However, failing to report models that included possible (if small) effects of phylogenetic relatedness might cause some readers to question what those models might have implied. For the moment, we are opting for the most transparent reporting approach here.

Comment:

(27) L241: Rather say directly XX of XX species in our data....

(28) L245: Same here. Provide the actual numbers, please.

Thank you for this suggestion, we made this change on Line 332 and Line 334.

Comment:

(29) L247-249: Then not necessary.

This issue highlights a challenge in the global biology literature and around the issue of biodiversity monitoring for understanding global change impacts on species. Almost no studies have been able to report simultaneous range and phenology shifts, and the literature addresses these biotic responses to global change predominantly as distinct phenomena. Differences in numbers of species for which these observations exist, even among the extremely widely-observed odonates, seems to us to be a meaningful issue to report on. If the reviewer prefers that we abbreviate or remove this sentence, we are happy to do so.

Comment:

(30) L251:261: That is discussion as you interpret your results.

Following your suggestion and the suggestion of another reviewer, we moved the following lines to the discussion section: “Species that did not shift their ranges northwards or advance their phenology included Coenagrion mercuriale, a European species that is listed as near threatened by the IUCN Red List (IUCN, 2021), and is projected to lose 68% of its range by 2035 (Jaeschke et al., 2013).” (Lines 517-527).

Comment:

(31) 252: Good to mention, but why is the discussion limited to C. mercurial?

We feel that it is important to link the broad-scale results to the specific biological characteristics of individual species, and C. mercurial is an IUCN threatened species. We are happy to expand links to natural history of this group and have added the following: “This group also includes Coenagrion resolutum, a common North American damselfly (Swaegers et al., 2014), for which we could not find evidence of decline. This may be due in part to the greater area of intact habitat available in North American compared to Europe, enabling C. resolutum to maintain larger populations that are less vulnerable to stochastic climate events. Still, this and other species failing to shift in range or phenology should be assessed for population health, as this species could be carrying an unobserved extinction debt.” (Lines 527-533).

Comment:

(32) L264: Insert 'being' before 'consistently'.

Thank you for the suggestion, we made this change (Line 373).

Comment:

(33) L271: .'. However,'.

Thank you for pointing out this grammatical error, we have corrected it (Line 382).

Comment:

(34) L273: 'affected' instead of 'predicted'

Thank you for the suggestion, we made this change (Line 383).

Comment:

(35) L279: 'despite pronounced recent warming' sounds not relevant in this context.

Thank you for this suggestion, we removed this portion of the sentence (Line 408).

Comment:

(36) L281: Rather 'the model performance did not improve....'

Thank you for the suggestion, we made this change (Line 409).

Comment:

(37) L288: Add 'but' before 'not'.

Thank you for the suggestion, we made this change (Line 416).

Comment:

(38) L311-316: Reconsider the causality here. maybe rather rephrase to are associated instead. Greater dispersal ability and developmental plasticity might well lead to higher growth rates, rather than the other way around.

We agree that plasticity/evolution at range edges is important to consider and have included it as an alternative explanation: “Adaptive evolution and plasticity may enable higher population growth rates in newly-colonized areas (Angert et al., 2020; Usui et al., 2023), but this possibility can only be directly tested with long term population trend data.” (Line 449-451).  

Comment:

(39) L313-316: Maybe delete the second 'should be able to'.

This phrase has been changed in response to other reviewer comments and now reads as follows:

“Emerging mean conditions in areas adjacent to the ranges of southern species may offer opportunities for range expansions of these relative climate specialists, which can then tolerate climate warming in areas of range expansion better than more cool-adapted historical occupants (Day et al., 2018).” (Lines 445-448).

Comment:

(40) L331: Limit this statement ending with 'in North American and European Odonata'.

Thank you for this suggestion, we made this addition (Lines 475-476).

Comment:

(41) L346-347: There are too many of these more-research-is-needed statements in the discussion (at least three in the last paragraphs). Please consider finishing the paragraphs rather with a significance statement.

Thank you for this suggestion, we have changed the final sentence here to the following: “The extent to which species’ traits actually determine rates of range and phenological shifts, rather than occasionally correlated with them, is worth considering further, but functional traits do not systematically drive patterns in these shifts among Odonates in North America and Europe.” (Lines 480-483).

We also made additional changes, removing a ‘more-research is needed’ statement from the following paragraph (Line 443), as well as from line 499.

Comment:

(42) L349: See also Franke et al. (2022, Ecology and Evolution).

Thank you for highlighting this excellent reference! We have added it to Line 501.

Comment:

(43) L363: Maybe a bit late in the text, but it is important to note that there is the third dimension 'abundance trends' or rather a common factor related to range and phenology shifts. I feel this fits better with the discussion of population growth.

Thank you for this suggestion, we have addressed the importance of abundance trends in the following sentences: “Further mechanistic understanding of these processes requires abundance data.” (Lines 442-443); “It remains unclear if range and phenology shifts relate to trends in abundance, but our results suggest that there are clear ‘winners’ and ‘losers’ under climate change.” (Lines 509-510).

Comment:

(44) L375-377: This last sentence is very similar to L371-373. Please reduce the redundancy. Focus more on specifically stating the process instead of vaguely saying 'new insights into patterns' and 'suggesting processes'. Rather, deliver a strong concluding message here.

Thank you for this suggestion, we feel that we now have a much stronger concluding message: “By considering both the seasonal and range dynamics of species, emergent and convergent climate change responses across continents become clear for this well-studied group of predatory insects.” (Lines 545-547).

Comment:

(45) Table 1: To me, the few estimates presented here do not justify a table. rather include them in the text. OR combine them with Table 2. Also, why not include the traits as predictors (from the range shift models) in these models as well?

We have clarified in the text that the results displayed in Table 1 are from the analysis of the relationship between range and phenology shifts: “The effect of species’ range shifts on phenology range shifts was significant in our model investigating the relationship between these responses, indicating that species shifting their northern range limits to higher latitudes also showed stronger advances in their emergence phenology (Figure 3).” (Lines 341-344).

As there were no significant effects in the model of phenology change drivers, we have not shown results of this model: “Emergence phenology shifts were not affected by species’ traits, range geography, nor climate variability; due to this, model results are not displayed here.” (Lines 383-384).

Comment:

(46) Table 2: L712-713: What does this mean? Are phenology shifts not used as a predictor of range shifts? (why then this comment?). Or do you want to say phenological shifts are not related to Southern range etc? Why do you present a phylosig here but not in Table 1? Why not include the traits as predictors (from the range shift models) in these models as well? Consider using the range size as a continuous predictor instead of 'Widespread'.

We are glad the reviewer pointed this out to us. We did not emphasize this issue sufficiently. We DID evaluate traits as predictors both of geographical range and phenological shifts, and species-specific biological traits did not significantly affect models predicting either of those sets of responses. We state this on Lines 312-323, but we have also noted in the discussion (Lines 473-476) that the most commonly assessed traits, like body size, do not alter observed trends here. Instead, where species are found, rather than the characteristics of species, is the key determinant of their overall responses.

Following this excellent suggestion, we re-analysed our data using range size, calculated as the number of quadrats occupied by a species in the historical time period, as a predictor. Range size was not significant in our models, but we believe this is the best way to analyze our data, and so have updated our methods (Lines 261-263) and results (375-378).

Comment:

(47) Figure 1: I don't see any grey points in the figure. Also, there is no A or B. If you are referring to the symbols then write cross and triangle instead and not use capital letters which usually refer to component plots of composite figures. Also, I highly recommend providing a similar figure based on your data (maybe each species as a dot for T1 and another symbol for T2). Given the small number of species, you could try to connect these points with arrows. For the set with only range shifts maybe play the T2-dots at the center of the 'Emergence' axis.

Thank you for pointing out this error: a previous version of Figure 1 included grey points and multiple panels. We have removed this text from the figure caption to be consistent with the final version of the figure (Line 989).

The graphical depictions of the conceptual and empirical discoveries in this paper were challenging to create. The reviewer might be suggesting effectively decomposing Figure 3 (change in range on the y axis vs change in phenology among all species into two sets of points on the same graph, where each pair of points is a before and after value for each species. This would make for a very busy figure indeed. We have modified the conceptual Figure 1 to illustrate more clearly, we believe, that species can (in principle) remain within tolerable niche spaces by shifting their activity periods in time (phenology) or in space (geographical range) or both.

Comment:

(48) Figure 2: Please add a legend. Also black is a poor background color. The maps appear to be stretched. Please check aspect ratios. Now here are capital letters without an explanation in the caption. From the context I assume the upper panel maps are for the data used to calculate range shifts at the bottom panel maps are for data used to calculate the phenological shifts.

We apologise for the error in the figure caption and have clarified the differences between panels in the text, as well as changing the map background colour and fixing the aspect ratio:

“Figure 2: Richness of 76 odonate species sampled in North America and Europe in the historic period (1980-2002; panes A and C) and the recent period (2008-2018; panes B and D). Species richness per 100 × 100 km quadrat is shown in panes A and B, while panes C and D show species richness per 200 × 200 km quadrat. Dark red indicates high species richness, while light pink indicates low species richness.” (Lines 1002-1006).

Comment:

(49) Figure 3: Why this citation? Of terrestrial taxa? Please explain. Consider adding some stats here, such as the r-squared value for each of the relationships.

We have better explained the citation in the figure caption, as well as adding r-squared values:

“Figure 3: Relationship between range shifts and emergence phenology shifts among North American and European odonate species (N = 66; model R2 = 17.08 for glm, 14.9% for MCMCglmm). For reference, the shaded area shows mean latitudinal range shifts of terrestrial taxa as reported by Lenoir et al. (2020; calculated as the yearly mean dispersal rate of 1.11 +/- 0.96 km per year over 38 years).” (Lines 679-682)

Comment:

(50) L801: What are these underscored references?

This was an issue with the reference software and has been resolved.

Comment:

(51) Table S1: L848: Consider starting with 'Samples of 76 North American and European odonate species from between ...'. Please use a horizontal line to separate the content from the table header. Add a horizontal line below the last row. Same for all tables.

Thank you for this suggestion, we have edited the caption for Figure S1 as suggested (Line 1124). We have also made the suggested line additions to Table S1, S2, and S3.

Comment:

(52) Table S3: This is confusing. In Table 1 (main text) both 'southern range' and 'widespread' are used as predictors. Please explain.

We originally included information on species range geography, including southern versus northern range, and widespread versus not, into one categorical variable. Following additional comments we re-analysed our data using range size, calculated as the number of quadrats occupied by a species in the historical time period, as a predictor. Now the methods section text (Lines 261-263) and Table 1 report results of that variable with distribution options northern, southern, or both. 

Comment:

(53) Figure S5 and S6: It would be more coherent if the colors refer to the continents and the suborders are indicated by shading. I would love to see a combination of the two figures with species ordered by the phylogenetic relationship and a dot matrix indicating the traits in the main text! This could really be a good starting point for a synthesis figure.

The reviewer presents an interesting challenge for us. We have a choice, as we understand things, to present a figure showing phylogeny and traits (as requested here), or an ordered list of species relative to effect sizes in the two main responses to global change. The latter choice centers on the discoveries of the paper, while the former would be valuable for dragonfly biology but would depict information that proved to be biologically uninformative relative to our discovery. That is to say, there is no phylogenetic trend and biological traits among species did not affect results. We have gone some way toward illustrating that issue by retaining phylogeny in the MCMC-GLMM models, but we feel that a figure illustrating phylogeny and traits would (for most readers, at least) illustrate noise, rather than signal. For this reason, we have opted to take on the previous reviewer’s suggestion for a modified, main-text Figure 4, which we include below.

Figure 4: Distribution of Northern range limit shifts (Panel A, kilometers) and emergence phenology shift (Panel B, Julian day) of 76 European and North American odonate species between a recent time period (2008 - 2018) and a historical time period (1980 - 2002). Anisoptera (dragonflies) are shown in pink, Zygoptera (damselflies) are shown in blue.

Change last: Figure 3: Relationship between range shifts and emergence phenology shifts among North American and European odonate species (N = 66; model R2 = 17.08 for glm, 14.9% for MCMCglmm). For reference, the shaded area shows mean latitudinal range shifts of terrestrial taxa as reported by Lenoir et al. (2020; calculated as the yearly mean dispersal rate of 1.11 +/- 0.96 km per year over 38 years).

eLife Assessment

This study is a valuable contribution to the field of neuronal modeling by way of providing a method for rapidly obtaining neuronal physiology parameters from electrophysiological recordings. The method is solid as the generated models reproduce both ground-truth simulated data and empirical data, and there is now a quantitative comparison with other approaches.

Reviewer #2 (Public review):

Summary:

Developing biophysically detailed computational models that accurately capture the characteristic physiological properties of neurons across diverse cell types is a key challenge in computational neuroscience. A major obstacle lies in determining the large number of model parameters, which are notoriously difficult to fit such that the model faithfully reproduces the empirically observed electrophysiological responses. Existing approaches require substantial computational resources to generate models for even a single neuron. Generating models for additional neurons typically requires starting from scratch, with no reuse of previous computations - making the process just as computationally expensive each time.

Kim et al. introduce an innovative approach based on a Generative Adversarial Network (GAN) to overcome these limitations. Once trained, the network takes empirically observed electrophysiological responses as input and predicts the biophysical parameters with which a Hodgkin-Huxley model can reproduce these responses. The authors demonstrate this for nine non-spiking neurons in C. elegans. The resulting models generally provide a good fit to the empirical data. As the GAN has learned general relationships between biophysical parameters and the resulting electrophysiology, it can be used to generate models of diverse cell types without retraining - enabling model generation at low computational cost.

Strengths:

The authors address an important and technically challenging problem. A noteworthy strength of their approach is that, once trained, the GAN can generate models from new empirical data at low computational cost. The generated models reproduce the responses to current injections well.

The authors have addressed all of my previous major concerns and have significantly improved their method:

(1) Most importantly, the generated models reproduce both ground-truth simulated and empirical data well. Responses - including resting membrane potentials - are now well captured.

(2) The comparison with other approaches has been extended to be more quantitative and rigorous.

(3) The authors now convincingly demonstrate that the improved EP-GAN is relatively robust to data ablation.

Weaknesses:

Slow dynamics (e.g., slow ramps) are still not reliably captured. However, as the approach excels at other frontiers - the generation of models for diverse cell types at low computational cost - I consider this to be a relatively minor limitation.

Author response:

The following is the authors’ response to the previous reviews

Reviewer #1 (Public review):

(1) The bad equilibria of the model still remain a concern, as well as other features like the transient overshoots that do not match with the data. I think they could achieve more accuracy here by assigning more weight to such specific features, through adding these as separate objectives for the generator explicitly. The traces contain a five-second current steps, and one second before and one second after the training step. This means that in the RMSE, the current step amplitude will dominate as a feature, as this is simply the state for which the data trace contains most time-points. Note that this is further exacerbated by using the IV curve as an auxiliary objective. I believe a better exploration of specific response features, incorporated as independently weighted loss terms for the generator, could improve the fit. E.g. an auxiliary term could be the equilibrium before and after the current step, another term could penalise response traces that do not converge back to their initial equilibrium, etc.

We thank the reviewer for the suggestion. We supplemented the membrane potential regression loss with errors computed for 3 intervals: pre- post- and mid- stimulation time intervals, improving the accuracy of EP-GAN for baseline membrane potential responses (Figure 2, 3, Table S2, S3). We also changed the simulation protocols for generated parameters by allowing a longer simulation time of 15 seconds, where the stimulation is applied during [5, 10] seconds and no stimulation at t = [0, 5) (pre-stimulation) and t = (10, 15] (post-stimulation). These time intervals are chosen to ensure sufficient stabilization periods before and after stimulation.  

(2) The explanation of what the authors mean with 'inverse gradient operation' is clear now. However, this term is mathematically imprecise, as the inverse gradient does not exist because the gradient operator is not injective. The method is simply forward integration under the assumption that the derivate of the voltage is known at the grid time-points, and should be described as such.

We thank the reviewer for the clarification on inverse gradient operation terminology. In the Methods section, we changed the term describing the inverse gradient operation to ‘forward integration’ which is a more accurate description describing the process.

(3) I appreciate that the authors' method provides parameters of models at a minimal computational cost compared to running an evolutionary optimization for every new recording. I also believe that with some tweaking of the objective, the method could improve in accuracy. However, I share reviewer 2's concerns that the evolutionary baseline methods are not sufficiently explored, as these methods have been used to successfully fit considerably more complex response patterns. One way out of the dilemma is to show that the EP-GAN estimated parameters provide an initial guess that considerably narrows the search space for the evolutionary algorithm. In this context, the authors should also discuss the recent gradient based methods such as Deistler et al. (https://doi.org/10.1101/2024.08.21.608979) or Jones et al (https://doi.org/10.48550/arXiv.2407.04025).

We supplemented the optimization setup for existing methods (GDE3, NSDE, DEMO, and NSGA2) by incorporating steady-state response constraints as the initial selection process. The process is similar to that of EP-GAN training data generation and DEMO parameter selection process [16] (see Results section, page 6 for detail). We also expanded the testing scenarios by evaluating all methods with respect to both small and large HH-model estimation. The small HH-model scenario estimates 47 parameters consisting of channel conductance, reversal potentials and initial conditions with the channel parameters (n = 129) frozen to default values in [41]. Large HH-model includes estimating channel parameters (i.e. 129) in addition to the 47 parameters by considering +-50% variations from their default values. For both small and large HH-model scenarios, we test total sample sizes of both 32k and 64k for all methods to evaluate their scalability with the number of simulated samples given during optimization. The results show that existing methods show good performances for small HH-model scenarios that scale with sample size consistent with literature. EP-GAN on the other hand shows overall better performance in predicting membrane potential responses on both small and large HH-model scenarios.  

Reviewer #2 (Public review):

Major 1: Models do not faithfully capture empirical responses. While the models generated with EPGAN reproduce the average voltage during current injections reasonably well, the dynamics of the response are generally not well captured. For example, for the neuron labeled RIM (Figure 2), the most depolarized voltage traces show an initial 'overshoot' of depolarization, i.e. they depolarize strongly within the first few hundred milliseconds but then fall back to a less depolarized membrane potential. In contrast, the empirical recording shows no such overshoot. Similarly, for the neuron labeled AFD, all empirically recorded traces slowly ramp up over time. In contrast, the simulated traces are mostly flat. Furthermore, all empirical traces return to the pre-stimulus membrane potential, but many of the simulated voltage traces remain significantly depolarized, far outside of the ranges of empirically observed membrane potentials. The authors trained an additional GAN (EPGAN Extended) to improve the fit to the resting membrane potential. Interestingly, for one neuron (AWB), this improved the response during stimulation, which now reproduced the slowly raising membrane potentials observed empirically, however, the neuron still does not reliably return to its resting membrane potential. For the other two neurons, the authors report a decrease in accuracy in comparison to EP-GAN. While such deviations may appear small in the Root mean Square Error (RMSE), they likely indicate a large mismatch between the model and the electrophysiological properties of the biological neuron. The authors added a second metric during the revision - percentages of predicted membrane potential trajectories within empirical range. I appreciate this additional analysis. As the empirical ranges across neurons are far larger than the magnitude of dynamical properties of the response ('slow ramps', etc.), this metric doesn't seem to be well suited to quantify to which degree these dynamical properties are captured by the models.

We made improvements to the training data generation and architecture of EP-GAN to improve its overall accuracy with predicted membrane potential responses. In particular, we divided training data generation into three neuron types found in C. elegans non-spiking neurons: 1) Transient outward rectifier, 2) Outward rectifier and 3) Bistable [8, 16]. Each randomly generated training sample is categorized into one of 3 types by evaluating its steady-state currents with respect to experimental dI/dV bound constraints (See generating training data section under Methods for more detail). The process is then followed by imposing minimum-maximum constraints on simulated membrane potential responses. The setup allows generations of training samples that are of closer distribution to experimentally recorded neurons. This is further described in Section Methods page 15 in the revised manuscript.

We also improved the EP-GAN training process by incorporating random masking of input membrane potential responses. The masking forces EP-GAN to make predictions even with missing voltage traces, improving overall accuracy and allowing EP-GAN to use membrane potential inputs with arbitrary clamping protocol (see Methods page 13 for more detail). For the training loss functions, we further supplemented the membrane potential regression loss with errors computed for 2 intervals: pre- and post-stimulation time intervals to improve EP-GAN prediction capabilities for baseline membrane potentials.

Taken together, these modifications improved EP-GAN’s overall ability to better capture empirical membrane potential responses and we show the results in Figure 2 – 5, Table S2, S3.

Major 2: Comparison with other approaches is potentially misleading. Throughout the manuscript, the authors claim that their approach outperforms the other approaches tested. But compare the responses of the models in the present manuscript (neurons RIM, AFD, AIY) to the ones provided for the same neurons in Naudin et al. 2022 (https://doi.org/10.1371/journal. pone.0268380). Naudin et al. present models that seem to match empirical data far more accurately than any model presented in the current study. Naudin et al. achieved this using DEMO, an algorithm that in the present manuscript is consistently shown to be among the worst of all algorithms tested. I therefore strongly disagree with the authors claim that a "Comparison of EP-GAN with existing estimation methods shows EP-GAN advantage in the accuracy of estimated parameters". This may be true in the context of the benchmark performed in the study (i.e., a condition of very limited compute resources - 18 generations with a population size of 600, compare that to 2000 generations recommended in Naudin et al.), but while EP-GAN wins under these specific conditions (and yes, here the authors convincingly show that their EP-GAN produces by far the best results!), other approaches seem to win with respect to the quality of the models they can ultimately generate.

We thank the reviewer for the feedback regarding the comparison with existing methods. We have revised the optimization setup for existing methods (GDE3, NSDE, DEMO, and NSGA2) by incorporating steady-state response constraints as the initial selection process. The process is similar to that of EP-GAN training data generation and DEMO parameter selection process [16] (see Results section, page 6 for detail). Incorporating this process has improved the accuracy of existing methods especially for small HH-model scenarios where DEMO stood out with the best performance alongside NSGA2 (Figure 5, Table 1, 2).

We also expanded the testing scenarios by evaluating all methods with respect to both small and large HH-model estimation. The small HH-model scenario estimates 47 parameters consisting of channel conductance, reversal potentials and initial conditions with the channel parameters (n = 129) frozen to default values in [41]. Large HH-model includes estimating channel parameters (i.e. 129) in addition to the 47 parameters by considering +-50% variations from their default values. For both small and large HH-model scenarios, we test total sample sizes of both 32k and 64k for all methods to evaluate their scalability with the number of simulated samples given during optimization. The results show that existing methods show good performances for small HH-model scenarios that scale with sample size. EP-GAN on the other hand shows overall better performance in predicting membrane potential responses on both small and large HH-model scenarios. 

In particular, with extended membrane potential error including pre-, mid- , post-activation periods, EP-GAN (trained with 32k samples, large HH-model, 9 neurons) mean membrane potential responses error of 2.82mV was lower than that of DEMO (12.2mV, 64k samples) trained on identical setup (Table 2) and DEMO (7.78mV, using 36,000k samples, 3 neurons) applied to simpler HHmodel in [16]. With respect to DEMO performance in [16], under identical simulation protocol (i.e., no stimulation during (0, 5s), (10, 15s) and stimulation during (5, 10s)), EP-GAN predicted RIM (large HH-model) showed membrane potential accuracy on par with that of DEMO (simpler HH-model) and EP-GAN predicted AFD showed better accuracy for post-activation membrane potential response where DEMO predicted membrane potentials overshoot above the baseline (not shown in the paper).

Major 3: As long as the quality of the models generated by the EP-GAN cannot be significantly improved, I am doubtful that it indeed can contribute to the 'ElectroPhysiome', as it seems likely that dynamics that are currently poorly captured, like slow ramps, or the ability of the neuron to return to its resting membrane potential, will critically affect network computations. If the authors want to motivate their study based on this very ambitious goal, they should illustrate that single neuron model generation with their approach is robust enough to warrant well-constrained network dynamics. Based on the currently presented results, I find the framing of the manuscript far too bold.

We thank the reviewer for the feedback regarding the paper's scope. With revised methods, the overall quality of EP-GAN models is improved with the most significant improvements in baseline membrane potential accuracy. While high quality neuron models could be attained with existing methods given sufficient sample size, our results suggest EP-GAN can predict models with enhanced quality with significantly fewer sample size without a need for retraining, thus complementing the main drawback of evolutionary based methods. While EP-GAN still has limitations (e.g., difficulty in predicting slow ramps) that need to be addressed in the future, we believe its overall performance combined with fast inference speed and flexibility in its input data format (e.g., missing membrane potential traces) is a step forward in the large-scale neuron modeling tasks that can contribute to network models.   

Major 4: The conclusion of the ablation study 'In addition the architecture of EP-GAN permits inference of parameters even when partial membrane potential and steady-state currents profile are given as inputs' does not seem to be justified given the voltage traces shown in Figure 3. For example, for RIM, the resting membrane potential stays around 0 mV, but all empirical traces are around -40mV. For AFD, all simulated traces have a negative slope during the depolarizing stimuli, but a positive slope in all empirically observed traces. For AIY, the shape of hyperpolarized traces is off. While it may be that by their metric neurons in the 25% category are classified as 'preserving baseline accuracy', this doesn't seem justified given the voltage traces presented in the manuscript. It appears the metric is not strict enough.

We improved EP-GAN’s training process by incorporating random masking of input membrane potential responses. The masking forces EP-GAN to make predictions even with missing voltage traces, improving overall accuracy and allowing EP-GAN to use membrane potential inputs with arbitrary clamping protocol.

Such input masking during training has improved the results with ablation studies where EP-GAN now retains baseline membrane potential error (3.3mV, averaged across pre-, mid-, post-activation periods) up to 50% of membrane potential inputs remaining (3.5mV) and up to 25% of steady-state currents remaining (3.5mV).

eLife Assessment

This valuable study investigates the implementation of an efference copy mechanism in the visual flight control system of Drosophila, a topic of broad interest to sensorimotor neuroscientists. Although the behavioral data and computational analyses are each individually solid, there is limited quantitative evaluation of how the model predictions compare to the experimental data.

Reviewer #1 (Public review):

This study provides an integrative model of the visuomotor control in Drosophila melanogaster. This model presents an experimentally derived model based on visually evoked wingbeat pattern recordings of three strategically selected visual stimulus types with well-established behavioral response characteristics. By testing variations of these models, the authors demonstrate that the virtual model behavior can recapitulate the recorded wing beat behavioral results and those recorded by others for these specific stimuli when presented individually. Yet, the novelty of this study and their model is that it allows predictions for natural visual scenes in which multiple visual stimuli occur simultaneously and may have opposite or enhancing effects on behavior. Testing three models that would allow interactions of these visual modalities, the authors show that using a visual efference copy signal allows visual streams to interact, replicating behavior recorded when multiple stimuli are presented simultaneously. Importantly, they validated the prediction of this model in real flies using magnetically tethered flies, e.g., presenting moving bars with varying backgrounds. In conclusion, the presented manuscript presents a commendable effort in developing and demonstrating the validity of a mixture model that enables predictions of Drosophila behavior in natural visual environments.

The manuscript employs a thorough, logical approach, combining computational modeling with experimental behavioral validation using magnetically tethered flies. This iterative integration of simulation and empirical behavioral evidence enhances the credibility of the findings. The quantitative models and validating behavioral experiments make this a valuable contribution to the field. This study is well executed and addresses a significant gap in the modeling of fly behavior and holistic understanding of visuomotor behaviors.

The associated code base is well documented and readily produces all figures in the document.

Reviewer #2 (Public review):

Summary:

The fly visual circuit and its behavioral response to simple visual stimuli have been well investigated, yet how they respond to more complex visual patterns is less understood. Canelo et al. first characterized a fly's steering to simple stimuli and examined how the combination of those stimuli impacts behavior. Combining behavioral experiments and simulation, the authors found that, for some combinations, a behavioral response can be explained by a linear summation of responses to individual stimuli. However, for looming and background motion combinations, the behavioral response to one was suppressed by the other. Furthermore, the effect was dependent on the onset timing of the pair of stimuli.

Strength:

The authors tested various visual stimulus patterns and time delays between combinations of visual stimuli and found novel interactions in behavior. Their findings support the idea that, depending on the visual context, additional mechanisms kick into the visual-motor circuit to coordinate steering behavior flexibly.

Weakness:

The manuscript does not provide conclusive evidence on the presence of an efference copy signal, though there appears to be an intention to associate it with the result. However, demonstrating it is likely to be beyond the main scope of the revised version.

The goal of this manuscript is to understand how the fly's steering behavior is coordinated upon complex visual stimuli, and a number of experiments and simulations support their conclusion.

The behavioral findings presented in this paper will be helpful in further dissecting the underlying neural mechanisms of contextual sensory processing and in understanding visual processing in other species.

Reviewer #3 (Public review):

Summary:

Canelo et al. used a combination of mathematical modeling and behavioral experiments to ask how flies orient to visual features and stabilize their gaze. In particular, the authors propose three models of visuomotor control, which lead to specific experimental predictions. With the goal of teasing out the suggested models, the authors design three flight experiments: 1) a bar-background experiment, 2) a looming-background experiment, and 3) a bar-background statistics experiment. The authors claim that: experiment 1 data favor the addition-only and graded EC model; experiment 2 data favor the all-or-none EC model; experiment 3 appears to suggest a graded EC model.

While the study is interesting, there are major issues with the conceptual framework. In general, there is a major disconnect between model and animal data. The manuscript lacks a statistical framework to support or refute the proposed models. In the end, it is unclear what are the main conclusions of the manuscript and contributions to the field.

Strengths:

They ask a significant question related to efference copies during volitional movement.

The figures are overall clear and salient.

Weaknesses:

Comparison of model to fly data:<br /> In general, the manuscript suffers from a lack of quantitative comparisons between proposed models and fly data, which compromises the main findings of the work. While Figure 1-Fig. supplement 1 shows a direct comparison between experiment and model predictions, puzzlingly there is no such quantitative comparison in the main manuscript for the faster moving stimuli. Please overlay model predictions and experimental data and provide statistical comparisons throughout. The 3 proposed models are hypotheses, but there is no statistical framework to reject or support the models/hypotheses. Further, there is a disconnect between the new flight experiments and models. In fact, we do not see the model predictions for the set of experimental conditions tested in Figs. 5-7.

Concerns about mechanical model: I have several concerns regarding the biomechanics block in Figure 2:

(1) The inertia coefficient, derived from free flight studies. does not take into account the fact that the center of rotation and center of mass do not align in the magnetic tether (see Bender & Dickinson, 2006 for estimates). This must be corrected using the parallel axis theorem. As the authors compare the model prediction to experimental data in a magnetic tether, it is critical that they revise their analysis.

(2) According to their chosen inertia and damping constants, they would estimate that the I/C time constant is ~1E-3 ms, which is much much smaller than what has been estimated for yaw turns in the magnetic tether (200 ms; Bender & Dickinson, 2006) or free flight saccades (~17 ms; see Cheng et al., 2010; 10.1242/jeb.038778). The bottom line is that the current model underestimates the influence of inertia in turn manoeuvres, i.e. the aerodynamic damping is cranked up too high relative to yaw inertia. This may explain the mismatch between data and model that the authors posit, "What causes the fly to undershoot the movement of the target object in the magnetically tethered assay? One hypothesis is that strong upward magnetic force or a blunt top end of the steel pin significantly dampens the flies' flight turns."

Loom response experiment:<br /> As nicely shown by 10.1242/jeb.02369, visual stimulation of looming stimuli in the magnetic tether evokes saccades. Is it the case as well in Fig. 6? Without showing individual trials, it is not possible to know whether this is the case. If indeed saccades are present, then the authors will have to reframe their results given the physiological evidence for saccade-related cancellation signals and the three proposed models.

Minor comments:

Missing Equation 13 for saccade model in Methods.

For the discussion and results related to flight responses to the mismatch between expected and actual visual feedback, which is germane to the proposed models, the authors should integrate a discussion of a recent paper which directly tested this idea through an augmented reality system: 10.1016/j.cub.2023.11.045. In particular, the authors argue that the optomotor response is not particularly flexible because it may not rely on an internal model, as suggested by recent physiological evidence (Fenk et al.). How do these findings relate to the 3 proposed models within your work?

Author response:

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

Reviewer #1 (Public Review):

Summary:

The manuscript "Drosophila Visuomotor Integration: An Integrative Model and Behavioral Evidence of Visual Efference Copy" provides an integrative model of the visuomotor control in Drosophila melanogaster. This model presents an experimentally derived model based on visually evoked wingbeat pattern recordings of three strategically selected visual stimulus types with well-established behavioral response characteristics. By testing variations of these models, the authors demonstrate that the virtual model behavior can recapitulate the recorded wing beat behavioral results and those recorded by others for these specific stimuli when presented individually. Yet, the novelty of this study and their model is that it allows predictions for natural visual scenes in which multiple visual stimuli occur simultaneously and may have opposite or enhancing effects on behavior. Testing three models that would allow interactions of these visual modalities, the authors show that using a visual efference copy signal allows visual streams to interact, replicating behavior recorded when multiple stimuli are presented simultaneously. Importantly, they validated the prediction of this model in real flies using magnetically tethered flies, e.g., presenting moving bars with varying backgrounds. In conclusion, the presented manuscript presents a commendable effort in developing and demonstrating the validity of a mixture model that allows predictions of the behavior of Drosophila in natural visual environments.

Strengths:

Overall, the manuscript is well-structured and clear in its presentation, and the modeling and experimental research are methodically conducted and illustrated in visually appealing and easy-to-understand figures and their captions.

The manuscript employs a thorough, logical approach, combining computational modeling with experimental behavioral validation using magnetically tethered flies. This iterative integration of simulation and empirical behavioral evidence enhances the credibility of the findings.

The associated code base is well documented and readily produces all figures in the document.

Suggestions:

However, while the experiments provide evidence for the use of a visual efference copy, the manuscript would be even more impressive if it presented specific predictions for the neural implementation or even neurophysiological data to support this model. Or, at the very least, a thorough discussion. Nonetheless, these models and validating behavioral experiments make this a valuable contribution to the field; it is well executed and addresses a significant gap in the modeling of fly behavior and holistic understanding of visuomotor behaviors.

We appreciate the reviewer’s thoughtful comments on the strengths and weaknesses of our manuscript. We agree that biophysically realistic model reflecting the structure of neural circuits as well as physiological data from them would be invaluable. However, we are currently unable to provide physiological evidence for EC-based suppression, nor provide circuit architecture for efference copy-based suppression of the stability circuit because the neural pathway underlying this behavior remains unidentified. Extensive recordings from the HS/VS system have revealed cell-type-specific motor-related inputs during both spontaneous and loom-evoked flight turns (Fenk et al., 2021; Kim et al., 2017, 2015). These studies predicted suppression of the optomotor stability response during such turns, and our new experiments confirmed this suppression specifically during loom-evoked turns (Figures 5, 6). However, these neurons are primarily involved in the head optomotor response, not the body optomotor response. We hope to extend our current model in future studies to incorporate more cellular-level detail, as the feedforward circuits underlying stability behavior become more clearly defined.

Here are a few points that should be addressed:

(1) The biomechanics block (Figure 2) should be elaborated on, to explain its relevance to behavior and relation to the underlying neural mechanisms.

We appreciate this suggestion. The mathematical representation of the biomechanics block has been developed by other groups in previous studies (Fry et al., 2003; Ristroph et al., 2010). We used exactly the same model, and its parameters were identical to those used in one of those studies (Fry et al., 2003; Ristroph et al., 2010), in which the parameters were estimated from the stabilizing response in response to magnetic “stumbling” pulses. In the previous version of the manuscript, we had a description of the biomechanics block in the Method section (see Equation 4). In response to the reviewer’s comment, we have made a few changes in Figure 2A and expanded the associated description in the main text, as follows.

(Line 160) “To test the orientation behavior of the model, we developed an expanded model, termed “virtual fly model” hereafter. In this model, we added a biomechanics block that transforms the torque response of the fly to the actual heading change according to kinematic parameters estimated previously (Michael H Dickinson, 2005; Ristroph et al., 2010) (Figure 2A, see Equation 4 in Methods and Movie S1). The virtual fly model, featuring position and velocity blocks that are conditioned on the type of the visual pattern, can now change its body orientation, simulating the visual orientation behavior of flies in the free flight condition.”

(2) It is unclear how the three integrative models with different strategies were chosen or what relevance they have to neural implementation. This should be explained and/or addressed.

Thank you for this valuable comment. We selected the three models based on previous studies investigating visuomotor integration across multiple species, under conditions where multiple sensory cues are presented simultaneously.

The addition-only model represents the simplest hypothesis, analogous to the “additive model” proposed by Tom Collett in his 1980 study (Collett, 1980). We used this model as a baseline to illustrate behavior in the absence of any efference copy mechanism. Notably, some modeling studies have proposed linear (additive) integration for multimodal sensory cues at the behavioral level (Liu et al., 2023; Van der Stoep et al., 2021). However, experimental evidence demonstrating strictly linear integration—either behaviorally or physiologically—remains limited. In our study, new data (Figure 5) show that bar-evoked and background movement-evoked locomotor responses are combined linearly, supporting the addition-only model.

The graded efference copy model has been most clearly demonstrated in the cerebellum-like circuit of Mormyrid fish during electrosensation (Bell, 1981; Kennedy et al., 2014). In this system, the efference copy signal forms a negative image of the predicted reafferent input and undergoes plastic changes as the environment changes—an idea that inspired our modifiable efference copy model (Figure 4–figure supplement 1). The all-or-none efference copy model is exemplified in the sensory systems of smaller organisms, such as the auditory neurons of crickets during stridulation (Poulet and Hedwig, 2006). Notably, in crickets, the motor-related input is referred to as corollary discharge rather than efference copy. Typically, “efference copy” refers to a graded, subtractive motor-related signal, while “corollary discharge” denotes an all-or-none signal, both counteracting the sensory consequences of self-generated actions. In this manuscript, we use the term efference copy more broadly, encompassing both types of motor-related feedback signals (Sommer and Wurtz, 2008).

In response to this comment, we have made the following changes in the main text to enhance its accessibility to general readers.

(Line#268) “This integration problem has been studied across animal sensory systems, typically by analyzing motor-related signals observed in sensory neurons (Bell, 1981; Collett, 1980; Kim et al., 2017; Poulet and Hedwig, 2006). Building on the results of these studies, we developed three integrative models. The first model, termed the “addition-only model”, assumes that the outputs of the object (bar) and the background (grating) response circuits are summed to control the flight orientation (Figure 4B, see Equation 14 in Methods).”

(Line#272) “In the second and third models, an EC is used to set priorities between different visuomotor circuits (Figure 4C,D). In particular, the EC is derived from the object-induced motor command and sent to the object response system to nullify visual input associated with the object-evoked turn (Bell, 1981; Collett, 1980; Poulet and Hedwig, 2006). These motor-related inputs fully suppress sensory processing in some systems (Poulet and Hedwig, 2006), whereas in others they selectively counteract only the undesirable components of the sensory feedback (Bell, 1981; Kennedy et al., 2014).”

(3) There should be a discussion of how the visual efference could be represented in the biological model and an evaluation of the plausibility and alternatives.

Thank you for this helpful comment. We have now added the following discussion to share our perspective on the circuit-level implementation of the visual efference copy in Drosophila.

(Line#481) “Efference copy in Drosophila vision

Under natural conditions, various visual features in the environment may concurrently activate multiple motor programs. Because these may interfere with one another, it is crucial for the central brain to coordinate between the motor signals originating from different sensory circuits. Among such coordination mechanisms, the EC mechanisms were hypothesized to counteract so-called reafferent visual input, those caused specifically by self-movement (Collett, 1980; von Holst and Mittelstaedt, 1950). Recent studies reported such EC-like signals in Drosophila visual neurons during spontaneous as well as loom-evoked flight turns (Fenk et al., 2021; Kim et al., 2017, 2015). One type of EC-like signals were identified in a group of wide-field visual motion-sensing neurons that were shown to control the neck movement for the gaze stability (Kim et al., 2017). The EC-like signals in these cells were bidirectional depending on the direction of flight turns, and their amplitudes were quantitatively tuned to those of the expected visual input across cell types. Although amplitude varies among cell types, it remains inconclusive whether it also varies within a given cell type to match the amplitude of expected visual feedback, thereby implementing the graded EC signal. A more recent study examined EC-like signal amplitude in the same visual neurons for loom-evoked turns, across events (Fenk et al., 2021). Although the result showed a strong correlation between wing response and the EC-like inputs, the authors pointed that this apparent correlation could stem from noisy measurement of all-or-none motor-related inputs.

Thus, these studies did not completely disambiguate between graded vs. all-or-none EC signaling. Another type of EC-like signals observed in the visual circuit tuned to a moving spot exhibited characteristics consistent with all-or-none EC. That is, it entirely suppressed visual signaling, irrespective of the direction of the self-generated turn (Kim et al., 2015; Turner et al., 2022). 

Efference-copy (EC)–like signals have been reported in several Drosophila visual circuits, yet their behavioral role remains unclear. Indirect evidence comes from a behavioral study showing that the dynamics of spontaneously generated flight turns were unaffected by unexpected background motion (Bender and Dickinson, 2006a). Likewise, our behavioral experiments showed that, during loom-evoked turns, responses to background motion are suppressed in an all-or-none manner (Figures 6 and 7). Consistent with this, motor-related inputs recorded in visual neurons exhibit nearly identical dynamics during spontaneous and loom-evoked turns (Fenk et al., 2021). Together, these behavioral and physiological parallels support the idea that a common efference-copy mechanism operates during both spontaneous and loom-evoked flight turns.

Unlike loom-evoked turns, bar-evoked turn dynamics changed in the presence of moving backgrounds (Figure 5), a result compatible with both the addition-only and graded EC models. However, when the static background was updated just before a bar-evoked turn—thereby altering the amplitude of optic flow—the turn dynamics remained unaffected (Figures 5 and 7), clearly contradicting the addition-only model. Thus, the graded EC model is the only one consistent with both findings. If a graded EC mechanism were truly at work, however, an unexpected background change should have modified turn dynamics because of the mismatch between expected and actual visual feedback (Figure 4–figure supplement 1)—yet we detected no such effect at any time scale examined (Figure 7–figure supplement 1). This mismatch would be ignored only if the amplitude of the graded EC adapted to environmental changes almost instantaneously—a mechanism that seems improbable given the limited computational capacity of the Drosophila brain. In electric fish, for example, comparable adjustments take more than 10 minutes (Bell, 1981; Muller et al., 2019). Further investigation is needed to clarify how reorienting flies ignore optic flow generated by static backgrounds, potentially by engaging EC mechanisms not captured by the models tested in this study.

Why would Drosophila rely on the all-or-none EC mechanism instead of the graded one for loom-evoked turns? A graded EC must be adjusted adaptively depending on the environment, as the amplitude of visual feedback varies with both the dynamics of self-generated movement and environmental conditions (e.g., empty vs. cluttered visual backgrounds) (Figure 4—figure supplement 1). Recent studies on electric fish have suggested that a large array of neurons in a multi-layer network is crucial for generating a modifiable efference copy signal matched to the current environment (Muller et al., 2019). Given their small-sized brain, flies might opt for a more economical design for suppressing unwanted visual inputs regardless of the visual environment. Circuits mediating such a type of EC were identified in the cricket auditory system during stridulation (Poulet and Hedwig, 2006), for example. Our study strongly suggests the existence of a similar circuit in the Drosophila visual system. 

We tested the hypothesis that efference-copy (EC) signals guide action selection by suppressing specific visuomotor reflexes when multiple visual features compete. An alternative motif with a similar function is mutual inhibition between motor pathways (Edwards, 1991; Mysore and Kothari, 2020). In Drosophila, descending neurons form dense lateral connections (Braun et al., 2024), offering a substrate for such competitive interactions. Determining whether—and how—EC and mutual inhibition operate will require recordings from the neurons that ensure visual stability, which remain unidentified. Mapping these pathways and assessing how they are modulated by visual and behavioral context are important goals for future work.”

Reviewer #2 (Public Review):

It has been widely proposed that the neural circuit uses a copy of motor command, an efference copy, to cancel out self-generated sensory stimuli so that intended movement is not disturbed by the reafferent sensory inputs. However, how quantitatively such an efference copy suppresses sensory inputs is unknown. Here, Canelo et al. tried to demonstrate that an efference copy operates in an all-or-none manner and that its amplitude is independent of the amplitude of the sensory signal to be suppressed. Understanding the nature of such an efference copy is important because animals generally move during sensory processing, and the movement would devastatingly distort that without a proper correction. The manuscript is concise and written very clearly. However, experiments do not directly demonstrate if the animal indeed uses an efference copy in the presented visual paradigms and if such a signal is indeed non-scaled. As it is, it is not clear if the suppression of behavioral response to the visual background is due to the act of an efference copy (a copy of motor command) or due to an alternative, more global inhibitory mechanism, such as feedforward inhibition at the sensory level or attentional modulation. To directly uncover the nature of an efference copy, physiological experiments are necessary. If that is technically challenging, it requires finding a behavioral signature that unambiguously reports a (copy of) motor command and quantifying the nature of that behavior.

We thank the reviewer for this insightful and constructive comment. We agree that our current behavioral evidence does not directly identify the underlying circuit mechanism, and that direct recordings from visual neurons modulated by an efference copy would be critical for distinguishing between potential mechanisms.

A prerequisite for such physiological investigations would be the identification of both (1) the feedforward neurons directly involved in the optomotor response, and (2) the neurons conveying motor-related signals to the optomotor circuit. Despite efforts by several research groups, the location of the feedforward circuit mediating the optomotor response remains elusive. This limitation has prevented us from obtaining direct cellular evidence of flight turn-associated suppression of optomotor signaling.

In light of the reviewer’s suggestion, we expanded our investigation to strengthen the behavioral evidence for efference copy (EC) mechanisms. In addition to our earlier experiments involving unexpected changes in the static background, we examined how object-evoked flight turns influence the optomotor stability reflex and vice versa (Figures 5 and 6). To quantify the interaction between different visuomotor behaviors, we systematically varied the temporal relationship between two types of visual motion—loom versus moving background, or moving bar versus moving background—and measured the resulting behavioral responses.

Our findings support pattern- and time-specific suppressive mechanisms acting between flight turns associated with the different visual patterns. Specifically:

The responses to a moving bar and a moving background add linearly, even when presented in close temporal proximity.

Loom-evoked turns and the optomotor stability reflex mutually suppress each other in a time-specific manner.

For both loom- and moving bar-evoked flight turns, changes in the static background had no measurable effect on the dynamics of the object-evoked responses.

These results provide a detailed behavioral characterization of a suppressive interaction between distinct visuomotor responses. This, in turn, offers correlative evidence supporting the involvement of an efference copy-like mechanism acting on the visual system. While similar efference copy mechanisms have been documented in other parts of the visual system, we acknowledge that our findings do not exclude alternative explanations. In particular, it is still possible that lateral inhibition within the central brain or ventral nerve cord contributes to the suppression we observed.

Ultimately, definitive proof will require identifying the specific neurons that convey efference copy signals and demonstrating that silencing these neurons abolishes the behavioral suppression. Until such experiments are feasible, our behavioral approach provides an important contribution toward understanding the nature of sensorimotor integration in this system.

Reviewer #3 (Public Review):

Summary:

Canelo et al. used a combination of mathematical modeling and behavioral experiments to ask whether flies use an all-or-none EC model or a graded EC model (in which the turn amplitude is modulated by wide-field optic flow). Particularly, the authors focus on the bar-ground discrimination problem, which has received significant attention in flies over the last 50-60 years. First, they use a model by Poggio and Reichardt to model flight response to moving small-field bars and spots and wide-field gratings. They then simulate this model and compare simulation results to flight responses in a yaw-free tether and find generally good agreement. They then ask how flies may do bar-background discrimination (i.e. complex visual environment) and invoke different EC models and an additive model (balancing torque production due to background and bar movement). Using behavioral experiments and simulation supports the notion that flies use an all-or-none EC since flight turns are not influenced by the background optic flow. While the study is interesting, there are major issues with the conceptual framework.

Strengths:

They ask a significant question related to efference copies during volitional movement.

The methods are well detailed and the data (and statistics) are presented clearly.

The integration of behavioral experiments and mathematical modeling of flight behavior.

The figures are overall very clear and salient.

Weaknesses:

Omission of saccades: While the authors ask a significant question related to the mechanism of bar-ground discrimination, they fail to integrate an essential component of the Drosophila visuomotor responses: saccades. Indeed, the Poggio and Reichardt model, which was developed almost 50 years ago, while appropriate to study body-fixed flight, has a severe limitation: it does not consider saccades. The authors identify this major issue in the Discussion by citing a recent switched, integrate-and-fire model (Mongeau & Frye, 2017). The authors admit that they "approximated" this model as a smooth pursuit movement. However, I disagree that it is an approximation; rather it is an omission of a motor program that is critical for volitional visuomotor behavior. Indeed, saccades are the main strategy by which Drosophila turn in free flight and prior to landing on an object (i.e. akin to a bar), as reported by the Dickinson group (Censi et al., van Breugel & Dickinson [not cited]). Flies appear to solve the bar-ground discrimination problem by switching between smooth movement and saccades (Mongeau & Frye, 2017; Mongeau et al., 2019 [not cited]). Thus, ignoring saccades is a major issue with the current study as it makes their model disconnected from flight behavior, which has been studied in a more natural context since the work of Poggio.

Thank you for this helpful comment. We agree that including saccadic turns is essential and qualitatively improves the model. In the revised manuscript, we therefore expanded our bar-tracking model to incorporate an integrate-and-saccade strategy, now presented in Figure 2—figure supplement

The manuscript now introduces this result as follows:

(Line#190) “Finally, one important locomotion dynamics that a flying Drosophila exhibits while tracking an object is a rapid orientation change, called a “saccade” (Breugel and Dickinson, 2012; Censi et al., 2013; Heisenberg and Wolf, 1979). For example, while tracking a slowly moving bar, flies perform relatively straight flights interspersed with saccadic flight turns (Collett and Land, 1975; Mongeau and Frye, 2017). During this behavior, it has been proposed that visual circuits compute an integrated error of the bar position with respect to the frontal midline and triggers a saccadic turn toward the bar when the integrated value reaches a threshold (Frighetto and Frye, 2023; Mongeau et al., 2019; Mongeau and Frye, 2017). We expanded our bar fixation model to incorporate this behavioral strategy (Figure 2--figure supplement 2). The overall structure of the modified model is akin to the one proposed in a previous study (Mongeau and Frye, 2017), and the amplitude of a saccadic turn was determined by the sum of the position and velocity functions (Figure 2--figure supplement 2A; see Equation 13 in Methods). When simulated, our model successfully reproduced experimental observations of saccade dynamics across different object velocities (Figure 2--figure supplement 2B-D) (Mongeau and Frye, 2017). Together, our models faithfully recapitulated the results of previous behavioral observations in response to singly presented visual patterns (Collett, 1980; Götz, 1987; H. Kim et al., 2023; Maimon et al., 2008; Mongeau and Frye, 2017).”

Apart from Figures 1 and 2, most of our data—whether from simulations or behavioral experiments—use brief visual patterns lasting 200 ms or less. These stimuli trigger a single, rapid orientation change reminiscent of a saccadic flight turn. In this part of the paper, we essentially have examined how multiple visuomotor pathways interact to determine the direction of object-evoked turns when several visual patterns occur simultaneously.

Critically, recent work showed that a group of columnar neurons (T3) appear specialized for saccadic bar tracking through integrate-and-fire computations, supporting the notion of parallel visual circuits for saccades and smooth movement (Frighetto & Frye, 2023 [not cited]).

Thanks for bringing up this critical issue. We have now added this paper in the following part of the manuscript.

(Line#193) “During this behavior, it has been proposed that visual circuits compute an integrated error of the horizontal bar position with respect to the frontal midline and triggers a saccadic turn toward the bar when the integrated value reaches a threshold (Frighetto and Frye, 2023; Mongeau and Frye, 2017).”

(Line#462) “Visual systems extract features from the environment by calculating spatiotemporal relationships of neural activities within an array of photoreceptors. In Drosophila, these calculations occur initially on a local scale in the peripheral layers of the optic lobe (Frighetto and Frye, 2023; Gruntman et al., 2018; Ketkar et al., 2020).”

A major theme of this work is bar fixation, yet recent work showed that in the presence of proprioceptive feedback, flies do not actually center a bar (Rimniceanu & Frye, 2023). Furthermore, the same study found that yaw-free flies do not smoothly track bars but instead generate saccades. Thus prior work is in direct conflict with the work here. This is a major issue that requires more engagement by the authors.

Thank you for your thoughtful comments and for drawing our attention to this important paper. In our experiments, bar fixation on oscillating vertical objects emerges during the “alignment” phase of the magneto-tether protocol. The pattern movement dynamics was similar those used by Rimniceanu & Frye (2023), yet the two studies differ in a key respect: Rimniceanu & Frye employed a motion-defined bar, whereas we presented a dark vertical bar against a uniform or random-dot background. The alignment success rate—defined as the proportion of trials in which the fly’s body angle is within ±25° of the target—was about 50 % (data not shown). Our alignment pattern consisted of three vertical stripes spanning ~40° horizontally; when we replaced it with a single, narrower stripe, the success rate was lowered (data not shown). These observations suggest that bar fixation in the magnetically tethered assay is less robust than in the rigid-tethered assay, although flies still orient toward highly salient vertical objects.

We also observed that bar-evoked turns were elicited more reliably when the bar moved rapidly (45° in 200 ms) in the magneto-tether assay, although the turn magnitude was significantly smaller than the actual bar displacement (Figure 3).

In response to the reviewer’s comment, we now added the following description in the paper regarding the bar fixation behavior, citing Rimniceanu&Frye 2023.

(Line#239) “Another potential explanation arises from recent studies demonstrating that proprioceptive feedback provided during flight turns in a magnetically tethered assay strongly dampens the amplitude of wing and head responses (Cellini and Mongeau, 2022; Rimniceanu et al., 2023).”

Relevance of the EC model: EC-related studies by the authors linked cancellation signals to saccades (Kim et al, 2014 & 2017). Puzzlingly, the authors applied an EC model to smooth movement, when the authors' own work showed that smooth course stabilizing flight turns do not receive cancellation signals (Fenk et al., 2021). Thus, in Fig. 4C, based on the state of the field, the efference copy signal should originate from the torque commands to initiate saccades, and not from torque to generate smooth movement. As this group previously showed, cancellation signals are quantitatively tuned to that of the expected visual input during saccades. Importantly, this tuning would be to the anticipated saccadic turn optic flow. Thus the authors' results supporting an all-or-none model appear in direct conflict with the author's previous work. Further, the addition-only model is not particularly helpful as it has been already refuted by behavioral experiments (Rimneceanu & Frye, Mongeau & Frye).

Thank you for this constructive comment. Efference copy is best established for brief, discrete actions like flight saccades. While motor-related modulation of visual processing has been reported across short- and long-duration behaviours (Chiappe et al., 2010; Fujiwara et al., 2017; Kim et al., 2015, 2017; Maimon et al., 2010; Turner et al., 2022), only flight saccade-associated signals exhibit the temporal profile appropriate to cancel reafferent input. However, von Holst & Mittelstaedt (1950) originally formulated efference copy to explain the smooth optomotor response of hoverflies. In HS/VS recordings in previous studies, however, we could not detect membrane-potential changes tied to baseline wing-beat amplitude (data not shown), but further work is needed. 

Note that visually evoked flight turns analyzed in this paper have relatively fast dynamics. Fenk et al. (2021) showed that HS cells carry EC-like motor signals during both loom-evoked turns and spontaneous saccades. Building on this, we tested whether object-evoked rapid turns modulate other visuomotor pathways. Although Fenk et al. also found that optomotor turns lack motor input to HS cells, the authors did not test whether the optomotor pathway suppresses other reflexes, such as loom-evoked turns. Our new behavioral data (Figure 6) show that optomotor turns indeed suppress loom-evoked turns, suggesting a potential EC signal arising from the optomotor pathway that inhibits loom-responsive visual neurons.

In Kim et al. (2017), the authors argued that HS/VS neurons receive a “quantitatively tuned” efference copy that varies across cell types: yaw-sensitive LPTCs are strongly suppressed, roll-sensitive cells receive intermediate input, and pitch-sensitive cells receive little or none. We also showed that when the amplitude of ongoing visual drive changes, the amplitude of saccade-related potentials (SRPs) scales linearly. This proportionality does not imply a genuinely graded EC, however, because SRP amplitude could vary solely through changes in driving force (Vm – Vrest) with a fixed EC conductance. Crucially, SRPs do not fully suppress feed-forward visual signalling, arguing against an all-or-none EC mechanism.

How, then, can the cellular and behavioural data be reconciled? Silencing HS/VS neurons—or their primary inputs, the T4/T5 neurons—does not markedly diminish the optomotor response in flight (Fenk et al., 2014; Kim et al., 2017), indicating the presence of additional, as-yet-unidentified pathways.

Physiological recordings from other visual neurons that drive the optomotor response in flying Drosophila are therefore needed to determine how strongly they are suppressed during loom-evoked turns.

Behavioral evidence for all-or-none EC model: The authors state "unless the stability reflex is suppressed during the flies' object evoked turns, the turns should slow down more strongly with the dense background than the sparse one". This hypothesis is based on the fact that the optomotor response magnitude is larger with a denser background, as would be predicted by an EMD model (because there are more pixels projected onto the eye). However, based on the authors' previous work, the EC should be tuned to optic flow and thus the turning velocity (or amplitude). Thus the EC need not be directly tied to the background statistics, as they claim. For instance, I think it would be important to distinguish whether a mismatch in reafferent velocity (optic flow) links to distinct turn velocities (and thus position). This would require moving the background at different velocities (co- and anti-directionally) at the onset of bar motion. Overall, there are alternative hypotheses here that need to be discussed and more fully explored (as presented by Bender & Dickinson and in work by the Maimon group).

We appreciate the reviewer’s important suggestion. In response, we performed the recommended experiment. In Figures 5 and 6 of the revised manuscript, we now present how bar- or loom-evoked flight turns affect the response to a moving background pattern. These experiments revealed that bar-evoked turns do not suppress the optic flow response, whereas loom-evoked turns strongly suppress it. Specifically, when background motion began 100 ms after the onset of loom expansion, the response to the background was significantly suppressed. Although weak residual responses to the background motion were observed in this case, this could be due to background motion occurring outside of the suppression interval, which may correspond in duration to the duration of flight turns (Figure 6C,D). 

The lack of suppression of the optic flow response during and after bar-evoked turns appears to suggest that the responses are added linearly (Figure 5), seemingly contradicting the lack of dynamic change when the background dot density was altered (Figure 7, Figure 7–figure supplement 1). That is, the experimental result in Figure 5 supports either an addition-only or a graded efference copy (EC) model. However, the result in Figure 7 supports an all-or-none EC model. If a graded EC were used, the amplitude of the EC should be updated almost instantaneously when the static background changes.

Another possibility is that the optic flow during self-generated turns in a static background is extremely weak compared to the optic flow input generated by physically moving the pattern, perhaps due to the rapid nature of head movements. Indeed, detailed kinematic analysis of head movement during spontaneous saccades in blow flies revealed that the head reaches the target angle before the body completes the orientation change, making the effective speed of reafferent optic flow higher than the speed of body rotation (Hateren and Schilstra, 1999). To test these hypotheses, further experiments will be needed for bar-evoked flight turns.

Publishing the reviewed preprint:

(1) The Reviewed Preprint (including the full text of the preprint we reviewed, the eLife assessment, and public reviews) will typically be published in two weeks' time.

Please let us know if you would like to provide provisional author responses to be posted at the same time (if so, please send these by email). Please do not resubmit within the next two/three weeks, as we will need to publish the first version of the Reviewed Preprint first.

If there are any factual errors in the eLife assessment or public reviews, or other issues we should be aware of, please let us know as soon as possible.

(2) After publication of the Reviewed Preprint, you can use the link below to submit a revised version. There is no deadline to resubmit. Before resubmitting, please ensure that you update the preprint at the preprint server to correspond with the revised version. Upon submitting a revised version, we will ask the editors and reviewers if it's appropriate to update their assessment and public reviews, which will be included alongside the revised Reviewed Preprint. At that time we will also post the recommendations to the authors and the author responses you provide with the revised version. In the author response, please respond to the public reviews (where relevant) and the recommendations to the authors.

(3) Alternatively, you can proceed with the current version of the Reviewed Preprint (once published), without revisions, and request an eLife Version of Record. See the Author Guide for further information: https://elife-rp.msubmit.net/html/elife-rp_author_instructions.html#vor. However, most authors decide to request a Version of Record after a round of revision.

(4) After publication of eLife's Reviewed Preprint, you also have the option to submit/publish in another journal instead: if you choose to do this, please let us know so we can update our records.

The reviewers identified two key revisions that could improve the assessment of the paper:

(1) Consideration of saccades within the model framework (outlined by reviewer 3).

(2) Addition of physiology data to support the conclusions of the paper (outlined by reviewer 2). If this is not feasible within the timescale of revisions, the paper would need to be revised to clarify that the model leads to a hypothesis that would need to be tested with future physiology experiments.

Thank you for these comments.

Regarding revision point #1, we have added Figure 2–figure supplement 2, where we incorporated our position-velocity model (estimated in Figure 1) into the framework of the integrate-and-saccade model. A detailed description of this model is now provided in the main text (Lines 190–203).

For revision point #2, obtaining electrophysiological evidence for efference copy remains challenging, as neither the visual neurons nor the efference-copy neuron has been identified for the wing optomotor response. As suggested by the reviewers, we have revised the title of the paper to reduce emphasis on efference copy and have noted electrophysiological recordings as a direction for future work.

old title: A visual efference copy-based navigation algorithm in Drosophila for complex visual environments

new title: Integrative models of visually guided steering in Drosophila

Specific recommendations are detailed below.

Reviewer #2 (Recommendations For The Authors):

To directly demonstrate if an efference copy is non-scaled, the following experiments can be helpful: record from HS/VS cells and examine the relation between the amplitude of the succade-suppression signal vs. succade amplitude.

Thanks for raising this important point. We previously carried out the suggested analysis for loom-evoked saccades in Fenk et al. (2021). There, significant correlations emerged between wing-response amplitude and saccade-related potentials (Figures 2F and 3C). However, we did not interpret the strong correlation (r ≈ 0.8) as evidence for a graded efference copy, because the amplitude of saccade-related potentials appeared to be bimodal. Upon presentation of the looming stimulus, flies either executed large evasive turns or showed minimal changes in wing-stroke amplitude. Large wing responses were accompanied by strong, saturated suppression of HS-cell membrane potential, whereas trials without wing responses produced only weak modulations—reflected in the bimodal distribution of saccade-related potential amplitudes (Figure 3C). 

Importantly, in rigidly tethered preparations—where these potentials are typically measured—the absence of proprioceptive feedback can itself drive wingbeat amplitudes to saturation during saccades. We therefore reasoned that the lack of intermediate-sized flight saccades would naturally yield correspondingly saturated saccade-related potentials, even if a graded EC system is in play. 

In Kim et al. (2017), we also performed a comprehensive analysis of spontaneous saccade-related potentials across all HS/VS cell types. When we later examined the relationship between saccade amplitude and the corresponding saccade-related potentials in each cell type, we could not find any statistically significant correlation (unpublished data).

measure how much a weak visual stimulus and a strong visual stimulus are suppressed by the suppression signal. If the signal is non-scaled, visual stimuli should always be suppressed independently of their intensities.

Thank you for this important suggestion. As mentioned in our response to the previous comment, we believe it is not feasible to record from neurons responsible for the body optomotor response at this point, as their identity remains unknown. Regarding the HS/VS cells, our previous study showed that HS cells are not always fully suppressed. The changes in saccade-related potential amplitude can be described as a linear function of the pre-saccadic visually-evoked membrane potential (Figure 7 in Kim et al., 2017). 

As suggested by Fenk et al. 2014 (doi: 10.1016/j.cub.2014.10.042), HS cells might also be responsive to a moving bar. If that is the case, and if you present a bar and background (either sparse or dense) in a closed-loop manner to a head-fixed fly, HS cells might be sensitive only to the bar but not to the background (independently of the density).

Thanks for pointing out this important issue. HS cells indeed respond strongly to the horizontal movement of a vertical bar, as expected given that their receptive fields are formed by the integration of local optic flow vectors. In one of our previous studies (Supplemental Figure 1 in Kim et al., 2015), we showed that the response amplitude to a single vertical bar is roughly equivalent to that elicited by a vertical grating composed of 12 bars of the same size. Therefore, we believe that HS cells are likely to contribute to the head response to a moving vertical bar. In a body-fixed flight simulator, HS cells would respond only to the bar if the bar runs in a closed loop with a static background. In this scenario, HS cells are likely to play a role in the head optomotor response.

Note also that the role of HS cells in the wing optomotor response remains unresolved. Unilateral activation of HS cells has been shown to elicit locomotor turns in walking Drosophila (Fujiwara et al., 2017), as well as in flying individuals (unpublished data from our lab). However, a previous study also showed that strong silencing of HS/VS cells significantly reduced the head optomotor response, but not the wing optomotor response (Kim et al., 2017).

If neurophysiology is technically challenging, an alternative way might pay attention to a head movement that exclusively follows the background (Fox et al., 2014 (doi: 10.1242/jeb.080192)). Because HS cells are thought to promote head rotation to background motion, a non-scaled suppression signal on HS cells would always suppress the head rotation independently of the background density.

Thanks for this helpful comment. We have analyzed head movements during bar-evoked flight turns (Figure 7–figure supplement 1B) and found no significant changes across different background dot densities. We think that this might suggest that HS cells are unlikely to receive suppressive inputs during bar-evoked turns, akin to the lack of modulation during optomotor turns.

Another way to separate a potential efference copy from other mechanisms (more global inhibition) is the directionality. A global inhibition would suppress the response to the background even if the background moves in the same direction as self-motion, but the efference copy would not.

Thanks for this important point. In Heisenberg and Wolf, 1979, it was proposed that modulation might be bidirectional, with behavioral effects observed only for perturbations in the “unexpected” direction. In our new data on loom-evoked turns (Figure 6), the suppression appears equally strong for background motion in either direction, supporting an all-or-none suppression mechanism.

Besides, in general, it is unclear if you think an efference copy operates both in smooth pursuits and saccades or if such a signal is only present during saccades. Your previous neurophysiological work supports the latter. Are your behavioral results consistent with the previous saccade suppression idea, or do you propose a new type of efference copy that also operates in smooth pursuits?

Thanks for raising this important point. von Holst and Mittelstaedt (1950) originally introduced the concept of efference copy to explain the smooth optomotor response. We previously analyzed electrophysiological recordings from HS cells for membrane-potential changes associated with slow deviations in wing-steering angle but found none. However, this negative result does not entirely rule out modulation of visual processing during smooth flight turns, given the slow drift in membrane potential observed in most whole-cell recordings.

In this study, We examined only the interactions among visuomotor pathways during these rapid flight turns as the dynamics of visually evoked turns are almost as rapid as spontaneous saccades. Our data reveal that interactions between distinct visuomotor reflexes are more diverse than previously appreciated.

Minor comments:

Line 108, 109: match the description between here and the labels in Fig. 1F.

Thank you for indicating this issue. We have defined the general equation to obtain the position and velocity components in the main text lines 108,109, but due to a slight asymmetry in the data (Fig. 1E) we used the approach indicated in Fig. 1F. and explained in lines 113-117.

Fig.1 F: If the position-dependent component is due to fatigue, the tuning curve's shape is likely changed (shrunk or extended) depending on the stimulus speed. How can you generalize the tuning curve shown here? Does the result hold even if the stimulus speed/contrast/spatial frequency is changed?

We appreciate this indication. We believed that fatigue may be the reason why the wing response to the grating stimulus showed that significant decay (Fig. 1E). As you mention, the stimulus speed would increase the amplitude of the fly’s response up to a saturation point. We addressed this in our model by multiplying the derived value by the angular velocity of the grating.

Regarding the contrast, and spatial frequency we did not test it experimentally, instead, we simulated our model for changing visual feedback (Fig. 4A, B), which can be seen as increasing/decreasing contrast of a grating. An increase in the contrast would increase the response of the fly to the grating and so will contribute to dampening the response to the foreground object (Fig. 4C).

Line 233-255: Here, the description sounds like you will consider several parallel objects (e.g., two stripes) in the visual field instead of the combination of the figure and background (which is referred to in the following paragraph).

Thank you for pointing it out. Indeed it was slightly ambiguous. We have addressed this by explaining the specific situation of a combination of an object and the background in lines 231-233.

Figure 6C: you kept the foreground visual field between sparse and dense random dot backgrounds to keep the bar's saliency. Is it sure that this does not influence the difference in the fly's response to these two backgrounds (in Figure 6B)?

This is a good point that we have also discussed internally. We also carried out similar experiments with a fully covered background and found no significant differences (Figure 7–figure supplement 1).

Reviewer #3 (Recommendations For The Authors):

Identify and analyze flight saccade dynamics in the raw trajectories (e.g., Fig. 3B). There should be some since the bar is near the 'sweet spot' for triggering saccades (see Mongeau & Frye, 2017).

Thank you for bringing up this interesting point. In previous work, it was reported that the fly fixated on a vertical bar through saccadic turns rather than smooth-tracking (Mongeau & Frye, 2017). When the bar width was thin (<15 deg) there was barely one saccade per second (Mongeau & Frye, 2017, Fig. 4). In our magno tether essay (Fig. 3A, B) the object width was 11.25 degrees, and the object moved for a short time window, and so the fly only generated the saccade related to the onset of the object. It could not be considered as a saccade some small turns of a few degrees that are likely related to small perturbations in comparison to those previously reported (Mongeau & Frye, 2017). Additionally, in our protocol (Fig. 3A) from onset time (‘go’ mark), only a single object moved, within an empty background, so in principle there is no trigger for a switch to a smooth movement. We addressed this in lines x-x.

Consider updating the Poggio model with flight saccades (switched, integrate-and-fire).

We appreciate this suggestion. Following previous work (Mongeau et al., 2017), we expanded our model to include a saccade mechanism: the torque produced by the summed position- and velocity-dependent components is now replaced by an integrate-and-fire saccade (Figure 2—figure supplement 2). We optimized the saccade interval and amplitude so that both vary linearly with stimulus amplitude and faithfully reproduce the kinematic properties reported previously (Mongeau et al., 2017).  

Please engage more with the literature, especially work that directly conflicts with your conclusions (see above). Also, highly relevant work by Bender & Dickinson was not sufficiently discussed. Spot results presented in Fig. 3 should be contextualized in light of the work of Mongeau et al., 2019, who performed similar experiments and identified a switch in saccade valence.

We appreciate your pointing out the relevant previous work. We have added references to the following papers and tried to describe the relationship between our data and previous ones.

Bender & Dickinson 2006

(Line#162) “This simulation experiment is reminiscent of the magnetically tethered flight assay, where a flying fly remains fixed at a position but is free to rotate around its yaw axis (Bender and Dickinson, 2006b; Cellini et al., 2022; G. Kim et al., 2023; Mongeau and Frye, 2017).”

(Line#218) “We tested the predictions of our models with flies flying in an environment similar to that used in the simulation (Figure 3A). A fly was tethered to a short steel pin positioned vertically at the center of a vertically oriented magnetic field, allowing it to rotate around its yaw axis with minimal friction (Bender and Dickinson, 2006b; Cellini et al., 2022; G. Kim et al., 2023).”

(Line#238) “To determine if our assay imposes additional friction compared to other assays used in previous studies, we analyzed the dynamics of spontaneous saccades during the “freeze” phase (Figure 3–figure supplement 1A). We found their duration and amplitude to be within the range reported previously (Bender and Dickinson, 2006b; Mongeau and Frye, 2017) (Figure 3–figure supplement 1B-D). 

Mongeau et al., 2019

(Line#196) “During this behavior, it has been proposed that visual circuits compute an integrated error of the bar position with respect to the frontal midline and triggers a saccadic turn toward the bar when the integrated value reaches a threshold (Frighetto and Frye, 2023; Mongeau et al., 2019; Mongeau and Frye, 2017). We expanded our bar fixation model to incorporate this behavioral strategy (Figure 2–figure supplement 2).”

This paper shows that the dynamics of saccadic flight turns elicited by a rotating bar or spot determine whether flies display attraction or aversion. In that study, the visual stimulus—a bar or spot—rotated slowly at a constant 75 deg s⁻¹. By contrast, in our Figure 3 the object moves much faster, driving the neural “integrator” to saturation and triggering an almost immediate flight turn. In Mongeau et al. (2019), saccades occur at variable times and their amplitudes and directions are more stochastic, again reflecting the slower stimulus speed. Because these differences all arise from the disparity in object speed, we did not cite Mongeau et al. (2019) in Figure 3 or the associated text.

In addition to the two papers cited above, we have incorporated several relevant studies on the Drosophila visuomotor control identified through the reviewers’ insightful comments. Examples include:

Frighetto G, Frye MA. 2023 (Line#195, 464)

Rimniceanu et al., 2023 (Line#241)

Cellini & Mongeau 2020 (Line#91)

Cellini & Mongeau 2022 (Line#241)

Cellini et al., 2022 (LIne#91, 162, 218)

Many citations are not in the proper format (e.g. using numbers rather than authors' last name).

Thank you for letting us know. We have changed the remaining citations to the proper format.

eLife Assessment

This valuable study reports evidence that items maintained in working memory can bias attention in an oscillatory manner, with the attentional capture effect fluctuating at theta frequency. The study provides incomplete evidence that this dynamic attentional bias is associated with oscillatory neural mechanisms, particularly in the alpha and theta bands, as measured by EEG. The study will be relevant for researchers studying attention, working memory, and neural oscillations, particularly those interested in how memory and perception interact over time.

Reviewer #1 (Public review):

Summary:

In the presented paper, Lu and colleagues focus on how items held in working memory bias someone's attention. In a series of three experiments, they utilized a similar paradigm in which subjects were asked to maintain two colored squares in memory for a short and variable time. After this delay, they either tested one of the memory items or asked subjects to perform a search task.

In the search task, items could share colors with the memory items, and the authors were interested in how these would capture attention, using reaction time as a proxy. The behavioral data suggest that attention oscillates between the two items. At different maintenance intervals, the authors observed that items in memory captured different amounts of attention (attentional capture effect).

This attentional bias fluctuates over time at approximately the theta frequency range of the EEG spectrum. This part of the study is a replication of Peters and colleagues (2020).

Next, the authors used EEG recordings to better understand the neural mechanisms underlying this process. They present results suggesting that this attentional capture effect is positively correlated with the mean amplitude of alpha power. Furthermore, they show that the weighted phase lag index (wPLI) between the alpha and theta bands across different electrodes also fluctuates at the theta frequency.

Strengths:

The authors focus on an interesting and timely topic: how items in working memory can bias our attention. This line of research could improve our understanding of the neural mechanisms underlying working memory, specifically how we maintain multiple items and how these interact with attentional processes. This approach is intriguing because it can shed light on neuronal mechanisms not only through behavioral measures but also by incorporating brain recordings, which is definitely a strength.

Subjects performed several blocks of experiments, ranging from 4 to 30, over a few days, depending on the experiment. This makes the results - especially those from behavioral experiments 2 and 3, which included the most repetitions - particularly robust.

Weaknesses:

One of the main EEG results is based on the weighted phase lag index (wPLI) between oscillations in the alpha and theta bands. In my opinion, this is problematic, as wPLI measures the locking of oscillations at the same frequency. It quantifies how reliably the phase difference stays the same over time. If these oscillations have different frequencies, the phase difference cannot remain consistent. Even worse, modeling data show that even very small fluctuations in frequency between signals make wPLI artificially small (Cohen, 2015).

Another result from the electrophysiology data shows that the attentional capture effect is positively correlated with the mean amplitude of alpha power. In the presented scatter plot, it seems that this result is driven by one outlier. Unfortunately, Pearson correlation is very sensitive to outliers, and the entire analysis can be driven by an extreme case. I extracted data from the plot and obtained a Pearson correlation of 0.4, similar to what the authors report. However, the Spearman correlation, which is robust against outliers, was only 0.13 (p = 0.57), indicating a non-significant relationship.

The behavioral data are interesting, but in my opinion, they closely replicate Peters and colleagues (2020) using a different paradigm. In that study, participants memorized four spatial positions that formed the endpoints of two objects, and one object was cued. Similarly, reaction times fluctuated at theta frequency, and there was an anti-phase relationship between the two objects. The main novelty of the present study is that this bias can be transferred to an unrelated task. While the current study extends Peters and colleagues' findings to a different task context, the lack of a thorough, direct comparison with Peters et al. limits the clarity of the novel insights provided.

Cohen, M. X. (2015). Effects of time lag and frequency matching on phase-based connectivity. Journal of Neuroscience Methods, 250, 137-146.

Peters, B., Kaiser, J., Rahm, B., & Bledowski, C. (2020). Object-based attention prioritizes working memory contents at a theta rhythm. Journal of Experimental Psychology: General, 150(6), 1250-1256.

Reviewer #2 (Public review):

The information provided in the current version of the manuscript is not sufficient to assess the scientific significance of the study.

(1) In many cases, the details of the experiments or behavioral tasks described in the main text are not consistent with those provided in the Materials and Methods section. Below, I list only a few of these discrepancies as examples:

a) For Experiment 1, the Methods section states that the detection stimulus was presented for 2000 ms (lines 494 and 498), but Figure 1 in the main text indicates a duration of 1500 ms.

b) For Experiment 2, not only is the range of SOAs mentioned in the Methods section inconsistent with that shown in the main text and the corresponding figure, but the task design also differs between sections.

c) For Experiment 3, the main text indicates that EEG recordings were conducted, but in the Methods section, the EEG recording appears to have been part of Experiment 2 (lines 538-540).

(2) The results described in the text often do not match what is shown in the corresponding figure. For example:

a) In lines 171-178, the SOAs at which a significant difference was found between the two conditions do not appear to match those shown in Figure 2A.

b) In Figure 4, the figure legend (lines 225-228) does not correspond to the content shown in the figure.

c) In Figure 9, not sufficient information is provided within the figure or in the text, making it difficult to understand. Consequently, the results described in the text cannot be clearly linked to the figure.

(3) Insufficient information is provided regarding the data analysis procedures, particularly the permutation tests used for the data presented in Figures 2B, 4, and 10. The results shown in these figures are critical for the main conclusions drawn in the manuscript.

Given these issues, it is not possible to provide a detailed review of the study, particularly regarding its scientific significance.

eLife Assessment

This study presents valuable computational findings on the neural basis of learning new motor memories and the savings using recurrent neural networks. The evidence supporting the claims of the authors is solid, but it would benefit from more controls and from considering the role of explicit strategies and other brain regions. This work will be of interest to computational and experimental neuroscientists working in motor learning.

Reviewer #1 (Public review):

Summary:

Shahbazi et al used a recurrent neural network model trained to control a musculoskeletal model of the arm to investigate how neural populations accommodate activity patterns underpinning savings. The paper draws upon the recent finding of a "uniform shift" in preparatory activity in monkey motor cortex associated with savings, and leverages full access to a computational model to establish causality.

Strengths:

The paper is well written, and the figures are clearly presented. The key finding that the uniform shift first reported based on neural recordings by Sun et al. emerges in artificial neural networks performing a similar task is interesting and well-backed by their analyses. Manipulating this uniform shift to show that it drives behavioural savings is an important causal confirmation of the proposal by Sun et al.

Weaknesses / Comments:

As mentioned earlier, the core results are well backed by the analyses. Most of my comments relate to adding more controls and additional questions that could be explored with the model to strengthen the paper.

(1) Savings are quantified as more rapid relearning of the FF upon re-exposure (e.g., Figure 3). This finding is based on backpropagation through time, but would this hold when using a different optimiser, e.g., FORCE?

(2) The authors should include a "null model" showing that training on a different reaching task following NF, as opposed to FF2, won't show something akin to a uniform shift during preparation due to the adoption of TDR and having similar targets.

(3) The analyses of network activity during movement preparation (Figure 4) nicely replicate the key finding in Sun et al, but I think the authors could leverage the full access to their network and go further, e.g., by examining changes (or the lack of) during execution in FF2 with respect to FF (and perhaps in a future NF2 with respect to NF), including whether execution activity lives also lives in parallel hyperplanes, etc.

(4) Related to the above, while the results are interesting and the paper is well done, I kept wishing that the authors had done "more" with their model. This could be one or two final sections on "predictions" that would nicely complement their "validation" of the uniform shift, and that, in my opinion, would greatly increase the impact of the paper. In particular:<br /> a) What would be the effect of learning more "tasks"? For example, is there a limit on how many fields can be learned? (You show something related by manipulating network size, but this is slightly different.)<br /> b) Figure 5 is a nice causal demonstration that the uniform shift is related to savings. However, and related to comment #3, it'd be interesting to see more details about how the behaviour and the network activity changes as preparatory activity shifts along this axis, in particular regarding how moving the preparatory states affect the organisation and dynamics of upcoming execution activity -these are the kind of intuitions that modelling studies like this one can provide.<br /> c) The authors focus on a task design that spans baseline, FF, NF, FF2 to replicate the original study by Sun et al. However, it would be interesting if they generated predictions for neural changes to other types of tasks that have been studied behaviourally. These could include, for example: (i) modelling a visuomotor rotation or a mirror reversal task; (ii) having to adapt to a FF in the opposite direction; (iii) investigating the role of adding an explicit context and having the networks learn multiple FF; and (iv) trying to learn FF fields in opposite directions, perhaps restricted to specific targets. As the authors know, all these questions and more have been studied with similar behavioural paradigms, and it would be nice to see what neural predictions are generated by this model.

(5) On the Discussion: When extrapolating from neural network results to animals, the fact that your networks can learn implicitly doesn't mean that animals do learn implicitly. Indeed, I think the consensus view is that different perturbations may lead to the expression of different types of savings (e.g., FF vs VR, which seems to be more explicit). Besides, these different mechanisms may be primarily implemented by brain regions less directly tied to motor control (e.g., cerebellum, parietal cortex?), which are not directly implemented in the authors' model.

These aspects (limitations) should be discussed in the paper.

Reviewer #2 (Public review):

Summary:

Shahbazi et al. trained recurrent neural networks (RNNs) to simulate human upper limb movement during adaptation to a force field perturbation. They demonstrated that throughout adaptation, the pattern of motor commands to the muscles of the simulated arm changed, allowing the perturbed movements to regain their typical, perturbation-free straight-line paths. After this initial learning block (FF1), the network encountered null-fields to wash out the adaptation, before re-experiencing the force in a second learning block (FF2). Upon re-exposure, the network learned faster than during initial learning, consistent with the savings observed in behavioral studies of adaptation. They also found that as the number of hidden units in the RNN increased, so did the probability of exhibiting savings. The authors concluded that these results propose a neural basis for savings that is independent of context and strategic processes.

Strengths:

The paper addresses an important and controversial topic in motor adaptation: the mechanism underlying motor memory. The RNN simulation reproduces behavioral hallmarks of adaptation, and it provides a useful illustration of the pattern of muscle activity underlying human-like movements under both normal and perturbing conditions. While the savings effect produced by the network, though significant, appears somewhat small, the simulation demonstrating an increase in savings with a greater number of hidden units is particularly intriguing.

Weaknesses:

(1) To be transparent, savings in motor adaptation have been a primary focus of my own research. Some core findings presented in this paper are at odds with the ideas I and others have previously put forward. While I don't want to impose my agenda on the authors of this paper, I do think the authors should address these issues.

a) The authors acknowledge the ongoing debate in the literature regarding the mechanisms underlying savings, particularly whether it stems from explicit or implicit learning processes. However, it remains unclear how the current work addresses this debate. There is already a considerable body of research, particularly in visuomotor adaptation, demonstrating that savings is predominantly driven by explicit strategies. For example, when people are asked to report their strategy, they recall a strategy that was useful during the first learning block (Morehead et al. 2015). Furthermore, savings are abolished under experimental manipulations designed to eliminate strategic contributions (e.g., Haith et al., 2015; Huberdeau et al., 2019; Avraham et al., 2021). The authors briefly state that their findings support the hypothesis that a neural basis of memory retention underlying savings can be independent of cognitive or strategic learning components, and that savings can be characterized as implicit. While these statements may be true, it is not clear how this work substantiates these claims.<br /> b) Our research has also demonstrated that if implicit adaptation is completely washed out after the initial learning block, it not only fails to exhibit savings but is actually attenuated relative to the first learning block (Avraham et al., 2021). This phenomenon of attenuation upon relearning can also be seen in other studies of visuomotor adaptation (e.g., Leow et al., 2020; Yin and Wei, 2020; Hamel et al., 2021; Hamel et al., 2022; Wang and Ivry, 2023; Hadjiosif et al., 2023). More recently, we have shown that this attenuation is due to anterograde interference arising from the experience with the washout block experience (Avraham and Ivry, 2025). We illustrated that the implicit system is highly susceptible to interference; it doesn't require exposure to salient opposite errors and can occur even following prolonged exposure to veridical feedback. The central thesis of this paper, namely that implicit savings can emerge through RNNs, is at odds with these empirical results. The authors should address this discrepancy.

(2) This brings me to the question about neural correlates: The results are linked to activity in the primary motor cortex. How does that align with the well-established role of the cerebellum in implicit motor adaptation? And with the studies showing that savings are due to explicit strategies, which are generally associated with prefrontal regions?

(3) The analysis on the complexity of the neural network (i.e., the number of hidden units) and its relationship to savings is very interesting. It makes sense to me that more complex networks would show more savings. I'm not sure I follow the author's explanation, but my understanding is that increased network complexity makes it more difficult to override the formed memory through interference (e.g., from the experience with NF2). Also, the results indicate that a network with 32 units led to a less-than-chance level of networks exhibiting savings (Figure 3b). What behavioral output does this configuration produce? Could this behavior manifest as attenuation upon relearning? Furthermore, if one were to examine an even smaller, simpler network (perhaps one more closely reflecting cerebellar circuits), would such a model predict attenuation rather than savings?

(4) The authors emphasize that their network did not receive any explicit contextual signals related to the presence or absence of the force field (FF), thus operating in a 'context-free' manner. From my understanding, some existing models of context's role in motor memories (e.g., Oh and Schweighofer, 2019; Heald et al., 2021) propose that memory-related changes can be observed even without explicit contextual information, as contextual changes can be inferred from sudden or significant environmental shifts (e.g., the introduction or removal of perturbations). Given this, could the observed savings in the current simulation be explained by some form of contextual retrieval, inferred by the network from the re-presentation of the perturbation in FF2?

(5) If there is residual hidden unit activity related to the FF at the end of the NF2 phase, how does the simulated movement revert back to baseline? Are there any differences in the movement trajectory, beyond just lateral deviation, between NF1 and NF2? The authors state that "changes in the preparatory hidden unit activity did not result in substantive changes in the motor commands (Figure 5b), which emphasizes that the uniform shift resides in the null space of motor output." However, Figure 5b appears to show visible changes in hidden unit activity. Don't these changes reflect a pattern of muscle activity that is the basis for behavior? These changes are indeed small, but it seems that so is the effect size for savings (Figure 3a). Could this suggest that there is not, in fact, a complete washout of initial learning during NF2 within the network?

eLife Assessment

This useful study replicates a previous finding that information about peripherally presented visual stimuli is represented in the foveal visual cortex, and extends it by demonstrating that these representations are similar to those evoked by foveally presented stimuli. The authors' gaze-contingent fMRI design provides solid evidence for these findings. Some of the stronger theoretical claims, such as that the effects are due to predictive pre-saccadic remapping, are not fully supported by the current results.

Reviewer #1 (Public review):

Summary:

The main contributions of this paper are: (1) a replication of the surprising prior finding that information about peripherally-presented stimuli can be decoded from foveal V1 (Williams et al 2008), (2) a new demonstration of cross-decoding between stimuli presented in the periphery and stimuli presented at the fovea, (3) a demonstration that the information present in the fovea is based on shape not semantic category, and (4) a demonstration that the strength of foveal information about peripheral targets is correlated with the univariate response in the same block in IPS.

Strengths:

The design and methods appear sound, and finding (2) above is new, and importantly constrains our understanding of this surprising phenomenon. The basic effect investigated here is so surprising that even though it has been replicated several times since it was first reported in 2008, it is useful to replicate it again.

Weaknesses:

(1) The paper, including in the title ("Feedback of peripheral saccade targets to early foveal cortex") seems to assume that the feedback to foveal cortex occurs in conjunction with saccade preparation. However, participants in the original Williams et al (2008) paper never made saccades to the peripheral stimuli. So, saccade preparation is not necessary for this effect to occur. Some acknowledgement and discussion of this prior evidence against the interpretation of the effect as due to saccade preparation would be useful. (e.g., one might argue that saccade preparation is automatic when attending to peripheral stimuli.)

(2) The most important new finding from this paper is the cross-decodability between stimuli presented in the fovea and stimuli presented in the periphery. This finding should be related to the prior behavioral finding (Yu & Shim, 2016) that when a foveal foil stimulus identical to a peripheral target is presented 150 ms after the onset of the peripheral target, visual discrimination of the peripheral target is improved, and this congruency effect occurred even though participants did not consciously perceive the foveal stimulus (Yu, Q., & Shim, W. M., 2016). Modulating foveal representation can influence visual discrimination in the periphery (Journal of Vision, 16(3), 15-15).

(3) The prior literature should be laid out more clearly. For example, most readers will not realize that the basic effect of decodability of peripherally-presented stimuli in the fovea was first reported in 2008, and that that original paper already showed that the effect cannot arise from spillover effects from peripheral retinotopic cortex because it was not present in a retinotopic location between the cortical locus corresponding to the peripheral target and the fovea. (For example, this claim on lines 56-57 is not correct: "it remains unknown 1) whether information is fed back all the way to early visual areas".) What is needed is a clear presentation of the prior findings in one place in the introduction to the paper, followed by an articulation and motivation of the new questions addressed in this paper. If I were writing the paper, I would focus on the cross-decodability between foveal and peripheral stimuli, as I think that is the most revealing finding.

Reviewer #2 (Public review):

Summary:

This study investigated whether the identity of a peripheral saccade target object is predictively fed back to the foveal retinotopic cortex during saccade preparation, a critical prediction of the foveal prediction hypothesis proposed by Kroell & Rolfs (2022). To achieve this, the authors leveraged a gaze-contingent fMRI paradigm, where the peripheral saccade target was removed before the eyes landed near it, and used multivariate decoding analysis to quantify identity information in the foveal cortex. The results showed that the identity of the saccade target object can be decoded based on foveal cortex activity, despite the fovea never directly viewing the object, and that the foveal feedback representation was similar to passive viewing and not explained by spillover effects. Additionally, exploratory analysis suggested IPS as a candidate region mediating such foveal decodability. Overall, these findings provide neural evidence for the foveal cortex processing the features of the saccade target object, potentially supporting the maintenance of perceptual stability across saccadic eye movements.

Strengths:

This study is well-motivated by previous theoretical findings (Kroell & Rolfs, 2022), aiming to provide neural evidence for a potential neural mechanism of trans-saccadic perceptual stability. The question is important, and the gaze-contingent fMRI paradigm is a solid methodological choice for the research goal. The use of stimuli allowing orthogonal decoding of stimulus category vs stimulus shape is a nice strength, and the resulting distinctions in decoded information by brain region are clean. The results will be of interest to readers in the field, and they fill in some untested questions regarding pre-saccadic remapping and foveal feedback.

Weaknesses:

The conclusions feel a bit over-reaching; some strong theoretical claims are not fully supported, and the framing of prior literature is currently too narrow. A critical weakness lies in the inability to test a distinction between these findings (claiming to demonstrate that "feedback during saccade preparation must underlie this effect") and foveal feedback previously found during passive fixation (Williams et al., 2008). Discussions (and perhaps control analysis/experiments) about how these findings are specific to the saccade target and the temporal constraints on these effects are lacking. The relationship between the concepts of foveal prediction, foveal feedback, and predictive remapping needs more thorough treatment. The choice to use only 4 stimuli is justified in the manuscript, but remains an important limitation. The IPS results are intriguing but could be strengthened by additional control analysis. Finally, the manuscript claims the study was pre-registered ("detailing the hypotheses, methodology, and planned analyses prior to data collection"), but on the OSF link provided, there is just a brief summary paragraph, and the website says "there have been no completed registrations of this project".

Specifics:

(1) In the eccentricity-dependent decoding results (Figure 2B), are there any statistical tests to support the results being a U-shaped curve? The dip isn't especially pronounced. Is 4 degrees lower than the further ones? Are there alternative methods of quantifying this (e.g., fitting it to a linear and quadratic function)?

(2) In the parametric modulation analysis, the evidence for IPS being the only region showing stronger fovea vs peripheral beta values was weak, especially given the exploratory nature of this analysis. The raw beta value can reflect other things, such as global brain fluctuations or signal-to-noise ratio. I would also want to see the results of the same analysis performed on the control condition decoding results.

(3) Many of the claims feel overstated. There is an emphasis throughout the manuscript (including claims in the abstract) that these findings demonstrate foveal prediction, specifically that "image-specific feedback during saccade preparation must underlie this effect." To my understanding, one of the key aspects of the foveal prediction phenomenon that ties it closely to trans-saccadic stability is its specificity to the saccade target but not to other objects in the environment. However, it is not clear to what degree the observed findings are specific to saccade preparation and the peripheral saccade target. Should the observers be asked to make a saccade to another fixation location, or simply maintain passive fixation, will foveal retinotopic cortex similarly contain the object's identity information? Without these control conditions, the results are consistent with foveal prediction, but do not definitively demonstrate that as the cause, so claims need to be toned down.

(4) Another critical aspect is the temporal locus of the feedback signal. In the paradigm, the authors ensured that the saccade target object was never foveated via the gaze-contingent procedure and a conservative data exclusion criterion, thus enabling the test of feedback signals to foveal retinotopic cortex. However, due to the temporal sluggishness of fMRI BOLD signals, it is unclear when the feedback signal arrives at the foveal retinotopic cortex. In other words, it is possible that the feedback signal arrives after the eyes land at the saccade target location. This possibility is also bolstered by Chambers et al. (2013)'s TMS study, where they found that TMS to the foveal cortex at 350-400 ms SOA interrupts the peripheral discrimination task. The authors should qualify their claims of the results occurring "during saccade preparation" (e.g., pg 1 ln 22) throughout the manuscript, and discuss the importance of temporal dynamics of the effect in supporting stability across saccades.

(5) Relatedly, the claims that result in this paradigm reflect "activity exclusively related to predictive feedback" and "must originate from predictive rather than direct visual processes" (e.g., lines 60-65 and throughout) need to be toned down. The experimental design nicely rules out direct visual foveal stimulation, but predictive feedback is not the only alternative to that. The activation could also reflect mental imagery, visual working memory, attention, etc. Importantly, the experiment uses a block design, where the same exact image is presented multiple times over the block, and the activation is taken for the block as a whole. Thus, while at no point was the image presented at the fovea, there could still be more going on than temporally-specific and saccade-specific predictive feedback.

(6) The authors should avoid using the terms foveal feedback and foveal prediction interchangeably. To me, foveal feedback refers to the findings of Williams et al. (2008), where participants maintained passive fixation and discriminated objects in the periphery (see also Fan et al., 2016), whereas foveal prediction refers to the neural mechanism hypothesized by Kroell & Rolfs (2022), occurring before a saccade to the target object and contains task irrelevant feature information.

(7) More broadly, the treatment of how foveal prediction relates to saccadic remapping is overly simplistic. The authors seem to be taking the perspective that remapping is an attentional phenomenon marked by remapping of only attentional/spatial pointers, but this is not the classic or widely accepted definition of remapping. Within the field of saccadic remapping, it is an ongoing debate whether (/how/where/when) information about stimulus content is remapped alongside spatial location (and also whether the attentional pointer concept is even neurophysiologically viable). This relationship between saccadic remapping and foveal prediction needs clarification and deeper treatment, in both the introduction and discussion.

(8) As part of this enhanced discussion, the findings should be better integrated with prior studies. E.g., there is some evidence for predictive remapping inducing integration of non-spatial features (some by the authors themselves; Harrison et al., 2013; Szinte et al., 2015). How do these findings relate to the observed results? Can the results simply be a special case of non-spatial feature integration between the currently attended and remapped location (fovea)? How are the results different from neurophysiological evidence for facilitation of the saccade target object's feature across the visual field (Burrow et al., 2014)? How might the results be reconciled with a prior fMRI study that failed to find decoding of stimulus content in remapped responses (Lescroart et al, 2016)? Might this reflect a difference between peripheral-to-peripheral vs peripheral-to-foveal remapping? A recent study by Chiu & Golomb (2025) provided supporting evidence for peripheral-to-fovea remapping (but not peripheral-to-peripheral remapping) of object-location binding (though in the post-saccadic time window), and suggested foveal prediction as the underlying mechanism.

Reviewer #3 (Public review):

Summary:

In this paper, the authors used fMRI to determine whether peripherally viewed objects could be decoded from the foveal cortex, even when the objects themselves were never viewed foveally. Specifically, they investigated whether pre-saccadic target attributes (shape, semantic category) could be decoded from the foveal cortex. They found that object shape, but not semantic category, could be decoded, providing evidence that foveal feedback relies on low-mid-level information. The authors claim that this provides evidence for a mechanism underlying visual stability and object recognition across saccades.

Strengths:

I think this is another nice demonstration that peripheral information can be decoded from / is processed in the foveal cortex - the methods seem appropriate, and the experiments and analyses are carefully conducted, and the main results seem convincing. The paper itself was very clear and well-written.

Weaknesses:

There are a couple of reasons why I think the main theoretical conclusions drawn from the study might not be supported, and why a more thorough investigation might be needed to draw these conclusions.

(1) The authors used a blocked design, with each object being shown repeatedly in the same block. This meant that the stimulus was entirely predictable on each block, which weakens the authors' claims about this being a predictive mechanism that facilitates object recognition - if the stimulus is 100% predictable, there is no aspect of recognition or discrimination actually being tested. I think to strengthen these claims, an experiment would need to have unpredictable stimuli, and potentially combine behavioural reports with decoding to see whether this mechanism can be linked to facilitating object recognition across saccades.

(2) Given that foveal feedback has been found in previous studies that don't incorporate saccades, how is this a mechanism that might specifically contribute to stability across saccades, rather than just being a general mechanism that aids the processing/discrimination of peripherally-viewed stimuli? I don't think this paper addresses this point, which would seem to be crucial to differentiate the results from those of previous studies.

eLife Assessment

This important study uses a combination of eye-tracking and computational models based on Active Inference to explain behavior in a gaze-contingent cued-reversal paradigm with 6 - 10-month-old infants. The study demonstrates solid evidence that the same rigorous computational modeling standards commonly applied in studies in adults can also be applied in studies of infants' learning, and a cluster analysis reveals that the parameters of the winning model provide better pattern separation between identified subgroups than behavior or questionnaire data alone. However, the evidence for some specific claims is incomplete, due to poor behavioral performance, unclear significance of the pupil data, and complexity of the model fitting; the claims regarding implications for psychiatry were also considered to be too strong and unsupported by evidence. This work will be of interest to developmental psychologists and cognitive neuroscientists.

Reviewer #1 (Public review):

Summary:

The authors developed a new gaze-based reversal task to study 6 - 10-month-old infants, in what would typically be a very challenging age group to study behavior related to learning, exploration, and perseveration. Here, the research question is excellently motivated by pointing out the limitation of past work that has typically studied adult clinical populations using similar approaches, which presents only the endpoint of the developmental process. Thus, there is important clinical and scientific value in studying much earlier stages in the developmental process. Here, the authors accomplish this with a new gaze-based paradigm that allows them to fit a variety of complex computational models to data from 41 infants. The main advantage of their winning model is that the parameters provide better pattern separation between two identified clusters of participants compared to behavioral variables alone.

Strengths:

Overall, the paper is well-written, and the models and analyses are applied in a principled and thorough fashion. The authors do an excellent job of both motivating their research question and addressing it through their task and set of computational models. The scope is also quite ambitious, modeling both choices and pupillary responses, while also using the models to generate behavior that is comparable to the experimental data and performing a cluster analysis to compare the suitability of the model parameters vs. other behavioral/questionnaire data in performing pattern separation between participants.

Weaknesses:

However, despite these strengths, I had a number of concerns that may limit the reliability of the findings.

First, given the fact that the rewards for the initial pre-reversal setting are defined by the first choice of the infants, it was unclear to me whether the behavioral patterns in Figure 2 really support the fact that there was in fact, (prediction-error-based) learning in the task at all. The behavioral analyses proceed very briskly without really addressing this question, before rapidly jumping off the complexity cliff to present the models. However, even with the models, the winning model only had free parameters for preference (c) and a left-right dominance (epsilon), which don't really capture mechanisms related to learning. The epistemic and extrinsic components included in the model at the 2nd stage could potentially help shed light on this question, but (unless I've misunderstood) they seem to be all-or-nothing parts of the model, and thus don't reappear in later analyses (e.g., cluster analysis) because they are not individual-specific parameters. Thus, the main learning-relevant aspects of the model seem divorced from the ability to perform clustering or other clinically relevant diagnoses downstream. Thus, it was unclear to me whether the results really capture mechanisms related to cognitive flexibility that motivate the manuscript in the introduction.

My other main concern was the complexity of the models and the way model comparison was performed using the three stages. First of all, the set of models is quite complex and risks alienating many developmental psychologists who would otherwise be very interested in these findings. Thus, I'm curious why the authors didn't consider including much simpler context-based RL models (e.g., Rescorla-Wagner/Q-learning models) that explicitly use prediction-error updates and whose simplicity might better match the simplicity of the behavior that 6-10 month infants are capable of displaying. Certainly, preference (as an inverse temperature parameter for a softmax policy) and left-right dominance (as a bias) could be implemented with these much simpler models. Second, while the three-stage model comparison seems somewhat principled, it left me questioning whether the 1st stage or 2nd stage results might be impacted by later stages. For instance, if the Simple-discard model were to still win in the first stage, once omega and eta have been eliminated as free parameters. Of course, I understand that there may be feasibility issues with testing all combinatorial variants of the model. But it was unclear why this specific order was chosen and what consequences this sequential dependency in the model fitting may have for the conclusions. And while model identifiability is stated in the abstract as one of the strengths of this approach, there don't seem to be any clear analyses supporting this fact. I would have loved to see a model recovery analysis (see Wilson & Collins et al., eLife 2019) to support this statement.

Reviewer #2 (Public review):

Summary:

This paper examines infants' learning in a novel gaze-contingent cued reversal learning task. The study provides strong evidence that infants learn in the task, and they characterize individual differences in learning using computational modeling. The best-fitting model of the set compared reflects a learning of mappings between context cues and outcomes that do not carry over across blocks. Infants are then clustered into two groups based on model parameter estimates capturing primacy bias and reward sensitivity. These groupings exhibited differences in infant temperament and other developmental measures. The modeling is rigorous, with model predictions accounting for substantial variance in infants' choices, and parameter estimates showing high recoverability. This study is important in that it demonstrates that such rigorous standards in computational modeling of behavior can be successfully deployed in infant studies.

Strengths:

The study provides evidence that infants exhibit cognitive flexibility within a reversal learning task and do not simply perseverate.

The methods used within the novel gaze-contingent will be useful for other groups interested in studying learning and decision-making in infants.

The study applies rigorous computational modeling approaches to infants' choices (inferred from gaze) and their physiological responses (i.e., pupil dilation) in the task, demonstrating that infants' reward learning is well-captured by an error-driven learning process.

The authors conduct model comparison, posterior predictive checks, and parameter recoverability analyses and demonstrate that model parameters can be well estimated and that the model can recapitulate infant choice behavior.

Physiological pupil dilation measures that correlate with prediction error signals from the model further validate the model as capturing the learning process.

Weaknesses:

It is not entirely clear that the individual differences in reversal learning identified between the two clusters of infants (ostensibly reflecting differences in cognitive flexibility) have construct validity or specificity for the associated developmental abilities that differ between groups (daily living, communication, motor function, and socialization).

Similarly, it's not clear why the paper is framed as an advance for infant computational *psychiatry* rather than simply an advance in computational modeling of infant behavior. It seems to me that a more general framing is warranted. Basic cognitive development research can also benefit from cognitive hypothesis testing via computational model comparison and precise measurement of infants' behavior in reward learning tasks. Is there reason to believe that infants' behavior in this task might have construct validity for mental health problems related to cognitive flexibility later in development? Do the Vineland or IBQ-R-VSF prospectively predict clinical symptoms?

A large proportion of the recruited infants (14 of 55) were excluded, but few details are provided on why and when they were excluded. Did the excluded infants differ on any of the non-task measures? This information would be helpful to understand limitations in the utility of the task or the generalizability of the findings.

It is stated that: "The infants who completed at least three trials following the reversal were included in the analysis, as it is more likely that their expectations were violated in this interval." Are three trials post-reversal sufficient to obtain reliable estimates of model parameters? More details should be provided on the number of trials completed for all of the included/excluded infants.

Reviewer #3 (Public review):

This paper used computational modeling of infants' performance in a reversal learning paradigm to identify two subgroups of infants, one that initially learned a bit faster but then perseverated more and failed to switch after the reversal (yellow cluster), and those who sampled more before the switch but then perseverated less/switched better (magenta cluster - though see below for comments about infants' overall weak performance). The authors describe magenta babies as showing a profile of greater cognitive flexibility, which they note in adults is linked to better outcomes and a lower incidence of psychiatric disorder. Indeed, the yellow cluster scored less well on several scales of the Vineland and showed lower surgency on the IBQ than the magenta cluster. The authors argue that this paper paves the way for the field of "infant computational neuropsychiatry."

In general, I think this is a fun and intriguing paper. That said, I have a number of concerns with how it is currently written.

First, the role of pupil dilation in the models was really unclear -- I've read it through a few times and came away with different impressions each time. I am now pretty sure the models were only based on infants' behavioural responses (e.g., choice for the correct versus incorrect location) rather than differences in pupil size, but pupil size kept popping up throughout, and so I initially thought the clusters were based on that. The authors should clarify this so other readers are not confused. (One thing that might help is avoiding the word "behaviour" on its own, unless it is further specified as looking behaviour or not, as I assume that some would characterize pupil dilation as a behaviour as well.)

If clusters were NOT based on pupil size (e.g., reaction to prediction error), why not? Was this attempted, and did no clusters emerge? Did the yellow and magenta group also differ in reaction to prediction error, or not? It seems like the argument that this work will be the basis of infant computational psychiatry would require that there not simply be a link between behaviour in an infant study and other measurements of their functioning - because many other papers to date have demonstrated such relationships, many longitudinally - but instead with the link to something where the neurobiology of the behaviour being studied is better understood. I assume this is why pupil dilation kept coming up, but again, it didn't actually seem to be part of the modelling unless I missed something. That is, although I think that this is a nice finding, currently I think the novelty of the finding, as well as the suggestion that it will start a whole new field, may be overblown. I certainly think the pupillometry data has promise, as does the LUMO data, which the authors alluded to being in the works. But perhaps the implications should be toned down a bit in this paper, until those data are further along.

My final substantial comment (a few more minimal ones below) is that overall, babies did quite poorly at this task. Even after 9 post-switch trials, the magenta group was still responding at chance, and the yellow group seemed not to switch at all. Infants then all seemed to perform very well again during block 2, which makes it seem like they still had the original contingency in mind. That said, from what I could see, no data was provided about how many babies looked to the original correct first during Block 2. But based on the data, I assume they basically all went back to predicting on the first side, as otherwise their return to high levels of successful trials would not make sense, unless they somehow forgot the entire thing. It would be good to know for sure, and to have that data (specifically, how many babies looked to the original side again at the start of block 2) in the main paper. Given this overall lack of sensitive performance in the paradigm, even despite the cues signaling where the rewarding video would be changing completely (that is, the contingency between cue and outcome did not itself switch, the cues themselves did), it seems odd to discuss things like statistical or even skillful learning alongside these data.

eLife Assessment

This valuable study shows the impact of the metabolic state of bacteria on phage infection. The experimental results, based on various phages infecting E. coli, are solid and consistent with a two-step adsorption mathematical model, although the detailed evidence supporting this model is currently incomplete. This study should be of interest to the communities working on cell metabolism and on host-pathogen interactions.

Reviewer #1 (Public review):

In the wild, bacteria can be found in a wide range of metabolic states, including states in which they are resource-limited. Because phages heavily rely on the infected cell's molecular machinery to replicate, it is natural to wonder how phage-bacteria interactions depend on the metabolic state of the cell. In this work, Marantos et al. investigate specifically how the rate of infection of 5 different phages changes between cells grown in energy-rich conditions and cells grown in energy-depleted conditions. Their results clearly show that 4 out of the 5 phages studied display a significant reduction in infection rate in cells that are energetically depleted and provide a potential explanation for this observation by looking into the mechanisms that these phages use to irreversibly infect their host cells.

The work also tries to explain the observation using a mathematical/mechanistic model that describes infection as the sequence of two steps, where a phage first needs to bind to a cell receptor, from which it can potentially unbind, and then irreversibly infects by injecting its genome. While the model is sensible from a mechanistic perspective, the experimental evidence that supports how each model's rate is affected by the cell metabolic state is weak, as only ratios of these rates can be inferred from the data.

Reviewer #2 (Public review):

Summary:

The authors investigate the dependence of phage adsorption rates on host metabolic state, using 5 coliphages that differ in their infection cycles and host receptors. They find that four of the 5 phages showed significantly reduced infection under low metabolic states, with phages that generally have weaker adsorption being more strongly affected by low metabolism. The authors complement their findings with a 2-step infection model where phages can disengage from their hosts after initial adsorption. The paper illustrates the power of standardized experimental protocols for quantitative trait comparisons and highlights the dependence of phage infection success on host physiology.

Strengths:

The paper is well written and clearly structured.

The experiments are well-designed, and particularly commendable is the diligent use of control scenarios to allow for quantitative comparison between phages. This standardized protocol will be valuable for the entire phage community.

The authors convincingly show the impact of host physiology on phage adsorption success. This dependence has so far mainly been considered for intracellular phage replication, and the paper shows that host physiology has to be taken into account at all steps of phage infection.

Weaknesses:

There are some concerns about the experimental setup and which conclusions can be drawn from it:

Before phage infection, bacterial cultures are grown to exponential growth, washed, and then resuspended with glucose or arsenate-azide for 10min. It is however, questionable that 10 minutes is enough to simulate high and low metabolic states realistically. 10 minutes seems to be quite short to go from exponential growth to a low metabolic state, given the transcriptional memory of previous environments. It seems more likely that the population will be quite heterogeneous, with cells in various states of transition towards low metabolic states.

Given that arsenate and azide inhibit cellular metabolism, i.e., have antimicrobial effects, cells might not just downregulate metabolism but also activate the stress response, and this causes some of the observed effects on phage adsorption. Therefore, the 'low metabolic state' of the cells in this paper could mean that cells are starved or that they are stressed or both.

The abundance of receptors could change between the high and low metabolic media conditions and contribute to the observed differences in adsorption, while the authors seem to assume in their model that the initial adsorption rate always remains the same.

Reviewer #3 (Public review):

Summary:

Marantos et al. showed that for some coliphages, the energetic state of the bacterial host cell has a strong impact on whether phage infection is initiated. The authors drew this conclusion from the observation that there are more free phages remaining in the medium after infection of arsenate-azide-treated cells as compared to after infection of untreated cells. These data were analyzed and reported both as ratios of the treated vs. untreated conditions and using a mass-action kinetic model of phage-cell collision in the infection mixture. The data supported the findings that for four phages infecting Escherichia coli bacteria, namely, phages λ, 𝜙80, m13, and T6, the phages are less likely to initiate infection if the host bacteria are energy-depleted. However, for phage T5, the authors found that their infection propensity is not impacted.

Strengths:

The data presented by the authors clearly supported the principal conclusion of the study ("Viral commitment to infection depends on host metabolism"). The five phages chosen by the authors represent different viral lifestyles and infection mechanisms, highlighting the potential applicability to other Escherichia coli phages. Finally, the authors successfully used a classic mass-action model of phage-cell collision to interpret their data. The simplicity of their experimental assay, combined with the use of this mathematical model, offers other investigators who study phage-bacterial interactions in other contexts a potentially useful toolkit to examine infection in general, and specifically, the dependence of phage infection on the host's metabolic state.

Weaknesses:

(1) The authors isolated and measured the numbers of free phages in the medium after infection of bacteria under different treatments. These measurements were analyzed in two different ways: (1) simply as ratios (corrected/normalized using different controls), and (2) fitted using a simple mathematical model. I have concerns regarding both analyses.

1.1) For the first method, having different time points at which the sample of each phage is collected critically complicates data interpretation. As one incubates the phage-bacteria mixture for a longer time, more infection occurs, and the number of phages collected from the mixture decreases. Therefore, the different incubation time forfeits the goal of "a systematic and quantitative comparison across different phages [...]" (line 81), just as the authors self-criticized. Conceivably, the authors could have used the shortest measurement time for all phages (i.e., 10 minutes, as for phage λ). Alternatively, the authors could have applied a systematic criterion such as half (or any other fraction) of the latent period of each phage, which would still "maximize the incubation period while ensuring that manipulations were completed before the first infection cycle concluded" (lines 126-127). In my view, the seemingly arbitrary measurement time for each phage renders the entire first analysis very challenging to interpret. It also goes against the author's proposition that the protocol was "standardized" (line 92) or "consistent" (line 200). It is not clear what the readers are supposed to take away from this first analysis, or rather, which evidence, finding, or conclusion the manuscript would lose if the authors only presented the modeling-based analysis.

1.2) The second method of analysis sought to remove the dependence of the measurements on time. I completely agree with this goal, and the findings extracted from this analysis significantly contributed to the merits of this manuscript. However, the authors achieved this goal using a single time point for each phage to calculate the infection rate (η). As shown in Figure S3, each of the phage depletion curves is anchored by only one data point (note that the P(t)/P(0) = 1 at t = 0 is assumed, not measured). This goes against the typical way this collision model is used in the literature, where a time series is measured and used to fit the model (e.g., DOI 10.1007/978-1-60327-164-6 18, or more recently, PMID 39700139). This practice in the current manuscript reduced the robustness of the inferred η values. This problem is exacerbated by assumptions used by the authors in formulating this model. For instance, the authors used a constant value for the bacterial concentration, B, because "bacterial growth and lysis were negligible" (lines 135-136). However, considering that the bacteria were cultured at 37oC in a very rich medium (first in YT broth, then in 2% glucose), the measurement times of 20, 30, and 55 minutes are most likely one or a few generations of bacterial growth and division.

Related note: I suggest that one of the panels in Figure S3 should be moved to the main text, since it is critical to the second method of analysis.

(2) The data were able to distinguish phages that successfully infected bacteria and those that remained free in the medium, and the authors appropriately interpreted the data as such throughout the Results section. However, in the Discussion (starting from the very first sentence, line 172), the authors used terms that include "adsorption" and "entry" more interchangeably (for example, see the three sentences in lines 310-313, for "viral entry efficiency is shaped by [...]", then "adsorption kinetics modeling"). I do not see how the authors' data could distinguish between adsorption (the phage particles attaching to the outside of the cell) and entry (the phage DNA being injected into the cell). Conceivably, any phage particles that irreversibly attach to a cell but do not yet inject their genome into the cell would still be removed from the medium and therefore not quantified. Another example: in lines 189-191, the authors interpreted that "[...] when the bacterium is in a low metabolic state, the phage does not bind irreversibly to the host", but how do the authors eliminate the case of no phage binding (i.e., the reversible step) to begin with? Similarly, in lines 283-293, how do the authors delineate whether energy depletion would increase the k_off term or decrease the k_inj term, because either would result in more free phages in the medium as observed in the data? I believe that the writing of the Discussion, as it stands now, is doing a disservice to the conclusions presented in the Results section.

(3) The authors presented an argument that performing infection of all five phages in the same condition is an advantage, allowing for comparison across different phages. While this goal is a completely valid one, it is difficult to reconcile that with the fact that different phages require different optimal conditions for successful infection. For instance, phage T5 famously requires Ca2+ for successful infection into the host bacterium (and later successful replication); see PMID 13174489. However, all infections were performed in TMG, which lacks Ca2+. Perhaps the absence of T5 dependence on the host metabolism is because the infection condition used by the authors was not optimal for T5 to begin with? Similar arguments could be made for other phages.

(4) Whereas the manuscript examined five coliphages, only phage T5 and phage λ were discussed extensively. I believe some discussion points for these two phages need clarification.

4.1) Phage T5: The data obtained by the authors show that the infection rate of phage T5 is not impacted by the metabolic state of the host cell. Considering that the authors used the terms "infection", "adsorption", and "entry" interchangeably to refer to the irreversible commitment of a phage to a host cell (see point 2), this discussion regarding phage T5 lacks one critical literature context: DNA entry of phage T5 is known to occur in two phases (first-step transfer and second-step transfer). Critically, the second step can only occur if phage proteins encoded by the phage DNA transferred in the first step are expressed (see PMID 10577483 and the cited papers therein). In that context, metabolic poisoning of the host bacteria should have impeded T5 infection. The authors should comment on this point.

4.2) Phage λ: The experiment using phage λ in this current study shares many resemblances to that in Brown et al. 2022. That feature alone is not a problem, but at many places in the text, the writing is ambiguous as to whether it is discussing the results in Brown et al. 2022 or in the current manuscript. I am giving three examples below, but this is not exhaustive: (i) Lines 67-69, there is no Brown et al. 2022 reference immediately after "a mutant phage variant (λh) could bypass this dependency [...]" (not just in the previous sentence); (ii) Line 228 should clearly say "Our previous findings suggested that phage λ is capable of [...]", since it concerns Brown et al., 2022, not the current study; and (iii) Lines 245-246, there is no Brown et al., 2022 reference immediately after "we observed that a mutant variant [...] even energy-depleted host" (without a reference, it reads like the authors "observed" that finding in this current manuscript).

Also, regarding phage λ: The discussion between line 230 and line 249 is very interesting, but since it concerns the differences between λ PaPa and Ur-λ, the authors should consider mentioning and discussing a very relevant recent study, PMCID: PMC6312755.

(5) Control experiments, or references to prior studies, are needed to support that the As/Az treatment at this concentration and duration (at least 10 minutes) is sufficient to deplete the metabolic state of the cell. For instance, this can be shown by impeded or null cell growth, arrested motility (using a standard swimming assay), or a fluorescent reporter for the energetic state of the cell.

eLife Assessment

Zandvoort and colleagues describe respiration-brain coupling in the context of apnoea in human newborns. The authors have addressed an important question and supported their claims with solid data. The rigor of the findings could perhaps be further strengthened with some relatively minor changes to the analysis methodology.

Reviewer #1 (Public review):

Summary:

The authors investigated the extent to which phase-amplitude coupling (PAC) of respiratory and electrophysiological brain activity recordings was related to episodes of life-threatening apnoea in human newborns.

Strengths:

I want to commend the authors for acquiring unique and illuminating data; the difficulty in recording and handling these data has to be appreciated. As far as I can tell, Zandvoort and colleagues are the first to provide robust evidence for respiration-brain coupling in newborns. Their creative use of the phase-slope index for peripheral-central interactions is innovative and credible. If proven to be robust, the authors' findings have important implications well beyond the field of brain-body research.

Weaknesses:

While the analyses were overall competently conducted and well-justified, I was not entirely convinced by a few methodological choices, specifically i) the computation of PAC surrogates, ii) details of the linear mixed-effects model, and iii) the electrode selection for linking phase-amplitude coupling to apnoea frequency.

Reviewer #2 (Public review):

Summary:

The author's central hypothesis was that the strength of cortico-respiratory coupling in infants is negatively associated with apnoea rate. To prove this, they first investigated the existence of cortico-respiratory coupling in premature and term-born infants, the spatial localisation of the cortical activity and its relationship with the phase of the respiratory cycle, and the directionality of coupling.

Strengths:

The researchers used synchronised EEG and impedance pneumography to detect the phase amplitude coupling.

They have studied a wide range of gestations, from 28 weeks to 42 weeks, including males and females. Their exclusion criteria ensured that healthy babies were studied and potential confounders of impaired respiratory activity were avoided. Their sequential approach in addressing the objectives was appropriate.

Weaknesses:

As a neonatal clinician and neuroscientist, I have commented based on my expertise. I have not commented on signal processing.

I did not identify any major weaknesses in the study. Some minor weaknesses include:

(1) Data relating to the cortical oscillations and the respiratory phase is given. However, whether this would lead to their hypothesis that the strength of cortico-respiratory coupling is negatively associated with apnoea rate is unclear. What preceding data enabled the authors to link the strength of coupling to the rate of apnoea?

(2) If we did not know of data showing the existence of cortico-respiratory coupling in newborn infants, then should it not be the first research question to examine?

(3) What are the characteristics of the infants who contributed data to establish the cortico-respiratory coupling (Figures 2 and 3)?

(4) Although it is the most plausible direction of the relationship, with neural activation driving respiratory muscle contraction, how can the authors prove this with their data? Given that they show coherence between signals, how do we know that the cortical signal precedes the respiratory muscle contraction?

(5) Apgar score is an ordinal variable. The authors should summarise this as median (range).

Reviewer #3 (Public review):

Summary:

This is a strong and important report that presents a framework for understanding cortical contributions to neonatal respiration. Overall, the authors successfully achieved their goal of linking cortical activity to respiratory drive. Despite the correlational nature of this study, it is a crucial step in establishing a foundation for future work to elucidate the interaction between cortical activity and breathing.

Strengths:

(1) The introduction and use of workflows that establish correlational relationships between breathing and brain activity.

(2) The execution of these workflows in human neonates.

Weaknesses:

Interpretations related to causal inference, confounds of sleep and caffeine, and the spatial interpretation of EEG data need to be addressed to ensure that the data appropriately support the conclusions.

Author response:

We would like to thank the reviewers for their helpful comments and critique of our manuscript. We plan to make the following revisions, which will improve the clarity of our manuscript and the robustness of our findings.

We will revise methodological details and interpretation throughout the manuscript. In particular, we will consider alternative methods for calculating surrogates. We intend to investigate the relationship between apnoea rate and phase-amplitude coupling at other electrodes as suggested by Reviewer 1, and we will revise the details of the linear-mixed effects models.

In relation to the comments raised by both Reviewers 2 and 3, we will carefully address the wording throughout the manuscript, including addressing the order of hypotheses, our interpretation of the directionality of the relationship between cortical and respiratory activity, and the connection between cortical-respiratory coupling and apnoea. We will further clarify the limitations of our recording setup and approach, in particular the limited EEG montage, and add further details with regards to sleep state and caffeine.

eLife Assessment

This paper reports a valuable finding that gastric fluid DNA content can be used as a potential biomarker for human gastric cancer. The evidence supporting the claims of the authors is solid, although an inclusion of explanations for the methodological limitations, moderate diagnostic performance, and the unexpected survival correlation would have strengthened the study. The work will be of interest to medical biologists working in the field of breast cancer.

Reviewer #1 (Public review):

The study analyzes the gastric fluid DNA content identified as a potential biomarker for human gastric cancer. However, the study lacks overall logicality, and several key issues require improvement and clarification. In the opinion of this reviewer, some major revisions are needed:

(1) This manuscript lacks a comparison of gastric cancer patients' stages with PN and N+PD patients, especially T0-T2 patients.

(2) The comparison between gastric cancer stages seems only to reveal the difference between T3 patients and early-stage gastric cancer patients, which raises doubts about the authenticity of the previous differences between gastric cancer patients and normal patients, whether it is only due to the higher number of T3 patients.

(3) The prognosis evaluation is too simplistic, only considering staging factors, without taking into account other factors such as tumor pathology and the time from onset to tumor detection.

(4) The comparison between gfDNA and conventional pathological examination methods should be mentioned, reflecting advantages such as accuracy and patient comfort.

(5) There are many questions in the figures and tables. Please match the Title, Figure legends, Footnote, Alphabetic order, etc.

(6) The overall logicality of the manuscript is not rigorous enough, with few discussion factors, and cannot represent the conclusions drawn

Reviewer #2 (Public review):

Summary:

The authors investigated whether the total DNA concentration in gastric fluid (gfDNA), collected via routine esophagogastroduodenoscopy (EGD), could serve as a diagnostic and prognostic biomarker for gastric cancer. In a large patient cohort (initial n=1,056; analyzed n=941), they found that gfDNA levels were significantly higher in gastric cancer patients compared to non-cancer, gastritis, and precancerous lesion groups. Unexpectedly, higher gfDNA concentrations were also significantly associated with better survival prognosis and positively correlated with immune cell infiltration. The authors proposed that gfDNA may reflect both tumor burden and immune activity, potentially serving as a cost-effective and convenient liquid biopsy tool to assist in gastric cancer diagnosis, staging, and follow-up.

Strengths:

This study is supported by a robust sample size (n=941) with clear patient classification, enabling reliable statistical analysis. It employs a simple, low-threshold method for measuring total gfDNA, making it suitable for large-scale clinical use. Clinical confounders, including age, sex, BMI, gastric fluid pH, and PPI use, were systematically controlled. The findings demonstrate both diagnostic and prognostic value of gfDNA, as its concentration can help distinguish gastric cancer patients and correlates with tumor progression and survival. Additionally, preliminary mechanistic data reveal a significant association between elevated gfDNA levels and increased immune cell infiltration in tumors (p=0.001).

Weaknesses:

The study has several notable weaknesses. The association between high gfDNA levels and better survival contradicts conventional expectations and raises concerns about the biological interpretation of the findings. The diagnostic performance of gfDNA alone was only moderate, and the study did not explore potential improvements through combination with established biomarkers. Methodological limitations include a lack of control for pre-analytical variables, the absence of longitudinal data, and imbalanced group sizes, which may affect the robustness and generalizability of the results. Additionally, key methodological details were insufficiently reported, and the ROC analysis lacked comprehensive performance metrics, limiting the study's clinical applicability.

eLife Assessment

This study presents valuable and compelling evidence that β-glucan-induced trained immunity can protect against intestinal inflammation by reprogramming innate immune cells toward a reparative phenotype. The authors employ a convincing combination of functional assays, adoptive transfers, and single-cell transcriptomics to uncover mechanistic insights and demonstrate the therapeutic potential of innate immune memory in IBD. While the work is robust, addressing the underlying epigenetic mechanisms and including additional controls would further reinforce the trained immunity-specific interpretation.

Reviewer #1 (Public review):

Summary:

This study presents an interesting investigation into the role of trained immunity in inflammatory bowel disease, demonstrating that β-glucan-induced reprogramming of innate immune cells can ameliorate experimental colitis. The findings are novel and clinically relevant, with potential implications for therapeutic strategies in IBD. The combination of functional assays, adoptive transfer experiments, and single-cell RNA sequencing provides comprehensive mechanistic insights. However, some aspects of the study could benefit from further clarification to strengthen the conclusions.

Strengths:

(1) This study elegantly connects trained immunity with IBD, demonstrating how β-glucan-induced innate immune reprogramming can mitigate chronic inflammation.

(2) Adoptive transfer experiments robustly confirm the protective role of monocytes/macrophages in colitis resolution.

(3) Single-cell RNA sequencing provides mechanistic depth, revealing the expansion of reparative Cx3cr1⁺ macrophages and their contribution to epithelial repair.

(4) The work highlights the therapeutic potential of trained immunity in restoring gut homeostasis, offering new directions for IBD treatment.

Weaknesses:

While β-glucan may exert its training effect on hematopoietic stem cells, performing ATAC-seq on HSCs or monocytes to profile chromatin accessibility at antibacterial defense and mucosal repair-related genes would further validate the trained immunity mechanism. Alternatively, the authors could acknowledge this as a study limitation and future research direction.

Reviewer #2 (Public review):

Summary:

The study investigates whether β-glucan (BG) can reprogram the innate immune system to protect against intestinal inflammation. The authors show that mice pretreated with BG prior to DSS-induced colitis experience reduced colitis severity, including less weight loss, colon damage, improved gut repair, and lowered inflammation. These effects were independent of adaptive immunity and were linked to changes in monocyte function.

The authors show that the BG-trained monocytes not only help control inflammation but confer non-specific protection against experimental infections (Salmonella), suggesting the involvement of trained immunity (TI) mechanisms. Using single-cell RNA sequencing, they map the transcriptional changes in these cells and show enhanced differentiation of monocytes into reparative CX3CR1⁺ macrophages. Importantly, these protective effects were transferable to other mice via adoptive cell transfer and bone marrow transplantation, suggesting that the innate immune system had been reprogrammed at the level of stem/progenitor cells.

Overall, this study provides evidence that TI, often associated with heightened inflammatory programs, can also promote tissue repair and resolution of inflammation. Moreover, this BG-induced functional reprogramming can be further harnessed to treat chronic inflammatory disorders like IBD.

Strengths:

(1) The authors use advanced experimental approaches to explore the potential therapeutic use of myeloid reprogramming by β-glucan in IBD.

(2) The authors follow a data-to-function approach, integrating bulk and single-cell RNA sequencing with in vivo functional validation to support their conclusions.

(3) The study adds to the growing evidence that TI is not a singular pro-inflammatory program, but can adopt distinct functional states, including anti-inflammatory and reparative phenotypes, depending on the context.

Weaknesses:

(1) The epigenetic and metabolic basis of TI is not explored, which weakens the mechanistic claim of TI. This is especially relevant given that a novel reparative, anti-inflammatory TI program is proposed.

(2) The absence of a BG-only group limits interpretation of the results. Since the authors report tissue-level effects such as enhanced mucosal repair and transcriptional shifts in intestinal macrophages (colonic RNA-Seq), it is important to rule out whether BG alone could influence the gut independently of DSS-induced inflammation.<br /> Without a BG-only control, it is hard to distinguish a true trained response from a potential modulation caused directly by BG.

(3) Although monocyte transfer experiments show protection in colitis, the fate of the transferred cells is not described (e.g., homing or differentiation into Cx3cr1⁺ macrophage subsets). This weakens the link between specific monocyte subsets and the observed phenotype.

(3) While scRNA-seq reveals distinct monocyte/macrophage subclusters (Mono1-3..), their specific functional roles remain speculative. The authors assign reparative or antimicrobial functions based on transcriptional signatures, but do not perform causal experiments (depletion or in vitro assays). The biological roles of these cells remain correlative.

(4) While Rag1⁻/⁻ mice were used to rule out adaptive immunity, the potential role of innate lymphoid cells (ILCs), particularly ILC2s and ILC3s, which are known to promote mucosal repair (PMID: 27484190), was not explored. Given the reparative phenotype observed, the contribution of ILCs remains a confounding factor.

Reviewer #3 (Public review):

Summary:

In the present work, Yinyin Lv et al offer evidence for the therapeutic potential of trained immunity in the context of inflammatory bowel disease (IBD). Prior research has demonstrated that innate cells pre-treated (trained) with β-glucan show an enhanced pro-inflammatory response upon a second challenge.

While an increased immune response can be beneficial and protect against bacterial infections, there is also the risk that it will worsen symptoms in various inflammatory disorders. In the present study, the authors show that mice preconditioned with β-glucan have enhanced resistance to Staphylococcus aureus infection, indicating heightened immune responses.

The authors demonstrate that β-glucan training of bone marrow hematopoietic progenitors and peripheral monocytes mitigates the pro-inflammatory effects of colitis, with protection extending to naïve recipients of the trained cells.

Using a dextran sulfate sodium (DSS)-induced model of colitis, β-glucan pre-treatment significantly dampens disease severity. Importantly, the use of Rag1^-/- mice, which lack adaptive immune cells, confirms that the protective effects of β-glucan are mediated by innate immune mechanisms. Further, experiments using Ccr2^-/- mice underline the necessity of monocyte recruitment in mediating this protection, highlighting CCR2 as a key factor in the mobilization of β-glucan-trained monocytes to inflamed tissues. Transcriptomic profiling reveals that β-glucan training upregulates genes associated with pattern recognition, antimicrobial defense, immunomodulation, and interferon signaling pathways, suggesting broad functional reprogramming of the innate immune compartment. In addition, β-glucan training induces a distinct monocyte subpopulation with enhanced activation and phagocytic capacity. These monocytes exhibit an increased ability to infiltrate inflamed colonic tissue and differentiate into macrophages, marked by increased expression of Cx3cr1. Moreover, among these trained monocyte and macrophage subsets, other gene expression signatures are associated with tissue and mucosal repair, suggesting a role in promoting resolution and regeneration following inflammatory insult.

Strengths:

(1) Overall, the authors present a mechanistically insightful investigation that advances our understanding of trained immunity in IBD.

(2) By employing a range of well-characterized murine models, the authors investigate specific mechanisms involved in the effects of β-glucan training.

(3) Furthermore, the study provides functional evidence that the protection conferred by the trained cells persists within the hematopoietic progenitors and can be transferred to naïve recipients. The integration of transcriptomic profiling allows the identification of changes in key genes and molecular pathways underlying the trained immune phenotype.

(4) This is an important study that demonstrates that β-glucan-trained innate cells confer protection against colitis and promote mucosal repair, and these findings underscore the potential of harnessing innate immune memory as a therapeutic approach for chronic inflammatory diseases.

Weaknesses:

However, FPKM is not ideal for between-sample comparisons due to its within-sample normalization approach. Best practices recommend using raw counts (with DESeq2) for more robust statistical inference.

Author response:

Public Reviews:

Reviewer #1 (Public review):

Summary:

This study presents an interesting investigation into the role of trained immunity in inflammatory bowel disease, demonstrating that β-glucan-induced reprogramming of innate immune cells can ameliorate experimental colitis. The findings are novel and clinically relevant, with potential implications for therapeutic strategies in IBD. The combination of functional assays, adoptive transfer experiments, and single-cell RNA sequencing provides comprehensive mechanistic insights. However, some aspects of the study could benefit from further clarification to strengthen the conclusions.

We are grateful for the reviewer’s positive assessment of our study and constructive suggestions to improve the manuscript.

Strengths:

(1) This study elegantly connects trained immunity with IBD, demonstrating how β-glucan-induced innate immune reprogramming can mitigate chronic inflammation.

(2) Adoptive transfer experiments robustly confirm the protective role of monocytes/macrophages in colitis resolution.

(3) Single-cell RNA sequencing provides mechanistic depth, revealing the expansion of reparative Cx3cr1⁺ macrophages and their contribution to epithelial repair.

(4) The work highlights the therapeutic potential of trained immunity in restoring gut homeostasis, offering new directions for IBD treatment.

Weaknesses:

While β-glucan may exert its training effect on hematopoietic stem cells, performing ATAC-seq on HSCs or monocytes to profile chromatin accessibility at antibacterial defense and mucosal repair-related genes would further validate the trained immunity mechanism. Alternatively, the authors could acknowledge this as a study limitation and future research direction.

We agree that further epigenetic profiling—such as ATAC-seq analysis on HSCs or monocytes—would provide additional mechanistic depth to our current findings. We will acknowledge this as a limitation of the present study and highlight it as an important direction for future research.

Comment (1): It’s better to include a schematic summarizing the proposed mechanism for reader clarity.

We agree that a visual summary will enhance the clarity and accessibility of our findings. We will add a new schematic diagram (Figure 6) illustrating the proposed mechanism of β-glucan–induced myeloid reprogramming and its protective effects in the experimental colitis model.

Comment (2): Discuss potential off-target effects of β-glucan-induced trained immunity (e.g., risk of exacerbated inflammation in other contexts).

We appreciate this important comment regarding the potential off-target effects of β-glucan pretreatment. As trained immunity is known to amplify inflammatory responses upon heterologous stimulation and has been implicated in chronic inflammation–prone conditions such as atherosclerosis, this is an important consideration. Previous in vivo studies have shown that β-glucan pretreatment can enhance antibacterial or antitumor responses without inducing basal inflammation after one week of administration (PMID: 22901542, PMID: 30380404, PMID: 36604547, PMID: 33125892). Nevertheless, it remains possible that β-glucan–induced trained immunity could have unintended effects in certain contexts, which warrants further investigation and caution. We will expand the Discussion section to include a dedicated paragraph addressing these potential off-target effects.

Reviewer #2 (Public review):

Summary:

The study investigates whether β-glucan (BG) can reprogram the innate immune system to protect against intestinal inflammation. The authors show that mice pretreated with BG prior to DSS-induced colitis experience reduced colitis severity, including less weight loss, colon damage, improved gut repair, and lowered inflammation. These effects were independent of adaptive immunity and were linked to changes in monocyte function.

The authors show that the BG-trained monocytes not only help control inflammation but confer non-specific protection against experimental infections (Salmonella), suggesting the involvement of trained immunity (TI) mechanisms. Using single-cell RNA sequencing, they map the transcriptional changes in these cells and show enhanced differentiation of monocytes into reparative CX3CR1⁺ macrophages. Importantly, these protective effects were transferable to other mice via adoptive cell transfer and bone marrow transplantation, suggesting that the innate immune system had been reprogrammed at the level of stem/progenitor cells.

Overall, this study provides evidence that TI, often associated with heightened inflammatory programs, can also promote tissue repair and resolution of inflammation. Moreover, this BG-induced functional reprogramming can be further harnessed to treat chronic inflammatory disorders like IBD.

Strengths:

(1) The authors use advanced experimental approaches to explore the potential therapeutic use of myeloid reprogramming by β-glucan in IBD.

(2) The authors follow a data-to-function approach, integrating bulk and single-cell RNA sequencing with in vivo functional validation to support their conclusions.

(3) The study adds to the growing evidence that TI is not a singular pro-inflammatory program, but can adopt distinct functional states, including anti-inflammatory and reparative phenotypes, depending on the context.

We are grateful for the reviewer’s positive assessment of our study and recognition of its translational implications. We particularly appreciate the acknowledgment that our work expands the therapeutic potential of β-glucan–mediated trained immunity in ameliorating colitis.

Weaknesses:

(1) The epigenetic and metabolic basis of TI is not explored, which weakens the mechanistic claim of TI. This is especially relevant given that a novel reparative, anti-inflammatory TI program is proposed.

We appreciate the reviewer’s valuable comment highlighting the importance of the epigenetic and metabolic basis of TI in providing mechanistic insight. While previous studies, including work from our group (S.-C. Cheng), have extensively characterized the epigenetic and metabolic signatures of monocytes from BG-trained mice—primarily in the context of inflammatory genes—we acknowledge that these aspects are not directly addressed in our current manuscript.

To strengthen the mechanistic component, we plan to: 1. Reanalyze relevant public datasets, focusing on pathways related to reparative and antibacterial function. 2. Perform monocyte ATAC-seq in our current model to validate the epigenetic changes in these pathways.

(2) The absence of a BG-only group limits interpretation of the results. Since the authors report tissue-level effects such as enhanced mucosal repair and transcriptional shifts in intestinal macrophages (colonic RNA-Seq), it is important to rule out whether BG alone could influence the gut independently of DSS-induced inflammation.

Without a BG-only control, it is hard to distinguish a true trained response from a potential modulation caused directly by BG.

We thank the reviewer for this important suggestion. Although we did not perform qPCR for mucosal repair genes in Figure S1C and Figure S1D, our colon RNA-seq analysis in Figure 5G included a BG-only control group (Colitis_d0). The results from this group indicate that BG preconditioning alone does not alter baseline expression of colon mucosal repair genes, supporting the conclusion that the observed effects occur in the context of DSS-induced inflammation.

(3) Although monocyte transfer experiments show protection in colitis, the fate of the transferred cells is not described (e.g., homing or differentiation into Cx3cr1⁺ macrophage subsets). This weakens the link between specific monocyte subsets and the observed phenotype.

(4) While scRNA-seq reveals distinct monocyte/macrophage subclusters (Mono1-3.), their specific functional roles remain speculative. The authors assign reparative or antimicrobial functions based on transcriptional signatures, but do not perform causal experiments (depletion or in vitro assays). The biological roles of these cells remain correlative.

We agree that the functional role of CX3CR1⁺ macrophages is not comprehensively validated and is currently inferred from scRNA-seq clustering. While our flow cytometry data show increased CX3CR1⁺ macrophages in the BG-TI group, and our CCR2 KO and monocyte adoptive transfer experiments indicate these macrophages are monocyte-derived, we lack direct depletion experiments due to the unavailability of effective depletion antibodies for this subset.

We acknowledge this as a limitation and will clarify in the Discussion that our conclusions regarding CX3CR1⁺ macrophage function are based on transcriptional profiling and association with protective phenotypes, rather than direct causal evidence.

(5) While Rag1^-/- mice were used to rule out adaptive immunity, the potential role of innate lymphoid cells (ILCs), particularly ILC2s and ILC3s, which are known to promote mucosal repair (PMID: 27484190IF: 7.6 Q1 IF: 7.6 Q1 IF: 7.6 Q1 IF: 7.6 Q1 IF: 7.6 Q1 IF: 7.6 Q1 ), was not explored. Given the reparative phenotype observed, the contribution of ILCs remains a confounding factor.

We appreciate the reviewer’s valuable comment regarding the potential role of ILCs in the observed mucosal repair. Indeed, in examining the BG-trained immunity effect, the contribution of ILCs was not evaluated. We will explicitly acknowledge in the Discussion that Rag1⁻/⁻ mice retain ILCs (including ILC3s) and that BG-induced activation of these cells remains possible.

The literature (PMID: 21502992; PMID: 32187516) supports a role for ILC3-mediated IL-22 production in tissue repair, which could overlap with our observed effects. However, our monocyte adoptive transfer experiments show that monocytes alone can alleviate DSS-induced colitis, suggesting a dominant role for monocytes in this context. Nonetheless, we will make it clear that ILC contributions cannot be excluded.

Reviewer #3 (Public review):

Summary:

In the present work, Yinyin Lv et al offer evidence for the therapeutic potential of trained immunity in the context of inflammatory bowel disease (IBD). Prior research has demonstrated that innate cells pre-treated (trained) with β-glucan show an enhanced pro-inflammatory response upon a second challenge.

While an increased immune response can be beneficial and protect against bacterial infections, there is also the risk that it will worsen symptoms in various inflammatory disorders. In the present study, the authors show that mice preconditioned with β-glucan have enhanced resistance to Staphylococcus aureus infection, indicating heightened immune responses.

The authors demonstrate that β-glucan training of bone marrow hematopoietic progenitors and peripheral monocytes mitigates the pro-inflammatory effects of colitis, with protection extending to naïve recipients of the trained cells.

Using a dextran sulfate sodium (DSS)-induced model of colitis, β-glucan pre-treatment significantly dampens disease severity. Importantly, the use of Rag1^-/- mice, which lack adaptive immune cells, confirms that the protective effects of β-glucan are mediated by innate immune mechanisms. Further, experiments using Ccr2^-/- mice underline the necessity of monocyte recruitment in mediating this protection, highlighting CCR2 as a key factor in the mobilization of β-glucan-trained monocytes to inflamed tissues. Transcriptomic profiling reveals that β-glucan training upregulates genes associated with pattern recognition, antimicrobial defense, immunomodulation, and interferon signaling pathways, suggesting broad functional reprogramming of the innate immune compartment. In addition, β-glucan training induces a distinct monocyte subpopulation with enhanced activation and phagocytic capacity. These monocytes exhibit an increased ability to infiltrate inflamed colonic tissue and differentiate into macrophages, marked by increased expression of Cx3cr1. Moreover, among these trained monocyte and macrophage subsets, other gene expression signatures are associated with tissue and mucosal repair, suggesting a role in promoting resolution and regeneration following inflammatory insult.

Strengths:

(1) Overall, the authors present a mechanistically insightful investigation that advances our understanding of trained immunity in IBD.

(2) By employing a range of well-characterized murine models, the authors investigate specific mechanisms involved in the effects of β-glucan training.

(3) Furthermore, the study provides functional evidence that the protection conferred by the trained cells persists within the hematopoietic progenitors and can be transferred to naïve recipients. The integration of transcriptomic profiling allows the identification of changes in key genes and molecular pathways underlying the trained immune phenotype.

(4) This is an important study that demonstrates that β-glucan-trained innate cells confer protection against colitis and promote mucosal repair, and these findings underscore the potential of harnessing innate immune memory as a therapeutic approach for chronic inflammatory diseases.

We thank the reviewer for their positive evaluation and constructive feedback on our manuscript.

Weaknesses:

However, FPKM is not ideal for between-sample comparisons due to its within-sample normalization approach. Best practices recommend using raw counts (with DESeq2) for more robust statistical inference.

We appreciate the reminder about best practices for RNA-seq analysis. We apologize for the inaccurate description in the Materials and Methods section. For all differential expression analyses, we have in fact used raw count data as input for DESeq2. FPKM values were only used for visualization purposes, such as in heatmaps and clustering analyses. We will correct this description in the revised manuscript to accurately reflect our analysis workflow.

eLife Assessment

The study by Takagi and colleagues is an important contribution to the question of how homologous neuronal circuits might be wired differently to elicit specific behaviours. The authors combine genetic, neuroanatomical, and behavioral data to provide convincing evidence that Dfz2/DWnt4 signaling controls the innervation pattern of wave command neurons in the fly larva, and thereby behavioral locomotion program selection.

Reviewer #1 (Public review):

Summary

In this study Takagi and colleagues demonstrate that changes in axonal arborization of the segmental wave motor command neurons are sufficient to change behavioral motor output.

The authors identify the Wnt receptors DFz2 and DFz4 and the ligand Wnt4 as modulators of the stereotypic segmental arborization pattern of segmental wave neurons along the anterior-posterior body axis. Based on both embryonic expression pattern analysis and genetic manipulation of the signaling components in wave neurons (receptors) and the neuropil (Wnt4) the authors convincingly demonstrate that Wnt4 acts as a repulsive ligand for DFz2 that restricts posterior axon guidance of both anterior and posterior wave neurons. They also provide first evidence that Wnt4 potentially acts as an attractive ligand for Df4 to promote posterior extension of p-wave neurons. Interestingly, artificial optogenetic activation of all wave neurons that normally induces a backward locomotion due to the activity of anterior wave neurons, fails to induce backward locomotion in a DFz2 knock down condition with altered axonal extensions of all wave neurons towards posterior segments. In addition, the authors now observe enhanced fast forward locomotion a feature normally induced by posterior wave neurons. Consistent with these findings, they observe that the natural response to an anterior tactile stimulus is similarly altered in DFz2 knock down animals. The animals respond with less backward movement and increase fast forward motion. These results suggest that alterations in the innervation pattern of wave motor command neurons are sufficient to switch behavioral response programs.

Strengths

The authors convincingly demonstrate the importance of Wnt signaling for anterior-posterior axon guidance of a single class of motor command neurons in the larval CNS. The demonstration that alteration of the expression level of a single axon guidance receptor is sufficient to not only alter the innervation pattern but to significantly modify the behavioral response program of the animal provides a potential entry point to understand behavioral adaptations during evolution.

Weaknesses

The authors demonstrate an alteration of the behavioral response to a natural tactile stimulus and correlate this to morphological alterations observed in the single-neuron analyses. As the authors suggest an alteration of the command circuitry, a direct observation of the downstream activation pattern in response to selective optogenetic stimulation of anterior wave neurons (if possible with appropriate genetic tools in the future) would further strengthen their claims.

Reviewer #2 (Public review):

Summary:

In the manuscript, the authors aim to determine the molecular mechanisms involved in wiring the segmentally homologous a- and p -Wave neurons distinctively and thus are functionally different in modulating forward or backward locomotion. The genetic screen focused on Wnt/Fz-signaling due to its known anterior-to-posterior guidance roles in mammals and nematodes.

Strengths:

The conclusion that Frizzled receptors DFz2 and DFz4 as well as the DWnt4 ligand is essential for normal segment-specific axon projections of Wave command neurons is strongly supported by the elaborate morphological analyses of numerous Wnt/Fz in gain and loss of function mutants. The distinctive Wnt/Fz ligand-receptor gradients also imply that they contribute to the diversification of Wave neurons in a location-dependent manner and that DFz2 and DFz4 may have opposing effects on axon extension.

Labeling of synaptic marker Bruchpilot in DFz2 mutants in this revised manuscript, now supports that the ectopic projections in a-Wave neurons make synaptic connections. Finally, the altered responses in two behavioral assays (optogenetic stimulation of all Wave neurons or tactile stimuli on heads using a von Frey filament) further strongly support the main conclusion, that Wnt/Fz-signaling is essential for the guidance of both Wave neurons and in diversifying their protection pattern in a segment-specific manner.

Weaknesses:

There are no major weaknesses in the revised version of this work.

Re-analysis of DFz2 expression now shows it is bidirectionally distributed. This new result does not affect the previous and current conclusions for the a-Wave neurons but leaves alternative interpretations for p-Wave neurons, which the author now included in their discussions. Evidently, it seems unlikely that the complex wiring of the numerous segmental a- and p-Wave neurons will be solely dependent on Wnt4-DFz2/4 but are likely to also involve other Wnt/Fz (see, Figure 1-figure supplement 2) or distinct guidance signaling pathways. However, unraveling all factors involved is certainly beyond the scope of this study, and the main conclusions made by the authors are well supported by the data provided.

Author response:

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

Reviewer #1 (Public Review):

Summary

In this study, Takagi and colleagues demonstrate that changes in axonal arborization of the segmental wave motor command neurons are sufficient to change behavioral motor output.

The authors identify the Wnt receptors DFz2 and DFz4 and the ligand Wnt4 as modulators of stereotypic segmental arborization patterns of segmental wave neurons along the anterior-posterior body axis. Based on both embryonic expression pattern analysis and genetic manipulation of the signaling components in wave neurons (receptors) and the neuropil (Wnt4) the authors convincingly demonstrate that Wnt4 acts as a repulsive ligand for DFz2 that restricts posterior axon guidance of both anterior and posterior wave neurons. They also provide the first evidence that Wnt4 potentially acts as an attractive ligand for Df4 to promote the posterior extension of p-wave neurons. Interestingly, artificial optogenetic activation of all wave neurons that normally induces backward locomotion due to the activity of anterior wave neurons, fails to induce backward locomotion in a DFz2 knockdown condition with altered axonal extensions of all wave neurons towards posterior segments. In addition, the authors now observe enhanced fast-forward locomotion, a feature normally induced by posterior wave neurons. Consistent with these findings, they observe that the natural response to an anterior tactile stimulus is similarly altered in DFz2 knockdown animals. The animals respond with less backward movement and increased fast forward motion. These results suggest that alterations in the innervation pattern of wave motor command neurons are sufficient to switch behavioral response programs.

Strengths

The authors convincingly demonstrate the importance of Wnt signaling for anteriorposterior axon guidance of a single class of motor command neurons in the larval CNS. The demonstration that alteration of the expression level of a single axon guidance receptor is sufficient to not only alter the innervation pattern but to significantly modify the behavioral response program of the animal provides a potential entry point to understanding behavioral adaptations during evolution.

Weaknesses

While the authors demonstrate an alteration of the behavioral response to a natural tactile stimulus the observed effects, a reduction of backward motion and increased fast-foward locomotion, currently cannot be directly correlated to the morphological alterations observed in the single-neuron analyses. The authors do not report any loss of innervation in the "normal" target region but only a small additional innervation of more posterior regions. An analysis of synaptic connectivity and/or a more detailed morphological analysis that is supported by a larger number of analyzed neurons both in control and experimental animals would further strengthen the confidence of the study. As the authors suggest an alteration of the command circuitry, a direct observation of the downstream activation pattern in response to selective optogenetic stimulation of anterior wave neurons would further strengthen their claims (analogous to Takagi et al., 2017, Figure 4).

We sincerely thank the reviewer for their insightful comments, which were instrumental in improving our manuscript. In response to the reviewers’ suggestion, we have now studied Brp expression and demonstrate that the ectopically extending Wave axons in the posterior region do contain synapses (new Figure 2). This finding supports the idea that these axons are functionally connected to ectopic downstream circuits. 

Additionally, we have increased the number of analyzed Wave clones in Figure 1F-J (WT and DFz2 KD) and new Figure 3C-G (WT; formerly Figure 2C-G) to strengthen the morphological analyses. We fully agree with the reviewer that “direct observation of the downstream activation pattern in response to selective optogenetic stimulation” would further reinforce our conclusions. However, this was not feasible in the current study since we found that the Wave-Gal4 driver used in this study, which drives expression during embryonic stages, does not drive sufficiently strong expression in the larvae to enable selective optogenetic stimulation (please see below for details). 

Reviewer #2 (Public Review):

Summary:

The authors previously demonstrated that anterior-located a-Wave neurons (neuromeres A1-A3) extend axons anteriorly to connect to circuits inducing backward locomotion, while p-Wave axon (neuromeres A4-A7) project posteriorly to promote forward locomotion in Drosophila larvae. In the manuscript, the authors aim to determine the molecular mechanisms involved in wiring the segmentally homologous Wave neurons distinctively and thus are functionally different in modulating forward or backward locomotion. The genetic screen focused on Wnt/Fz-signaling due to its known anterior-to-posterior guidance roles in mammals and nematodes.

Strengths:

Knock-down (KD) DFz2 with two independent RNAi-lines caused ectopic posterior axon and dendrite extension for all a- and p-Wave neurons, with a-Wave axon extending into regions where p-Wave axons normally project. Both behavioral assays (optogenetic stimulation of all Wave neurons or tactile stimuli on heads using a von Frey filament) show that backward movement is reduced or absent and that the speed of evoked fast-forward locomotion is increased. This demonstrates that altered projections of Wave do alter behavior and the DFz2 KD phenotype is consistent with the potential aberrant wiring of a-Wave neurons to forward locomotion-promoting circuits instead of to backward locomotion-promoting circuits.

The main conclusion, that Wnt/Fz-signaling is essential for the guidance of Wave neurons and in diversifying their protection pattern in a segment-specific manner, is further supported by the results showing that DFz2 gain of function causes shortening of a-Wave but not p-Wave axon extensions towards the posterior end and that KD of DFz4 causes axonal shortening only in A6-p-Wave neurons but does not affect dendrites or processes of other Wave neurons. A role for ligand Wnt4 is demonstrated by results indicating that WNT4 mutants' posterior extension of aWave axons was elongated similar to DFz2 KD animals and p-Wave axon extension towards the posterior end was shortened similar to DFz2 KD animals. Finally, a DWnt4 gradient decreasing from the posterior (A8) to the anterior end (A2), similar to that described in other species, is supported by analyses of DWnt4 gene expression (using Wnt4 Trojan-Gal4) and protein expression (using antibodies). In contrast, DFz2 receptor levels seemed to decrease from the anterior (A2) to the posterior end (A5/6). Together the results support the conclusion that opposing Wnt/Fz ligand-receptor gradients contribute to the diversification of Wave neurons in a location-dependent manner and that DFz2 and DFz4 have opposing effects on axon extension.

Weaknesses:

Wave axon and dendrite projections are not exclusively determined by Wnt4, DFz2, and DFz4, and are likely to involve other Fz receptors, Wt ligands, and other types of receptor-ligand signaling pathways. This is in part supported by the fact that Wnt4 loss of function also resulted in phenotypes that do not mimic DFz2 KD or DFz4 KD (Figures 3D, E, and F) and that other Fz/Wnt mutants caused wave neuron phenotypes (Figure 1-supplement 2, D+E). This is not a weakness per se, since it doesn't affect the main conclusion of the manuscript. However, the description and analyses of the data in particular for Figure 1-supplement 2 D should be clarified in the legend. The number within the bars and the asterisks are not defined. It's presumed they refer to numbers of animals assessed and the asterisk next to DFz2 and DFz4 indicate statistically significant differences. However, only one p-value is provided in the legend. It is also unclear if p-values for the other mutants have not been determined or are non-significant. At least for mutants like Corin, which also exhibit altered axon projections, the p-values should be provided.

We appreciate this reviewer’s careful attention to detail and intellectual curiosity. We apologize for the confusions caused by the statistical reporting in Figure 1 – figure supplement 2D. The numbers shown in the bars represent the number of neurons (i.e. Wave neurons from left or right hemisphere). As mentioned in Materials and Methods section, we applied Chi-square test followed by Haberman's adjusted residual analysis to determine the statistical significance of each RNAi group. The p-value provided in the figure legend corresponds to the Chi-square test. P-values for Haberman's adjusted residual analysis were calculated for all RNAi groups and groups without the asterisk are not statistically significant. We have clarified these points in the corresponding figure legend.

Figure 4 D, F. The gradient for Wnt4 was determined by comparison of expression levels of other segments to A8 but the gradient for DFz2 was by comparison to A2 and the data supports opposing gradients. However, for DFz2 (Figure 4, F) it seems that the gradient is bi-directional with the lowest being in A5 and increasing towards A2 as well as A8. Analysis should be performed in reference to A8 as well to determine if it is indeed bi-directional. While such a finding would not affect the interpretation of aWave neurons, it may impact conclusions about p-Wave neuron projections.

We thank the reviewer for highlighting this interesting possibility. In response, we performed an additional analysis of the DFz2 gradient by comparing the signal from each neuromere to that from A8 (new Figure 5—figure supplement 3). This analysis confirmed that the gradient is indeed bidirectional. We revised the description of DFz2 expression accordingly in the revision. We believe this finding does not affect our main conclusions since only the anterior gradient is relevant for a-Wave axon guidance. 

As discussed above, the DFz2 KD phenotypes are consistent with the potential aberrant wiring of a-Wave neurons to forward locomotion-promoting circuits instead of to backward locomotion-promoting circuits. However, since the axon and dendrites of a-Wave and p-Wave are affected the actual dendritic and axonal contributions for the altered behavior remain elusive. The authors certainly considered a potential contribution of altered dendrite projection of a-Wave neurons to the phenotype and their conclusion that altered axonal projections are involved is supported by the optogenetic experiment "bypassing" sensory input (albeit it seems unlikely that all Wave neurons are activated simultaneously when perceiving natural stimuli).However, the author should also consider that altered perception and projection of pWave neuron may directly (e.g. extended P-wave axon projections increase forward locomotion input thereby overriding backward locomotion) or indirectly (e.g. feedback loops between forward and backward circuits) contribute to the altered behavioral phenotypes in both assays. It is probably noteworthy that the more complex behavioral alterations observed with mechanical stimulation are likely to also be caused by altered dendritic projections.

We fully agree with the reviewer’s thoughtful interpretation. We have now included these important possibilities in the revised Discussion section. Specifically, we acknowledge that while the DFz2 knockdown phenotypes are consistent with aberrant wiring of a-Wave neurons to forward locomotion-promoting circuits, the contributions of both axonal and dendritic alterations remain unclear. We also recognize that altered perception and projection of p-Wave neurons may directly or indirectly contribute to the observed behavioral phenotypes, particularly in response to mechanical stimulation.

Presynaptic varicosities of a-Wave neurons in DFz2 KD animals are indicated by orange arrows in Figure 1. However, no presynaptic markers have been used to confirm actual ectopic synaptic connections. At least the authors should more clearly define what parameters they used to "visually" define potential presynaptic varicosities. Some arrows seem to point to more "globular structures" but for several others, it's unclear what they are pointing at.

As mentioned in our response to Reviewer #1, we have now performed Brp immunostaining to confirm the presence of ectopic synaptic connections (new Figure 2). This analysis supports the interpretation that the presynaptic varicosities observed in DFz2 knockdown animals represent actual synaptic sites. We also clarified in the figure legend the visual criteria used to identify potential presynaptic varicosities.

Reviewing Editor (Recommendations For The Authors):

There are a few major concerns that we recommend the authors address:

(1) Neuroanatomy: The point aberrant synaptic connectivity of a-Wave neurons following Dfz2 knockdown could be substantiated. This could be done by using a presynaptic marker and showing ectopic posterior presynaptic sites ( and/or reduced anterior presynaptic sites) in a-wave neurons.

As mentioned in our response to the public review, we now have used Brp as a presynaptic marker to quantify the number and distribution of presynaptic sites along the normal and ectopic a-Wave axons (new Figure 2). We show that ectopic posterior Wave axons do contain presynaptic sites.  

(2) Gradient calculations: As detailed in the reviews below, the Dfz2 gradient looks like it may be bidirectional. Changing the way the gradient is calculated might help address this point.

As mentioned in our response above, we now have recalculated the gradient by comparing the DFz2 signal to A8 and show that it indeed is bidirectional (new Figure 5—figure supplement 2; formerly Figure 4—figure supplement 2).

(3)  Statistics and sample sizes: As detailed in the reviews, some of the statistical reporting could be improved. Further, increasing sample sizes could help bolster confidence in the data as well.

As mentioned above, we have added a description on the sample size, asterisks, and p-values in Figure 1 – figure supplement 2 legend. We also increased sample sizes of single Wave neurons in control and DFz2 knock-down animals (Figure 1F-J (WT and DFz2 KD) and new Figure 3C-G (WT; formerly Figure 2C-G)).

(4) It would help to include some discussion of the potential contributions of altered p-wave neurons to the observed phenotypes.

As described above, we have added in the Discussion potential contributions of altered p-wave neurons to the observed phenotypes. 

Reviewer #1 (Recommendations For The Authors):

(1) In the current model the authors assume that posterior elongation of a-wave neuron connectivity (axonal projections) induces a loss of connectivity to their natural targets, as backward motion is no longer induced, and a gain of connectivity to posterior wave neuron targets. Is this at the cost of innervation of p-wave neurons, e.g. did these neurons now lose connectivity to their natural targets as well? Therefore, it would be very interesting if the authors would test the behavioral responses to tactile stimuli in the posterior parts of the animal - does the response pattern change?

This is indeed an interesting possibility that p-Wave function is altered upon DFz2 knock-down and hence behavioral response to posterior touch is changed. However, it is technically challenging to test this with tactile stimuli, due to the difficulty of (1) distinguishing between normal and fast-forward locomotion and (2) delivering a posterior touch stimulus while the larva is moving forward, which is the default behavior of the larvae on an agar plate.

As highlighted above, the authors should provide additional evidence that the circuit response to a-wave neurons is changed after a DFz2 knockdown. The authors should monitor the activation wave in response to optogenetic activation of anterior wave neurons - analogous to the data provided in Figure 4 of their 2017 paper. If this response is now switched for a-wave activation but not p-wave activation it would greatly support their claims and this data would be less ambiguous compared to the behavioral locomotion data.

As described in our response to the public review, we attempted this approach but found that the in vitro optogenetics experiment is unfortunately not feasible due to relatively weak expression of R60G09-GAL4 in the larvae. Local activation of control aWave induced fictive backward locomotion only at low frequencies, making comparison with the experimental a-Wave very difficult.  The MB120B-spGAL4 used in our 2017 study could not be employed in this study as it does not drive expression during the embryonic stages and thus cannot be used to knock down DFz2 during development. 

(2) Related to this point. Why would the normal "backward" circuitry of a-wave neurons be functionally suppressed in Dfz2 knockdowns? Do the authors observe reduced synaptic connectivity in these segments? Vesicle clustering of synaptotagmin or other presynaptic markers could be used as a first. As the innervation pattern is only extended by approximately one segment, it is surprising that the changes are so significant.

We agree that these are important and interesting points, which remain to be explored in the future study. As described above, we have performed Brp immunostaining and showed that the posterior ectopic axons of a-Wave do contain synapses (new Figure 2). We also found a slight decrease in the number of synapses in the anterior region, which could partially contribute to the weaker activation of downstream neurons responsible for eliciting backward locomotion. Another possibility is that backward suppression occurs through lateral interaction among downstream circuits. Since forward and backward locomotion do not occur simultaneously, it is likely that the circuits driving these two behaviors are mutually inhibitory. Upon DFz2 knock down in a-Wave, downstream neurons inducing fastforward locomotion may become more strongly activated than those inducing backward locomotion, resulting in inhibition of the latter via a “winner-take-all” mechanism. Since these discussions are highly speculative, we chose not to include them in the revised manuscript.  

(3) The low number of neurons analyzed per segment is of slight concern. This is particularly the case for the control data set used in Figure 1 and Figure 2. As stated, the same datasets are used for both figures. However, at most 6 neurons were analyzed (and for two segments only 3). The control morphology may be more variable than indicated by this data.

As mentioned above, we now have dissected 50 larvae each for the control and experimental groups, obtained seven and six clones respectively, and included these data in the revised manuscript. We apologize that the sample sizes are still relatively small but hope the reviewer understands the inherently low “hit rate” of the stochastic labelling method.

It is somewhat curious that in Figure 1- Supplement 3 the authors report the same number of control clones per segment as in Figure 1/2 - is this simply a coincidence? And if this is an independent dataset why did the author use new controls here but not for Figure 2? It is clear that it is very difficult to generate this data but increasing the n-number beyond 3-6 per segment would significantly increase the confidence in the presented data.

We apologize for the confusion. The data in Figure 1 – figure supplement 3 represent the innervation pattern of dendrites, not axons. We have corrected the figure caption accordingly. These data were obtained from the same samples used to analyze axonal innervation, as shown in the original version of Figure 1F-J.

(2) The name of the RNAi lines should be indicated in Figure 1 and Figure Supplement 3 to facilitate reading - at least the precise names should be given in both figure legends.

We have added these labels in the revised figure legends as requested.

(3) In Figure 4E again the control numbers of Figure 1 for the A2-wave axon are reused. This does not seem appropriate as now a different Gal4 driver is used and a different method to induce individual neuronal clones. Both components may induce significant variability in expression or arborization. As only 3 clones for the wnt4 mutant condition are analyzed (and compared to 5 control clones), this data does not allow for strong conclusions. The authors clearly state the reuse and different methods in the legend of Figure 4 F/G but should also highlight it for the E panel.

Here, we assume that the reviewer is referring to the former Figure 3 (now Figure 4). We have added a note in the legend that the control data, obtained using a different method, were reused in this panel.

(4) The expression levels of DWnt4 and DFz2 were analyzed at the end of embryogenesis. At what developmental stage does the axonal extension of wave neurons take place? Is the gradient maintained throughout the first larval stages?

Based upon the lateral view of Wave neurons in Figure 1—figure supplement 1D, we think that the axonal extension is already established by approximately 20 hr after egg laying. Previously, we performed Wnt4^{MI03717-Trojan-GAL4} > GFP.nls immunostaining in the third instar larva and observed a similar gradient of GFP signals towards the posterior end of the ventral nerve cord (VNC). We have included this data in the revised manuscript (new Figure 5—figure supplement 1).

(5) The authors state that either 2nd or 3rd instar larvae were used for the optogenetic experiments. This may induce unnecessary variation in their assay and should be avoided. As natural variance exists in larvae regarding forward stride duration, the comparison of "on" state forward stride duration between control and experimental genotype is potentially not the best measurement of effect size. What is the difference between OFF and ON stage within the control and experimental genotype? In both cases stride duration decreases but there may not be a significant difference between the delta of the two genotypes. Thus, the observed effect may in part be due to "slower" animals in the control pool. The authors should discuss this more carefully.

We thank the reviewer for bringing up this critical issue. Indeed, the stride durations of larvae between the control and DFz2 knock-down are slightly different in the OFF condition, although this is not statistically significant. In addition, the effect size of Wave activation on mean stride duration is -0.14 (s) in control while -0.21 (s) in DFz2 knock-down, which we interpret as DFz2 knock-down resulting in stronger fastforward locomotion upon Wave activation. We have incorporated this note in the corresponding figure legends (new Figure 6; formerly Figure 5).

(6) While the study clearly provides convincing evidence for their model, the authors should tune down their conclusions in the discussion a little bit and highlight that parts of their discussion are speculative.

We have revised the discussion as suggested.

Reviewer #2 (Recommendations For The Authors):

Albeit the optogenetic behavioral experiments strongly support that the altered axonal projection affect normal locomotion, simultaneous labeling of Wave neurons in DFz2 KD animals with presynaptic markers would strengthen the conclusion of ectopic connection of the extended axon with other circuits.

Please see our response to your public review.

Figure 1 K+L, Figure 2H, I, Figure 3 F+G: many of the individual data points are not visible in the Whisker plot- changing their color would be useful to visualize them better.

We have changed the outline width of the box plots to make the individual data points visible.

Figure 1-Supplement 2: In addition to the comments in the public review- a) the asterisk font size changes in the different panels, e.g. it is much smaller in G', b) font size in some graphs/legends should be increased - in particular in E the hyphenated letters in the genotypes are so small rendering them almost illegible.

We have unified the font size to make them readable in the figure. We thank the reviewer for the suggestions.

eLife Assessment

This valuable paper describes the crystal structure of a complex of the Sld3-Cdc45-binding domain (CBD) with Cdc45, which is essential for the assembly of an active Cdc45-MCM-GINS (CMG) double-hexamer at the replication origin. The structural and biochemical analyses of protein-protein interactions and DNA binding provided solid evidence to support the authors' conclusion. The results shown in the paper are of interest to researchers in DNA replication and genome stability.

Reviewer #1 (Public review):

Summary:

The crystal structure of the Sld3CBD-Cdc45 complex presented by Li et al. is a significant contribution that enhances our understanding of CMG formation during the rate-limiting step of DNA replication initiation. This structure provides crucial insights into the intermediate steps of CMG formation, and the particle analysis and model predictions compellingly describe the mechanism of Cdc45 loading.<br /> Building upon previously known Sld3 and Cdc45 structures, this study offers new perspectives on how Cdc45 is recruited to MCM DH through the Sld3-Sld7 complex. The most notable finding is the structural rearrangement of Sld3CBD upon Cdc45 binding, particularly the α8-helix conformation, which is essential for Cdc45 interaction and may also be relevant to its metazoan counterpart, Treslin. Additionally, the conformational shift in the DHHA1 domain of Cdc45 suggests a potential mechanism for its binding to Mcm2NTD.<br /> Furthermore, the ssDNA-binding experiments involving Sld3 further support a broader functional role in the replication process, beyond its established role in recruiting Cdc45. This adds an intriguing new layer to our understanding of Sld3's activity in the yeast.

Reviewer #2 (Public review):

Summary

The manuscript presents valuable findings, particularly in the crystal structure of the Sld3CBD-Cdc45 interaction and the identification of additional sequences involved in their binding. The modeling of the Sld7-Sld3CBD-CDC45 subcomplex is novel, and the results provide insights into potential conformational changes that occur upon interaction. Although the single-stranded DNA binding data from Sld3 of different species is a minor weakness, the experiments support a model in which the release of Sld3 from the complex may be promoted by its binding to origin single-stranded DNA exposed by the helicase.

Reviewer #3 (Public review):

Summary:

The paper by Li et al. describes the crystal structure of a complex of Sld3-Cdc45-binding domain (CBD) with Cdc45 and a model of the dimer of an Sld3-binding protein, Sld7, with two Sld3-CBD-Cdc45 for the tethering. In addition, the authors showed the genetic analysis of the amino acid substitution of residues of Sld3 in the interface with Cdc45 and biochemical analysis of the protein interaction between Sld3 and Cdc45 as well as DNA binding activity of Sld3 to the single-strand DNAs of the ARS sequence.

Author response:

The following is the authors’ response to the previous reviews

Reviewer #1 (Public review):

Summary:

The crystal structure of the Sld3CBD-Cdc45 complex presented by Li et al. is a significant contribution that enhances our understanding of CMG formation during the rate-limiting step of DNA replication initiation. This structure provides crucial insights into the intermediate steps of CMG formation, and the particle analysis and model predictions compellingly describe the mechanism of Cdc45 loading. Building upon previously known Sld3 and Cdc45 structures, this study offers new perspectives on how Cdc45 is recruited to MCM DH through the Sld3-Sld7 complex. The most notable finding is the structural rearrangement of Sld3CBD upon Cdc45 binding, particularly the α8-helix conformation, which is essential for Cdc45 interaction and may also be relevant to its metazoan counterpart, Treslin. Additionally, the conformational shift in the DHHA1 domain of Cdc45 suggests a potential mechanism for its binding to Mcm2NTD. Furthermore, Sld3's ssDNA-binding experiments provide evidence of its novel functions in the DNA replication process in yeast, expanding our understanding of its role beyond Cdc45 recruitment.

Strengths:

The manuscript is generally well-written, with a precise structural analysis and a solid methodological section that will significantly advance future studies in the field. The predictions based on structural alignments are intriguing and provide a new direction for exploring CMG formation, potentially shaping the future of DNA replication research. This research also opens up several new opportunities to utilize structural biology to unravel the molecular details of the model presented in the paper.

Weaknesses:

The main weakness of the manuscript lies in the lack of detailed structural validation for the proposed Sld3-Sld7-Cdc45 model, and its CMG bound models, which could be done in the future using advanced structural biology techniques such as single particle cryo-electron microscopy. It would also be interesting to explore how Sld7 interacts with the MCM helicase, and this would help to build a detailed long-flexible model of Sld3-Sld7-Cdc45 binding to MCM DH and to show where Sld7 will lie on the structure. This will help us to understand how Sld7 functions in the complex. Also, future experiments would be needed to understand the molecular details of how Sld3 and Sld7 release from CMG is associated with ssARS1 binding.

The proposals based on this study provide new knowledge of the CMG formation process. We agree that our Sld3-Sld7-Cdc45 model will be further confirmed by cryo-EM. We improved our ssARS1-binding assay and quantified data (See the response to Recommendations for the authors of #3 review).

Reviewer #2 (Public review):

Summary

The manuscript presents valuable findings, particularly in the crystal structure of the Sld3CBD-Cdc45 interaction and the identification of additional sequences involved in their binding. The modeling of the Sld7-Sld3CBD-CDC45 subcomplex is novel, and the results provide insights into potential conformational changes that occur upon interaction. Although the single-stranded DNA binding data from Sld3 of different species is a minor weakness, the experiments support a model in which the release of Sld3 from the complex may be promoted by its binding to origin single-stranded DNA exposed by the helicase.

Strengths

The Sld3CBD-Cdc45 structure is a novel contribution, revealing critical residues involved in the interaction.

The model structures generated from the crystal data are well presented and provide valuable insights into the interaction sequences between Sld3 and Cdc45.

The experiments testing the requirements for interaction sequences are thorough and conducted well, with clear figures supporting the conclusions.

The conformational changes observed in Sld3 and Cdc45 upon binding are interesting and enhance our understanding of the interaction.

The modeling of the Sld7-Sld3CBD-CDC45 subcomplex is a new and valuable addition to the field.

The proposed model of Sld3 release from the complex through binding to single stranded DNA at the origin is intriguing.

Weaknesses

The section on the binding of Sld3 complexes to origin single-stranded DNA is somewhat weakened by the use of Sld3 proteins from different species. The comparisons between Sld3-CBD, Sld3CBD-Cdc45, and Sld7-Sld3CBD-Cdc45 involve complexes from different species, limiting the comparisons' value.

Although the study reveals that Sld3 binds to different residues of Cdc45 than those previously shown to bind Mcm or GINS, the data in the paper do not shed any additional light on how GINS and Sld3 binding to Cdc45 or Mcms. would affect each other. Other previous research has suggested that the binding of GINS and Sld3 to Mcm or Cdc45 may be mutually exclusive. The authors acknowledge that a structural investigation of Sld3, Sld7, Cdc45, and MCM during the stage of GINS recruitment will be a significant goal for future research.

We agree that it is better to use all samples from a source; however, due to limitations in protein expression, we used Sld7-Sld3CBD-Cdc45 from a different source. The two sources used in this study belong to the same family, and the proteins Sld7, Sld3 and Cdc45 share sequence conservation with similar structures predicted by Alphafold3 (RMSD = 0.356, 1.392, and 0.891 for Ca atoms of Sld7CTD, Sld7NTD-Sld3NTD, and Sld3CBD-Cdc45). Such similarity in source and proteins allows us to do the comparison. We also mentioned that a cryo-EM study of Sld3-Sld7-Cdc45-MCM and Sld3-Sld7-CMG structures will be a significant goal for future research in our manuscript.

Reviewer #3 (Public review):

Summary:

The paper by Li et al. describes the crystal structure of a complex of Sld3-Cdc45-binding domain (CBD) with Cdc45 and a model of the dimer of an Sld3-binding protein, Sld7, with two Sld3-CBD-Cdc45 for the tethering. In addition, the authors showed the genetic analysis of the amino acid substitution of residues of Sld3 in the interface with Cdc45 and biochemical analysis of the protein interaction between Sld3 and Cdc45 as well as DNA binding activity of Sld3 to the single-strand DNAs of the ARS sequence.

Strengths:

The authors provided a nice model of an intermediate step in the assembly of an active Cdc45-MCM-GINS (CMG) double hexamers at the replication origin, which is mediated by the Sld3-Sld7 complex. The dimer of the Sld3-Sld7 complexes tethers two MCM hexamers together for the recruitment of GINS-Pol epsilon on the replication origin.

Weaknesses:

The biochemical analysis should be carefully evaluated with more quantitative ways to strengthen the authors' conclusion even in the revised version.

In this revision, we improved our ssARS1-binding assay in more quantitative ways (See the response to Recommendations for the authors).

Reviewer #1 (Recommendations for the authors):

I thank the authors for all their replies to my previous questions and for doing all the necessary corrections. I am satisfied with most of their replies, however, upon second reading I have a few more suggestions which could help to improve the manuscript further and make an impact in the field. My comments are listed below.

(1) In general, the manuscript is well structured, but I feel that it requires professional English correction. In many places it was difficult to understand the sentences and I had to read it several times to understand it. Also, very long sentences should be avoided. The flow should be easy to read and understand, and that is why I feel it requires professional English correction.

Following the comment, we checked English carefully and shortened the very long sentences.

(2) Page 5, line 103, please include molecule after the word complex to make it like- "Only one complex molecule exists within an asymmetric unit."

We revised this sentence (P5/L103).

(3) Line 113- more than the N-terminal half of the protruding long helix α7 113 was disordered in the Sld3CBD-Cdc45 complex. This sentence is not clear. What does it mean more than the N-terminal half? Please rewrite it.

We revised this sentence to give the corresponding residue number “(D219–H231)” (P5/L114).

(4) Page 5, result 2- Conformation changes in Sld3CBD and Cdc45 for binding each other, this section may require a little restructuring. Line 130-131- "Therefore, the helix α8CTP seems to be an intrinsically disordered segment when Sld3 alone but 130 folds into a helix coupled to the binding partner Cdc45 in the Sld3CBD-Cdc45 complex." This statement is the crux of the structural finding and therefore, I feel it should move after the first sentence.

Thank you for your comments. We rewrote this part (P5/L128-131).

(5) Line 121-122: Compared to the isolated form (PDBIDs: 5DGO 121 for huCdc45 [31] and 6CC2 for EhCdc45 [33]) and the CMG form (PDBID: 3JC6. Write it in the same format. Make 6CC2 in bracket like other PDB IDs. Restructure this sentence.

We revised this sentence (P5/122-123).

(6) Line 127-129: This sentence is also not very clear.

We revised this sentence together with above No (4). (P5/L128-131)

(7) In my question 4- "Can authors add a supplementary figure showing the probability of disordernes..."., I meant to use a disorder prediction tool like IUPred for the protein sequences and show that α8 is predicted to be a disordered upon sequence analysis. This will help to show the inherent property of α8 helix, and it could add up to the understanding that a disordered region is being structured in the complex structure.

The structures showed that α8CTP is stabilized by binding with Cdc45, but disordered in Sld3CBD alone, indicating that this part is flexible, like an intrinsically disordered segment. We have deposited the structure to PDB, so predictions like IUPred cannot show meaningful information.

(8) Question 9 regarding Supplementary Figure 8- Please include your statement in the figure legend - "WT Sld3CBD was prepared in a complex with Cdc45, while the mutants of Sld3CBD existed alone, we calculated the elements of secondary structure from the crystal structure of Sld3CBD-Cdc45. The concentration of samples was controlled to the same level for CD measurement."

Following the comment, we optimized the figure legend of Supplementary Figure 8.

(9) Question 13- I understand that negative staining and SEC-SAXS experiments could be very tricky for such protein complexes, which have very long loops and are flexible. Did authors try a GraFix cross-linking before doing the negative staining TEM? If it is not being tried, then it might be a good idea to try it and it may help to get much cleaner particles and easier class averaging. Although I completely understand the technical challenges the authors describe and I agree with them, I still feel that one good experiment that shows this dimer model would be very helpful to strengthen the claim. I am concerned because if people start using a similar DLS experiment to calculate intermolecular distances, citing your paper, in many cases it might be a wrong interpretation. In case the negative staining still does not work, at least discuss your technical challenges in the discussion section and mention that SEC-SAXS showed a similar length of the complex and show the Guinier plot and Porod plots in the supplementary data.

We believe that DLS is one of the methods for analyzing the single particle size. Of course, the confirmation by multiple methods will give compelling evidence. Following the comment, we added SEC-SAXS data in the [Results] (P7/L194-196) (Cdc45 recruitment to MCM DH by Sld3 with partner Sld7) and Supplementary Figure 11. The Sld7-Sld3-Cdc45 forms a flexible, long shape. Each binding domain is rigid but linked by the long loops. The flexibility problems are caused by the long loop linkers, but not by binding. So, we did not try to use the cross-linking method for analysis experiments.  

(10) Page 8, line 221- litter sequence specificity: Correct the word "litter" with little. Also, the word shaped is written as sharped at a few places in the manuscript. Please correct it.

We apologize for making such mistakes. We have modified these words.

(11) Page 9, line 237-238: Would it be possible to add a lane showing Sld7 binding to the ssDNA in figure 4. I recommend showing this to understand the ssDNA binding affinity of Sld7 by itself and it will also help us to compare when it is in complex with Sld3.

Considering that Sld7 on CMG is always a complex with Sld3, the ssDNA binding affinity should use the Sld3-Sld7 complex. Additionally, we attempted to overexpress Sld7, but could not obtain the target protein.

Reviewer #2 (Recommendations for the authors):

Thank you for the improved manuscript. The following sentence is unclear: "Cdc45 binds tighter to long ssDNA (>60 bases) with a litter sequence specificity".

We apologize for making such a mistake. We modified “litter” to “little”.

I found it challenging to understand which species were used while reading the results section and figure legends. I recommend that the authors revise the text in both the results and figure legends to clearly indicate when proteins from different species are being compared. Additionally, it would be valuable to explicitly acknowledge this limitation in the text.

Following the comment, we added a description for using different species in results (P8/L224-225) and figure legends (Supplementary Figure 14). We added more information in the Methods to explain why we used two species for preparing proteins.

Reviewer #3 (Recommendations for the authors):

Major points:

(1) The current title is not appropriate for the general readers. At least, DNA replication or DNA replication initiation should be added and abbreviations such as CBD should be avoided.

Following the comment, we added “DNA replication” into the title. Regarding “CBD”, since the full name of “Cdc45 binding domain” is too long, we continue to use Sld3CBD.

(2) As in my previous review, I asked for quantification of the EMSA assay shown in Figure 4 and Supplemental Figures 13 and 14. Since some signals of the bands are very weak, it is hard to conclude something. Given different protein concentrations used in the experiment, the authors should provide any kinds of value. For example, Sld3CBD-CDC45 shows weaker DNA binding than Sld3CBD alone (line 231). Is this true (or reproducible)? It is hard to conclude without any quantification.

We have repeated the EMSA assay four or more times with different rods of overexpression, purification and DNA synthesis, indicating that the EMSA assay is reproducible. In this revision, we changed the DNA stain and adjusted the ratio between the protein and ssDNA with increasing concentrations. The smeared bands of ssDNA with Sld7–Sld3ΔC–Cdc45 or Sld7–Sld3ΔC exhibit enhanced discernibility, and the ssDNA bands are intense enough for grayscale calculations (Figure 4 in the second revised version). We used a series of t-tests to confirm a significantly ssDNA residual level between Sld3CBD–Cdc45 to Sld3CBD, Sld7–Sld3ΔC–Cdc45, and Sld7–Sld3ΔCS (t-test, ****: P<0.0001). We also carefully controlled the sample amount in the EMAS assay and described it in the [Methods].

Moreover, in this EMSA assay (in Figure 4), the authors suggest that the disappearance of ssDNA bands corresponds with the binding of the protein to the DNA. However, it is also possible that the DNA is degraded. It is very important to show the band of protein-DNA complexes on the gel (a whole gel, not the parts of the gel shown in Figure). Why did the authors use this "insensitive" assay using SyberGreen, not radio-labelled ssDNA?

In this revision, we added a negative control of no ssDNA-binding by using ssARS1-3_3 for all protein samples (Sld3CBD, Sld3CBD–Cdc45, Sld7–Sld3ΔC–Cdc45 and Sld7–Sld3ΔC), which were the same rod of expression and purification for bound to ssARS1s (ssARS1-2 and ssARS1-5) (Figure 4), showing that the disappearance of ssDNA bands is caused by binding to proteins, not degradation. Moreover, this time, by changing the DNA stain and increasing the concentration of the samples, the smeared ssDNA bands exhibit enhanced discernibility in the high molecular weight regions when mixed with Sld7–Sld3ΔC–Cdc45 or Sld7–Sld3ΔC, whereas no bands appeared in the NC (ssARS1-3_1). The positions of smeared ssDNA bonds correspond to those of protein in the protein-stain pages, indicating that ssARS1 were complexed with proteins. Following the comment, we show all bands on the gel in Figure 4 and Supplementary Figure 14. Compared to Sld7–Sld3ΔC–Cdc45 or Sld7–Sld3ΔC, Sld3CBD and ssDNA bonds could not be observed because the pI value of Sld3CBD, which affects the entry of the samples into the gel.

We agree that using radio-labelled ssDNA can obtain a sensitive binding assay. However, current laboratory constraints did not allow us to use radio-labelled ssDNA. Furthermore, considering the characteristics of our target proteins, Sld3CBD, Sld3CBD–Cdc45, Sld7–Sld3ΔC–Cdc45, and Sld7–Sld3ΔC, we planned to perform the binding assay in a more natural state without any modifications, labelling or linkers. Additionally, we have attempted to use ITC experiments but failed in the measurements. Presumably, the conformational flexibility of Sld7-Sld3-Cdc45 and Sld7-Sld3 caused a thermodynamic anomaly.

Minor points:

(1) Line 215, 80b: This should be "80 nucleotides(nt)". Throughout the text, nucleotides is better than base to show the length of ssDNAs.

Thank you for your comments. We modified these words throughout the text.

eLife Assessment

This important study provides a description of how single-neuron firing rates in the human medial temporal lobe and frontal cortex are modulated by theta-burst stimulation of the basolateral amydala. The results are supported by convincing evidence obtained from a rigorous task design and analysis of an incredibly rare dataset. The results may help guide future studies incorporating amygdala stimulation to improve patient health.

Reviewer #1 (Public review):

In this manuscript, Campbell et al. assess how intracranial theta-burst stimulation (TBS) applied to the basolateral amygdala in 23 epilepsy patients affects neuronal spiking in the medial temporal lobe and prefrontal cortex during a visual recognition memory task. This is an incredibly rare dataset; collecting single-unit spiking data from behaving humans during active intracranial stimulation is a Herculean task, with immense potential for translational studies of how stimulation may be applied to modulate biological mechanisms of memory. The authors utilize careful, high quality methodology throughout (e.g. task design, spike recording and sorting, statistical analysis), providing high confidence in the validity of their findings.

In providing such a detailed and deep investigation into the single-unit responses to intracranial stimulation the authors provide a very useful resources to any researchers in the fields of brain stimulation and human neurophysiology. This work could be instrumental in guiding diverse research studies, from basic science investigating the role of theta oscillations in human cognition to translational work investigating deep-brain stimulation for memory.

The authors have adequately addressed all prior concerns.

Reviewer #2 (Public review):

Summary:

This study presents a valuable characterization of the effects of intracranial theta-burst stimulation of the basolateral amygdala on single units spiking activity in several areas in the human brain, associated with memory processing. It is written clearly and concisely, allowing readers to fully understand the analysis used.

The authors used a visual recognition memory task previously employed by their group to characterize the effects of basolateral amygdala stimulation upon memory consolidation (Inman et al, 2018). This current report presents an interesting analysis that complements the results reported in the 2018 paper.

Strengths:

Rare combination of human neurophysiology and behavior -<br /> The type of experiment performed in the manuscript, which contains both neurophysiological data, behavior, and a deep brain stimulation intervention (DBS), is incredibly rare, takes many years to accomplish with tight collaboration between clinical and research teams. Our understanding of spiking dynamics of human neurons is very limited, and this report is an important piece in the puzzle that allows DBS to be used in future interventions that will benefit patients' health.

Multiple brain areas included -<br /> It's important to note that the report analyzes brain areas with which the Amygdala has extensive connections (Fig. 1A) - Hippocampus, OFC, Amygdala, ACC. It seems that neurons in all these areas were modulated by the stimulation, except the ACC, in which firing rates were so low that only a handful of neurons were included in the analysis. This is an important demonstration that low-amplitude stimulation (even when reduced to 0.5mA) can travel far and wide across the human brain.

The experiment is cleverly designed to tease apart responses due to visual stimuli (image presentation) and electrical stimulation. Authors suggest that the units modulated by stimulation are largely distinct from those responsive to image offset during trials without stimulation. The subpopulation that responds strongly also tends to have a higher baseline firing rate. It's important to add that the chosen modulation index is more likely to be significant in neurons with higher firing rates (Figure S8). The authors discuss the tradeoff of using a nonparametric modulation index for vs. other methods (for example, percent change in trial-averaged firing rate from baseline).

Author response:

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

Reviewer #1 (Public review):

“This is an exploratory study that doesn't explore quite enough. Critically, the authors make a point of mentioning that neuronal firing properties vary across cell types, but only use baseline firing rate as a proxy metric for cell type. This leaves several important explorations on the table, not limited to the following:”

1a: “Do waveform shape features, which can also be informative of cell type, predict the effect of stimulation?”

To address this question, we modeled our approach to cell type classification after Peyrache et al. 2012. More specifically, we extracted two features from the mean unit waveforms—the valley-to-peak time (VP) and the peak half-width (PHW). These features were then used to classify units into two distinct clusters (k-means, clusters = 2, based on a strong prior from existing literature), representing putative excitatory and inhibitory neurons. Our approach recapitulated many of the same observations in Peyrache et al. 2012, namely (1) identification of two clusters (low PHW/VP: inhibitory, high PHW/VP: excitatory), (2) an ~80/20 ratio of excitatory/inhibitory neurons, and (3) greater baseline firing rates in the inhibitory vs. excitatory neurons. However, we did not observe a preferential modulation of one cell type compared to another (see newly created Figure 4). A description of this analysis and its takeaways has been incorporated into the manuscript.

Change to Text:

Created Figure 4 (Separation of presumed excitatory and inhibitory neurons by waveform morphology).

Caption: (A) Two metrics were calculated using the averaged waveforms for each detected unit: the valley-to-peak width (VP) and peak half-width (PHW). (B) Scatterplot of the relationship between VP and PHW; note that units with identical metrics are overlaid. Using k-means clustering, we identified two distinct response clusters, representing presumed excitatory (E, blue) and inhibitory (I, red) neurons. The units from which the example waveforms were taken are outlined in black. Probability distributions for each metric are shown along the axes. (C) Total number of units within each cluster, separated by region. (D) Comparison of baseline firing rates, separated by cluster. (E) Percent of modulated units in each cluster. * p < 0.05, NS = not significant.

Added a description of clustering methodology to lines 132-137: “We calculated two metrics from the averaged waveform from each detected unit: the valley-to-peak-width (VP) and the peak half-width (PHW) (Figure 4A); previously, these two properties of waveform morphology have been used to discriminate pyramidal cells (excitatory) from interneurons (inhibitory) in human intracranial recordings (Peyrache et al., 2012). Next, we performed k-means clustering (n = 2 clusters) on the waveform metrics, in line with previous approaches to cell type classification.

Added a section in the Results titled “Theta Burst Stimulation Modulates Excitatory and Inhibitory Neurons Equally”. Lines 370-378: “Using k-means clustering, we grouped neurons into two distinct clusters based on waveform morphology, representing neurons that were presumed to be excitatory (E) and inhibitory (I) (Figure 4B). Inhibitory (fast-spiking) neurons exhibited shorter waveform VP and PHW, compared with excitatory (regular-spiking) neurons (I cluster centroid: VP = 0.50ms, PHW = 0.51ms; E cluster centroid: VP = 0.32ms, PHW = 0.31ms), and greater baseline firing rates (U(N_I = 23, N<_E = 133) = 1074.50, p = 0.023) (Figure 4D). Although we observed a much greater proportion of excitatory vs. inhibitory neurons (E: 85.3%, I: 14.7%), stimulation appeared to affect excitatory and inhibitory neurons equally, suggesting that one cell type is not preferentially activated over another (Figure 4E).

Modified discussion of the effects of stimulation on different cell types. Lines 475-483: “…To test these hypotheses directly, we clustered neurons into presumed excitatory and inhibitory neurons based on waveform morphology. In doing so, we observed ~85% excitatory and ~15% inhibitory neurons, which is very similar what has been reported previously in human intracranial recordings (Cowan et al. 2024, Peyrache et al., 2012). Interestingly, stimulation appeared to modulate approximately the same proportion of neurons for each cell type (~30%), despite the differently-sized groups. Recent reports, however, have suggested that the extent to which electrical fields entrain neuronal spiking, particularly with respect to phase-locking, may be specific to distinct classes of cells (Lee et al., 2024).”

1b:  “Is the autocorrelation of spike timing, which can be informative about temporal dynamics, altered by stimulation? This is especially interesting if theta-burst stimulation either entrains theta-rhythmic spiking or is more modulatory of endogenously theta-modulated units.”

The reviewer is correct in suggesting that rate-modulation represents only one of many possible ways by which exogenous theta burst stimulation may influence neuronal activity. Indeed, intracranial theta burst stimulation has previously been shown to evoke theta-frequency oscillatory responses in local field potentials (Solomon et al. 2021), and other forms of stimulation (i.e., transcranial alternating current stimulation) may modulate the rhythm, rather than the rate, of neuronal spiking (Krause et al. 2019).

To investigate whether stimulation altered rhythmicity in neuronal firing, we contrasted the spike timing autocorrelograms, as suggested. More specifically, we computed the pairwise differences in spike timing for each trial, separating spikes into the same pre-, during-, and post-stimulation epochs described in the manuscript (bin size = 5 ms, max lag = 250 ms), grouped neurons by whether they were modulated, and then contrasted the differences in the latencies of the peak normalized autocorrelation value between epochs. Only neurons with a firing rate of ≥ 1 Hz (n = 70/203, 34.5%) were included in this analysis since sparse firing resulted in noisy autocorrelation estimates. Subsequent statistical testing of the peak latency differences between pre-/during- and pre-/post-stimulation did not reveal any group-level differences (Mann-Whitney U tests, p > 0.05). Thus, we were not able to identify neuronal responses suggestive of altered rhythmicity (see Figure S5). A description of this analysis and its takeaways has been incorporated into the manuscript.

Of note, there are two elements of the data that constrain our ability to detect modulation in the rhythm of firing. First, the baseline activity recorded across neurons modulated by stimulation was relatively low (i.e., median firing rate = 1.77 Hz). Second, stimulation often resulted in a suppression, rather than an enhancement, of firing rate. Taken together, the sparse firing afforded limited opportunity to characterize changes to subtle patterns of spiking. 

Change to Text:

Created Figure S5 (Analysis of modulation in spiking rhythmicity)

Caption: (A) Representative autocorrelograms ACG) for a single neuron. The pairwise differences in spike timing were computed for each trial and epoch (bin size = 5 ms, max lag = 250 ms), then smoothed with a Gaussian kernel. The peak in the normalized ACG across trials was computed for each epoch. (B) Kernel density estimate of the peak ACG lag, separated by epoch. (C) The peak ACG lags were split by whether the neuron was modulated (Mod) or unaffected by stimulation (NS = not significant) for each of the two contrasts: pre- vs. during-stim (left) and pre- vs. post-stim (right).

Details about the autocorrelation methodology have been incorporated. Lines 166-172: “To investigate whether stimulation altered rhythmicity in neuronal firing, we analyzed the spike timing autocorrelograms. More specifically, we computed the pairwise differences in spike timing for each trial (bin size = 5 ms, max lag = 250 ms) and then contrasted the differences in the latencies of the peak normalized autocorrelation value between epochs (pre-, during-, post-stimulation). Only neurons with a firing rate of ≥ 1 Hz (n = 70/203, 34.5%) were included in this analysis since sparse firing resulted in noisy autocorrelation estimates.

The results from contrasting the autocorrelograms are now mentioned briefly. Lines 297-298: “Stimulation, however, did not appear to alter the rhythmicity in neuronal firing, as measured by spiking autocorrelograms (Figure S5).”

1c: “The authors reference the relevance of spike-field synchrony (30-55 Hz) in animal work, but ignore it here. Does spike-field synchrony (comparing the image presentation to post-stimulation) change in this frequency range? This does not seem beyond the scope of investigation here.”

We agree that a further characterization of spike-field and spike-phase relationships may provide rich insights into more complex regional and interregional dynamics that may be altered by stimulation. Given that many metrics are biased by sample size (e.g., number of spikes), which can vary considerably, computing the pairwise phase consistency (PPC) between spikes and LFP is a preferred metric (Vinck et al. 2010). Although PPC is unbiased, its variance nonetheless increases considerably with low spike counts; pooling spike counts across trials, however, decouples the temporal relationship between spiking and the LFP phase for each trial, confounding results and yielding an unstable estimate.

To determine whether such an analysis is indeed possible, we calculated the percentage of stimulation trials with ≥ 10 spikes in both the 1s pre- and post-stimulation epochs (a relatively low threshold for inclusion). Only a very small proportion of the total number of trials across all neurons met this criterion (2.5%). Thus, because of the sparse spiking in our data, we are unable to reliably characterize spike-field or spike-phase modulation in detected neurons.

Change to Text:

In the manuscript, we have added a description of why our data is not well-suited to investigate these relationships.

Lines 532-538: “The present study did not investigate interactions between spiking activity and local field potentials because neuronal spiking was sparse at baseline and often further suppressed by stimulation; only a very small proportion of the total number of trials across all neurons exhibited ≥ 10 spikes in both the 1s pre- and post-stimulation epochs (~2.5%). Although certain metrics are not biased by sample size (e.g., pairwise phase consistency), low spike counts can dramatically affect variance and, therefore, result in unstable estimates (Vinck et al., 2011).

1d: “How does multi-unit activity respond to stimulation? At this somewhat low count of neurons (total n=156 included) it would be valuable to provide input on multi-unit responses to stimulation as well.”

We thank the reviewer for this suggestion. We have incorporated an analysis of multiunit activity (MUA), which similarly identifies robust modulation via permutation-based statistical testing and characterizes the different profiles of responses (i.e., increased vs. decreased MUA threshold crossings pre- vs. post-stimulation).

Change to Text:

Created Figure S8 (Analysis of multiunit activity response to stimulation)

Caption: (A) Example trace of multiunit activity (MUA) in one channel during a single stimulation trial. Threshold crossings are highlighted with a pink dot overlaid on the MUA signal with a corresponding hash below. (B) The percentage of channels with significantly modulated MUA, separated by the direction of effect. (C) The percentage of channels with significantly modulated MUA, separated by direction effect and region. Inc (red; post > pre) vs. Dec (blue; post < pre). HIP = hippocampus, OFC = orbitofrontal cortex, AMY = amygdala, ACC = anterior cingulate cortex. *** p < 0.001, NS = not significant.

Details about the MUA methodology have been incorporated. Lines 174-180: “Finally, we measured modulation in multiunit activity (MUA) by filtering the microleectrode signals in a 300-3,000 Hz window and counting the number of threshold crossings. Thresholds were determined on a per-channel basis and defined as -3.5 times the root mean square of the signal during the baseline period; activity during stimulation was excluded since stimulation artifact is difficult to separate from MUA in the absence of spike sorting.

MUA results are now incorporated. Lines 365-367: “Additional characterization of MUA revealed a dominant signature of increased activity post- vs. pre-stimulation, in line with these trends observed at the single-neuron level (Figure S8).”

1e: “Several intracranial studies have implicated proximity to white matter in determining the effects of stimulation on LFPs; do the authors see an effect of white matter proximity here?”

We thank the reviewer for the interesting question. Subsequent characterization revealed only small differences in the proximity of stimulation contacts to white matter (range 1.5-8.0 mm), likely because the chosen target (i.e., basolateral amygdala) has several nearby white matter structures (e.g., stria terminalis). Nonetheless, we performed a linear regression between the proximity to white matter and the stimulation-induced effect on behavior (stimulation vs. no-stimulation d’ difference), the results of which indicate no clear association (p > 0.05; see Figure S9). Critically, this is not to suggest that white matter proximity has no interaction with the reported behavioral effects, but rather, that we could not identify such an association within our data.

Change to Text:

Created Figure S9 (The effect of stimulation proximity to white matter and distance to recorded neurons).

Caption: (A) Kernel density estimate of the Euclidean distance from stimulation contacts to nearest WM structure (in mm); hash marks represent individual observations. (B) The change in memory performance (Δd’) was linearly regressed onto the distance from the stimulated contacts to white matter.

The following has been added to lines 405-426: “Proximity to white matter has been shown to influence the effects of stimulation on behavior and the strength of evoked responses (Mankin et al., 2021; Mohan et al., 2020; Paulk et al., 2022). Across all stimulated contacts, we observed only small differences in the proximity of stimulation contacts to white matter (median = 4.5 mm, range = 1.5-8.0 mm), likely because the chosen target (i.e., basolateral amygdala) has several nearby white matter structures (e.g., stria terminalis). Nonetheless, we performed a linear regression between the proximity to white matter and the stimulation-induced effect on behavior (stimulation vs. no-stimulation d’ difference), the results of which indicate no clear association (p > 0.05; see Figure S9).

Comment 2: “It is a little confusing to interpret stimulation-induced modulation of neuronal spiking in the absence of stimulation-induced change in behavior. How do the authors findings tell us anything about the neural mechanisms of stimulation-modulated memory if memory isn't altered? In line with point #1, I would suggest a deeper dive into behavior (e.g. reaction time? Or focus on individual sessions that do change in Figure 4A?) to make a stronger statement connecting the neural results to behavioral relevance.”

We agree that the connection between the observed stimulation-induced neuronal modulation and effects on behavior is unclear and has proven challenging to elucidate. Per the reviewer’s suggestion, we further focused our analyses on the neuronal modulation effects in the individual sessions that resulted in a robust change in memory performance (stimulation vs. no-stimulation d’ difference threshold of ± 0.5, based on a moderate effect size for Cohen’s d); both a positive and negative threshold were used to capture robust changes in memory performance associated with firing rate modulation, whether enhancement or suppression. To this end, we contrasted the proportion of modulated neurons in the sessions where stimulation resulted in a robust behavioral change (Δd’) with those that did not (~d’). We did not observe a difference in the proportions between groups when collapsed across all sampled regions, or when separately evaluated (Fisher’s exact tests, p > 0.05; see Figure 5C).

Given that this approach did not further clarify the connection between our neural and behavioral results, we believe it is most appropriate to deemphasize claims in the manuscript regarding the potential insights for behavioral modulation (e.g., memory enhancement), and have done so.

Change to Text:

Toned down reference to the memory-related effects of stimulation in the abstract by removing the following lines from the abstract: “Previously, we demonstrated that intracranial theta burst stimulation (TBS) of the basolateral amygdala (BLA) can enhance declarative memory, likely by modulating hippocampal-dependent memory consolidation…” and “…and motivate future neuromodulatory therapies that aim to recapitulate specific patterns of activity implicated in cognition and memory.”

Changed Figure 4 to Figure 5

Created Figure 5C (Interaction between behavioral effects and neuronal modulation)(C)  Change in recognition memory performance was split into two categories using a d’ difference threshold of ± 0.5: responder (positive or negative; Δd’, pink) and non-responder (~d’, grey). Individual d’ scores are shown (left) with points colored by outcome category; dotted lines demarcate category boundaries, and the grey-shaded region represents negligible change. The number of sessions within each outcome category (middle) and the proportion of modulated units as a function of outcome category, separated by region (right). NS = not significant.

The description of the behavioral results has been updated. Lines 394-403: “At the level of individual sessions, we observed enhanced memory (Δd’ > +0.5) in 36.7%, impaired memory (Δd’ < -0.5) in 20.0%, and negligible change (-0.5 ≤ Δd’ ≤ 0.5) in 43.3% when comparing performance between the stim and no-stim conditions; a threshold of Δd’ ± 0.5 was chosen for this classification based on the defined range of a “medium effect” for Cohen’s d. To test our hypothesis that neuronal modulation would be associated with changes in memory performance, we combined the sessions that resulted in either memory enhancement or impairment and contrasted the proportion of modulated units across regions sampled. We did not, however, observe a meaningful difference in the proportion of modulated units when grouped by behavioral outcome (all contrasts p > 0.05) (Figure 5C).

Lines 213-214 and 394-397 have been edited to reflect a change in the d’ threshold used for categorizing behavioral results (from Δd’ ± 0.2 to Δd’ ± 0.5).

Comment 3: “It is not clear to me why the assessment of firing rates after image onset and after stim offset is limited to one second - this choice should be more theoretically justified, particularly for regions that spike as sparsely as these.”

We thank the reviewer for this question and acknowledge that no clear justification was provided for this decision in the manuscript. Our decision to limit each of the analysis epochs to 1s was chosen for two reasons. First, the maximum possible length of the during-stimulation epoch was 1 s (stim on for 1 s). Although the pre- and post-stimulation epochs could be extended without issue, we were concerned that variable time windows could introduce a bias, for instance, resulting in different variances between epochs. Second, we anticipated, both from empirical observations and prior literature, that the neural response following stimulation or task features (e.g., image onset/offset) was likely to be transient, rather than sustained for a period of many seconds. By keeping the windows short, we ensured that our approach to detecting modulation (i.e., contrasting trial-wise spike counts between each pair of epochs) captured the intended effect rather than random noise. We have incorporated a discussion of this rationale in the Peri-Stimulation Modulation Analyses section.

Change to Text:

Lines 156-158 have been added: “Each epoch was constrained to 1 s to ensure that subsequent firing rate contrasts were unbiased and to capture potential transient effects (e.g., image onset/offset).”

Comment 4: “This work coincides with another example of human intracranial stimulation investigating the effect on firing rates (doi: https://doi.org/10.1101/2024.11.28.625915). Given how incredibly rare this type of work is, I think the authors should discuss how their work converges with this work (or doesn't).”

Thank you for bringing this highly relevant work to our attention. We were unaware of this recent preprint and have incorporated a discussion of its main findings into the manuscript.

Change to Text:

New citations: van der Plas et al. 2024 (bioRxiv), Cowan et al. 2024 (bioRxiv)

The discussion of related studies has been updated. Lines 447-457: “Few studies, however, have characterized the impact of electrical stimulation via macroelectrodes on the spiking activity of human cortical neurons, none of which involve intracranial theta burst stimulation. One study reported a long-lasting reduction in neural excitability among parietal neurons, with variable onset time and recovery following continuous transcranial TBS in non-human primates (Romero et al., 2022). In a similar vein, it was recently shown that human neurons are largely suppressed by single-pulse electrical stimulation (Cowan et al., 2024; Plas et al., 2024). Other emerging evidence suggests that transcranial direct current stimulation may entrain the rhythm rather than rate of neuronal spiking (Krause et al., 2019) and that stimulation-evoked modulation of spiking may meaningfully impact behavioral performance on cognitive tasks (Fehring et al., 2024).”

Comment 5: “What information does the pseudo-population analysis add? It's not totally clear to me.”

We recognize the need to further contextualize the motivation for the exploratory pseudo-population analysis and appreciate the reviewer for bringing the lack of detail to our attention. In brief, the analysis allowed us to observe trends in activity across populations of neurons, which, in principle, are not visible by characterizing modulation solely in discrete neurons. Additional details have been incorporated into the manuscript, as suggested.

Change to Text:

Additional justification has been incorporated in the description of the methodology. Lines 185-187: “…This approach enables the identification of dominant patterns of coordinated neural activity that may not be apparent when examining individual neurons in isolation.”, lines 192-194: “…By collapsing across subjects into a common pseudo-population, this analysis provides a mesoscale view of how stimulation modulates shared activity patterns across anatomically distributed neural populations.”

A summary interpretation has been added to the paragraph describing the results. Lines 326-328: “Taken together, these analyses reveal global structure in the state space of responses to BLA stimulation within hippocampal circuits.”

Reviewer #2 (Public review):

Comment 1 “Authors suggest that the units modulated by stimulation are largely distinct from those responsive to image offset during trials without stimulation. The subpopulation that responds strongly also tends to have a higher baseline of firing rate. It's important to add that the chosen modulation index is more likely to be significant in neurons with higher firing rates.”

This is an important point that was not previously addressed in our manuscript. We suspect there are likely two factors at play worth considering with respect to our chosen nonparametric modulation index: neurons with lower activity require smaller changes in spike counts to be significantly modulated (easier to flip ranks), and neurons with higher activity empirically exhibit greater absolute shifts in the number of spikes. Our further use of permutation testing, while mitigating false positives, may also somewhat constrain the ability to detect modulation in sparsely active neurons. Nonetheless, given that many trials entailed few or no spikes, we believe this approach is preferable to alternatives that may be more susceptible to noise (e.g., percent change in trial-averaged firing rate from baseline).

To better understand the tradeoffs with detection probability, we performed a sensitivity analysis. We generated synthetic data with different baseline firing rates (0.1-5.0 Hz) and effect sizes (± 0.1-0.7 Hz) and simulated the likelihood of detection with our given modulation index across neurons. The results of the simulation support the notion that the probability of detecting modulation is lower for sparsely active neurons (Figure S8C). Further discussion of this consideration for the chosen modulation index, as well as details regarding the sensitivity analysis, have been incorporated into the manuscript.

Change to Text:

Created Figure S7C (Detection probability analysis)

Caption: The same permutation-based analyses reported in the manuscript were repeated under different control conditions… (C) Visualization of the predicted probability of detecting modulation across synthetic neurons with variable firing rates and modulation effect sizes; FR = firing rate.

Lines 223-224 have been added to the Methods section titled “Firing Rate Control Analyses”: “We performed a series of control analyses to test whether our approach to firing rate detection was robust…”

A description of the simulation has been incorporated into the same section as above. Lines 234-237: “Finally, to better understand the tradeoffs with our statistical approach, we generated synthetic data with different baseline firing rates (0.1-5.0 Hz) and effect sizes (± 0.1-0.7 Hz), then simulated the likelihood of detecting modulation across variable conditions (Figure S7C).”

The description of the results from the control analyses has been updated. Lines 330-339: “Finally, we performed three supplementary analyses to evaluate the robustness of our approach to detecting firing rate modulation: a sensitivity analysis assessing the proportion of modulated units at different firing rate thresholds for inclusion/exclusion, a data dropout analysis designed to control for the possibility that non-physiological stimulation artifacts may preclude the detection of temporally adjacent spiking, and a synthetic detection probability analysis. These results recapitulate our observation that units with higher baseline firing are most likely to exhibit modulation (though the probability of detecting modulation is lower for sparsely active neurons) and suggest that suppression in firing rate is not solely attributable to amplifier saturation following stimulation (Figure S7).

Comment 2: “Readers can benefit from understanding with more details the locations chosen for stimulation - in light of previous studies that found differences between effects based on proximity to white matter (For example - PMID 32446925, Mohan et al, Brain Stimul. 2020 and PMID 33279717 Mankin et al Brain Stimul. 2021).”

This has been addressed in the above response to Reviewer’s 1 comment 1.1e.

Change to Text:

See changes related to Reviewer 1 comment 1.1e.

Comment 3: “Missing information in the manuscript…”

3a: “Images of stimulation anatomical locations for all subjects included in this study. Ideally information about the impedance of the contacts to be able to calculate the actual current used.”

As requested, we have provided an image from the coronal T1 MRI sequence, which highlights the position of the stimulated contacts for each of the 16 patients. Though we did not measure the impedances directly, the stimulation was current-controlled, which ensured that the desired current and charge density were consistent regardless of the tissue or electrode impedance.

Change to Text:

Created Figure S1 (Anatomical location of stimulated electrodes).

Caption: A coronal slice from the T1-weighted MRI scan is shown for each patient who participated in the study (n = 16). Electrode contacts within the same plane of the image are shown with blue circles, and the bipolar pair of stimulated contacts within the basolateral amygdala is highlighted in red.

Lines 144-145 have been edited to reflect that the delivered stimulation was current-controlled: “Specifically, we administered current-controlled, charge-balanced, …”

3b: “The studied population is epilepsy patients, and the manuscript lacks description of their condition, proximity to electrodes included in the study to pathological areas, and the number of units from each patient/hemisphere.”

We agree that additional information regarding patient demographics, experimental details, and clinical characteristics would further contextualize this unique patient population. A new table has been included, which contains the following information: patient ID, sex, age, # experimental session, # SEEG leads (and # microelectrodes), # detected units (L vs. R hemisphere), and suspected seizure onset zone.

Change to Text:

Created Table S1 (Patient demographics and clinical characteristics).

Lines 258-259 have been added: “…(see Table S1 for patient demographics).”

3c: “I haven't seen any comments on code availability (calculating modulation indices and statistics) and data sharing.”

For clarification, a section titled Resource Availability is already appended to the end of the manuscript following the Conclusion, which describes the data and code availability.

Change to Text:

None

3d: “Small comment - Figure legend 3E - Define gray markers (non-modulated units?)”

Thank you for highlighting this omission. We have updated the relevant figure caption.

Change to Text:

The following has been added to the Figure 3 caption: “…whereas units without a significant change in activity are shown in grey.”

eLife Assessment

This study presents an important discovery regarding the diversity and evolution of gall-forming microbial effectors. Supported by convincing computational structural predictions and analyses, the research provides insights into the unique mechanisms by which gall-forming microbes exert their pathogenicity in plants. This study also offers guidance that is of value for future studies on pathogen effector function and co-evolution with host plants.

Reviewer #1 (Public review):

Summary:

This manuscript presents a comprehensive structure-guided secretome analysis of gall-forming microbes, providing valuable insights into effector diversity and evolution. The authors have employed AlphaFold2 to predict the 3D structures of the secretome from selected pathogens and conducted a thorough comparative analysis to elucidate commonalities and unique features of effectors among these phytopathogens.

Strengths:

The discovery of conserved motifs such as 'CCG' and 'RAYH' and their central role in maintaining the overall fold is an insightful finding. Additionally, the discovery of a nucleoside hydrolase-like fold conserved among various gall-forming microbes is interesting.

Weaknesses:

Important conclusions are not verified by experiments.

Comments on revisions: I acknowledge the authors' revision efforts.

Reviewer #2 (Public review):

Summary:

Soham Mukhopadhyay et al. investigated the protein folding of the secretome from gall-forming microbes using the AI-based structure-modeling tool AlphaFold2. Their study analyzed six gall-forming species, including two Plasmodiophorid species and four others spanning different kingdoms, along with one non-gall-forming Plasmodiophorid species, Polymyxa betae. The authors found no effector fold specifically conserved among gall-forming pathogens, leading to the conclusion that their virulence strategies are likely achieved through diverse mechanisms. However, they identified an expansion of the Ankyrin repeat family in two gall-forming Plasmodiophorid species, with a less pronounced presence in the non-gall-forming Polymyxa betae. Additionally, the study revealed that known effectors such as CCG and AvrSen1 belong to sequence-unrelated but structurally similar (SUSS) effector clusters.

Strengths:

(1) The bioinformatics analyses presented in this study are robust, and the AlphaFold2-derived resources deposited in Zenodo provide valuable resources for researchers studying plant-microbe interactions. The manuscript is also logically organized and easy to follow.

(2) The inclusion of the non-gall-forming Polymyxa betae strengthens the conclusion that no effector fold is specifically conserved in gall-forming pathogens and highlights the specific expansion of the Ankyrin repeat family in gall-forming Plasmodiophorids.

(3) Figure 4a and 4b effectively illustrate the SUSS effector clusters, providing a clear visual representation of this finding.

(4) Figure 1 is a well-designed, comprehensive summary of the number and functional annotations of putative secretomes in gall-forming pathogens. Notably, it reveals that more than half of the analyzed effectors lack known protein domains in some pathogens, yet some were annotated based on their predicted structures, despite the absence of domain annotations.

Weaknesses:

(1) The effector families discussed in this paper remain hypothetical in terms of their functional roles, which is understandable given the challenges of demonstrating their functions experimentally. However, this highlights the need for experimental validation as a next step.

Authors' response: Thank you. Yes, there is a lot of work to do in the coming years.

Reviewer's response: Incorporating experimental validation substantially strengthened the manuscript. Did you try the AlphaFold-Multimer prediction of the interaction between PBTT_00818 and the GroES-like protein? Does the model indicate a high-confidence interface?

(2) Some analyses, such as those in Figure 4e, emphasize motifs derived from sequence alignments of SUSS effector clusters. Since these effectors are sequence-unrelated, sequence alignments might be unreliable. It would be more rigorous to perform structure-based alignments in addition to sequence-based ones for motif confirmation. For instance, methods described in Figure 3E of de Guillen et al. (2015, https://doi.org/10.1371/journal.ppat.1005228) or tools like Foldseek could be useful for aligning structures of multiple sequences.

Authors' response: In Fig. 4e, we highlight the conserved cysteine residues. While there is no clearly conserved overall motif, the figure illustrates that despite the high sequence divergence, the key cysteines involved in disulfide-bridge formation are consistently conserved across the sequences.

Reviewer's response: Understood. Nevertheless, if a reliable sequence alignment can indeed be generated, I would interpret this to mean that the CCG effectors constitute a highly diversified family rather than being truly sequence unrelated. By comparison, members of the MAX effector family share a common fold, yet their sequences are so divergent that sequence alignment is impossible.

(3) When presenting AlphaFold-generated structures, it is essential to include confidence scores such as pLDDT and PAE. For example, in Figure 1D of Derbyshire and Raffaele (2023, https://doi.org/10.1038/s41467-023-40949-9), the structural representations were colored red due to their high pLDDT scores, emphasizing their reliability.

Authors' response: Thank you for the observation. Due to the restrictive parameters used in our analysis, over 90 % of the structure would appear red. For this reason, we chose not to include the color scale, as it would not provide additional informative value in this context.

Reviewer's response: Understood.

Author response:

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

Reviewer #1 (Public review):

Summary:

This manuscript presents a comprehensive structure-guided secretome analysis of gall-forming microbes, providing valuable insights into effector diversity and evolution. The authors have employed AlphaFold2 to predict the 3D structures of the secretome from selected pathogens and conducted a thorough comparative analysis to elucidate commonalities and unique features of effectors among these phytopathogens.

Strengths:

The discovery of conserved motifs such as 'CCG' and 'RAYH' and their central role in maintaining the overall fold is an insightful finding. Additionally, the discovery of a nucleoside hydrolase-like fold conserved among various gall-forming microbes is interesting.

Weaknesses:

Important conclusions are not verified by experiments.

Thank you very much. There are many aspects of this study that could be further validated, each potentially requiring years of work. Therefore, we chose to focus on two specific hypotheses: are AlphaFol-Multimer predictions accurate? Can ANK target more than one host protein? Particularly, we focused on the identification of putative targets for one of the ankyrin repeat proteins, PBTT_00818 (Fig. 6). Using one-by-one yeast two-hybrid (Y2H) assays, we tested the AlphaFold-Multimer prediction of an interaction between PBTT_00818 and MPK3. The interaction did not occur in yeast, suggesting it might not take place under those conditions.

This negative result led us to perform a Y2H screen using an Arabidopsis cDNA library, which identified a GroES-like protein, highly expressed in roots, as a potential target of the ANK effector. Surprisingly, both the PBTT_00818–MPK3 and PBTT_00818–GroES-like protein interactions were later confirmed in planta using BiFC assays. These findings suggest two key points: (1) AlphaFold predictions can be accurate for ANK proteins, and (2) ANK domains, known for mediating protein-protein interactions, may enable these effectors to target multiple host proteins.

Although the precise biological implications remain unclear, it is possible that ANK proteins act as scaffolds or adaptors for other effectors during infection. The validations presented here open exciting avenues for further research into the role of ANK proteins in Plasmodiophorid pathogenesis and gall formation. This is presented in the corrected preprint and Fig. 7, Table S12, Fig. S7-S8.

Reviewer #2 (Public review):

Summary:

Soham Mukhopadhyay et al. investigated the protein folding of the secretome from gall-forming microbes using the AI-based structure modeling tool AlphaFold2. Their study analyzed six gall-forming species, including two Plasmodiophorid species and four others spanning different kingdoms, along with one non-gall-forming Plasmodiophorid species, Polymyxa betae. The authors found no effector fold specifically conserved among gall-forming pathogens, leading to the conclusion that their virulence strategies are likely achieved through diverse mechanisms. However, they identified an expansion of the Ankyrin repeat family in two gall-forming Plasmodiophorid species, with a less pronounced presence in the non-gall-forming Polymyxa betae. Additionally, the study revealed that known effectors such as CCG and AvrSen1 belong to sequence-unrelated but structurally similar (SUSS) effector clusters.

Strengths:

(1) The bioinformatics analyses presented in this study are robust, and the AlphaFold2-derived resources deposited in Zenodo provide valuable resources for researchers studying plant-microbe interactions. The manuscript is also logically organized and easy to follow.

(2) The inclusion of the non-gall-forming Polymyxa betae strengthens the conclusion that no effector fold is specifically conserved in gall-forming pathogens and highlights the specific expansion of the Ankyrin repeat family in gall-forming Plasmodiophorids.

(3) Figure 4a and 4b effectively illustrate the SUSS effector clusters, providing a clear visual representation of this finding.

(4) Figure 1 is a well-designed, comprehensive summary of the number and functional annotations of putative secretomes in gall-forming pathogens. Notably, it reveals that more than half of the analyzed effectors lack known protein domains in some pathogens, yet some were annotated based on their predicted structures, despite the absence of domain annotations.

Weaknesses:

(1) The effector families discussed in this paper remain hypothetical in terms of their functional roles, which is understandable given the challenges of demonstrating their functions experimentally. However, this highlights the need for experimental validation as a next step.

Thank you. Yes, there is a lot of work to do in the coming years.

(2) Some analyses, such as those in Figure 4e, emphasize motifs derived from sequence alignments of SUSS effector clusters. Since these effectors are sequence-unrelated, sequence alignments might be unreliable. It would be more rigorous to perform structure-based alignments in addition to sequence-based ones for motif confirmation. For instance, methods described in Figure 3E of de Guillen et al. (2015, https://doi.org/10.1371/journal.ppat.1005228) or tools like Foldseek could be useful for aligning structures of multiple sequences.

In Fig. 4e, we highlight the conserved cysteine residues. While there is no clearly conserved overall motif, the figure illustrates that despite the high sequence divergence, the key cysteines involved in disulfide bridge formation are consistently conserved across the sequences.

(3) When presenting AlphaFold-generated structures, it is essential to include confidence scores such as pLDDT and PAE. For example, in Figure 1D of Derbyshire and Raffaele (2023, https://doi.org/10.1038/s41467-023-40949-9), the structural representations were colored red due to their high pLDDT scores, emphasizing their reliability.

Thank you for the observation. Due to the restrictive parameters used in our analysis, over 90% of the structure would appear red. For this reason, we chose not to include the color scale, as it would not provide additional informative value in this context.

Reviewer #1 (Recommendations for the authors):

Experimental validation of the significance of 'CCG' and 'RAYH' motifs would further strengthen this study.

Regarding the Mig1-like protein in Ustilago maydis, the presence of four conserved cysteine residues that are pivotal for maintaining the stability of its folded structure raises an intriguing question. Specifically, while many Mig cluster effectors contain four cysteine residues that form two conserved disulfide bridges, this structure is notably absent in the Mig protein itself. The author has speculated that these four cysteine residues form two conserved disulfide bonds, which are crucial for the stability of Mig protein folding. However, this hypothesis remains unvalidated. To test this prediction, it would be prudent to simulate mutations in the cysteine residues corresponding to the disulfide bonds in Mig and employ molecular dynamics simulations to assess the stability of folding before and after the mutation.

Mig-1 does contain the four conserved cysteine residues responsible for forming disulfide bridges. However, due to the high divergence among Mig-1-like sequences, the alignment software was unable to properly align all the cysteine residues. As a result, Mig-1 may appear to lack these conserved cysteines in the alignment, although they are indeed present upon individual inspection. This is an area that research groups working with U. maidis as a model could explore further to expand our understanding of this effector family.

Could you please clarify why talking about Ankyrins and LRR in Arabidopsis thaliana (line 252)? Additionally, what are the structural and functional differences between the LRR sequences of P. brassicae and those of the host plants?

This sentence refers to the identification of the ANK motif in P. brassicae and S. spongospora, not in Arabidopsis thaliana. While the hydrophobic core of the ANK domains appears conserved between the host and the pathogen, the surface residues are highly polymorphic.

The evidence supporting the interaction between the ANK effector and Arabidopsis immunity-related proteins, as validated using AlphaFold-Multimer, is currently limited. To enhance the reliability of these data, it is advisable for the author to select several pairs of proteins predicted to interact for further experimental verification.

We conducted a large-scale yeast two-hybrid (Y2H) screen using the ANK domain effector PBTT_00818, which was selected due to its high iPTM+pTM score. The Y2H interactions were subsequently validated through BiFC assays. Our results show that PBTT_00818 interacts with Arabidopsis MPK3 in the nucleus, consistent with predictions from the AlphaFold2-multimer model. In addition, PBTT_00818 was also found to target AT3G56460, a GroES-like zinc-binding alcohol dehydrogenase, also localized in the nucleus.

While the manuscript is well-composed, certain sections could be enhanced for clarity and readability. For example, the discussion section could be expanded to include a more in-depth analysis of the implications of the findings for understanding the virulence mechanisms of gall-forming microbes. Additionally, a comparison of the findings with previous studies on related pathogens would provide a more comprehensive perspective.

Certain sections of the discussion have been expanded. However, we chose to focus on the novel aspects of the study and to avoid comparisons with other plant pathogens, as those mechanisms are already well known and extensively studied. Studies using AlphaFold in plant pathology are also limited.

*Reviewer #2 (Recommendations for the authors):*

The results of clustering analyses are highly dependent on the chosen thresholds. Given that the authors provide clear and well-designed visualizations of SUSS effectors in Figures 4a and 4b, applying the same presentation methods to Figures 5a and 5b could make these analyses more convincing.

We were able to generate the all-vs-all matrix for Figures 4a and 4b because it involved only 13 proteins. However, Figure 5b includes over 40 effectors, making it impractical to visualize the data in the same way. Instead, we presented the sequence-based clusters as nodes and connected them based on structural similarity.

eLife Assessment

This valuable study presents computational analyses of over 5,000 predicted extant and ancestral nitrogenase structures. The data analyses are convincing, it offers unique insights into the relationship between structural evolution and environmental and biological phenotypes. The data generated in this study provide a vast resource that can serve as a starting point for studies of reconstructed and extant nitrogenases.

Reviewer #1 (Public review):

This was a clearly written manuscript that did an excellent job summarizing complex data. In this manuscript, Cuevas-Zuviría et al. use protein modeling to generate over 5,000 predicted structures of nitrogenase components, encompassing both extant and ancestral forms across different clades. The study highlights that key insertions define the various Nif groups. The authors also examined the structures of three ancestral nitrogenase variants that had been previously identified and experimentally tested. These ancestral forms were shown in earlier studies to exhibit reduced activity in Azotobacter vinelandii, a model diazotroph.

Reviewer #2 (Public review):

Summary:

This work aims to study the evolution of nitrogenanses, understanding how their structure and function adapted to changes in environment, including oxygen levels and changes in metal availability.

The study predicts > 3000 structures of nitrogenases, corresponding to extant, ancestral and alternative ancestral sequences. It is observed that structural variations in the nitrogenases correlate with phylogenetic relationships. The amount of data generated in this study represents a massive and admirable undertaking. The study also provides strong insight into how structural evolution correlates with environmental and biological phenotypes.

Author response:

The following is the authors’ response to the previous reviews

Reviewer #1 (Public review):

Comments on revisions:

I appreciate the authors responding to my comments. I think Fig. S10 helps put the structural data into more context. It would be helpful to make clearer in the legend what proteins are being compared, especially in 10C.

Although I can see why the authors focus on the NifK extension and its potential connection to oxygen protection, I would point out that Vnf and Anf do not have this extension in their K subunit, and you find both Vnf and Anf in aerobic and facultative anaerobic diazotrophs. This is a minor point, but I think it is important to mention in the discussion.

We thank the reviewer for their thoughtful comments. We now added an additional line to the Discussion following their recommendation and moved Figure S10 to main text.

Reviewer #2 (Public review):

Summary: 

This work aims to study the evolution of nitrogenanses, understanding how their structure and function adapted to changes in environment, including oxygen levels and changes in metal availability. 

The study predicts > 3000 structures of nitrogenases, corresponding to extant, ancestral and alternative ancestral sequences. It is observed that structural variations in the nitrogenases correlate with phylogenetic relationships. The amount of data generated in this study represents a massive and admirable undertaking. The study also provides strong insight into how structural evolution correlates with environmental and biological phenotypes. 

We thank the reviewer for their summary and positive appraisal.

eLife Assessment

This study presents a valuable finding on the molecular mechanisms that govern GABAergic inhibitory synapse function. The authors propose that Endophilin A1 serves as a novel regulator of GABAergic synapses by acting as a component of the inhibitory postsynaptic density. The findings are convincing and likely to interest a broad audience of scientists focusing on inhibitory synaptic transmission, the excitation-inhibition balance, and its disruption in disorders such as epilepsy.

Reviewer #1 (Public review):

Summary:

In the present study, Chen et al. investigate the role of Endophilin A1 in regulating GABAergic synapse formation and function. To this end, the authors use constitutive or conditional knockout of Endophilin A1 (EEN1) to assess the consequences on GABAergic synapse composition and function, as well as the outcome for PTZ-induced seizure susceptibility. The authors show that EEN1 KO mice show a higher susceptibility to PTZ-induced seizures, accompanied by a reduction in the GABAergic synaptic scaffolding protein gephyrin as well as specific GABAAR subunits and eIPSCs. The authors then investigate the underlying mechanisms, demonstrating that Endophilin A1 binds directly to gephyrin and GABAAR subunits, and identifying the subdomains of Endophilin A1 that contribute to this effect. Overall, the authors state that their study places Endophilin A1 as a new regulator of GABAergic synapse function.

Strengths:

Overall, the topic of this manuscript is very timely, since there has been substantial recent interest in describing the mechanisms governing inhibitory synaptic transmission at GABAergic synapses. The study will therefore be of interest to a wide audience of neuroscientists studying synaptic transmission and its role in disease. The manuscript is well written and contains a substantial quantity of data. In the revised version of the manuscript, the authors have increased the number of samples analyzed and have significantly improved the statistical analysis, thereby substantially strengthening the conclusions of their study.

Reviewer #2 (Public review):

Summary:

The function of neural circuits relies heavily on the balance of excitatory and inhibitory inputs. Particularly, inhibitory inputs are understudied when compared to their excitatory counterparts due to the diversity of inhibitory neurons, their synaptic molecular heterogeneity, and their elusive signature. Thus, insights into these aspects of inhibitory inputs can inform us largely on the functions of neural circuits and the brain.

Endophilin A1, an endocytic protein heavily expressed in neurons, has been implicated in numerous pre- and postsynaptic functions, however largely at excitatory synapses. Thus, whether this crucial protein plays any role in inhibitory synapse, and whether this regulates functions at the synaptic, circuit, or brain level remains to be determined.

The three remaining concerns are:

(1) The use of one-way ANOVA is not well justified.

(2) The use of superplots to show culture to culture variability would make it more transparent.

(3) Change EEN1 in Figure 8B to EndoA1.

Comments on revised version:

The authors addressed the concerns adequately.

Reviewer #3 (Public review):

Chen et al. identify endophilin A1 as a novel component of the inhibitory postsynaptic scaffold. Their data show impaired evoked inhibitory synaptic transmission in CA1 neurons of mice lacking endophilin A1, and an increased susceptibility to seizures. Endophilin can interact with the postsynaptic scaffold protein gephyrin and promotes assembly of the inhibitory postsynaptic element. Endophilin A1 is known to play a role in presynaptic terminals and in dendritic spines, but a role for endophilin A1 at inhibitory postsynaptic densities has not yet been described, providing a valuable addition to the field.

To investigate the role of endophilin A1 at inhibitory postsynapses, the authors used a broad array of experimental approaches, including tests of seizure susceptibility, electrophysiology, biochemistry, neuronal culture and image analysis. The authors have addressed the remaining concerns in their revision. Taken together, their results expand the synaptic role of endophilin-A1 to include the inhibitory post synaptic element.

Author response:

The following is the authors’ response to the previous reviews

Reviewer #2 (Recommendations for the authors):

Comments on revised version:

The authors addressed the concerns adequately. The three remaining concerns are:

(1) The use of one-way ANOVA is not well justified.

The statement about statistical test in “Statistical analysis” section is as follows in the revised manuscript, “Data sets were tested for normality and direct comparisons between two groups were made using two-tailed Student’s t test (t test, for normally distributed data) as indicated. To evaluate statistical significance of three or more groups of samples, one-way ANOVA analysis with a Tukey test was used or repeated measures ANOVA analysis with a Tukey test was used in behavior assays. Statistical parameters are reported in the figures and the corresponding legends”.

We used a one-way ANOVA for the data about one categorical independent variable and one quantitative dependent variable. The independent variable should have at least three different groups or categories. And we conducted repeated measures ANOVA analysis for the data about behavioral tests according to the suggestion by Reviewer #1 (Point 18) in revised manuscript.

(2) The use of superplots to show culture to culture variability would make it more transparent.

Thanks for the nice suggestion. While superplots could more transparently show culture to culture variability, it is difficult to add more colors or even shades to the scatterplots in the current form, which have already been color coded for multiple groups of samples. The scatterplots we used effectively illustrate the variability across all collected data and do not affect the conclusions of our study. Therefore, we prefer not to change the way of data presentation in the revised manuscript.

(3) Change EEN1 in Figure 8B to EndoA1.

Thanks a lot for the sharp eye. Corrected.

Reviewer #3 (Recommendations for the authors):

Specific comments:

The authors have made a substantial effort to improve their manuscript. A number of issues, related to numbers of observations mentioned by the reviewers, are clarified in the revised manuscript. The authors have also clarified some of the other questions from the reviewers. The long list of issues brought up by the reviewers and the many corrections needed still raise questions about data quality in this manuscript.

In response to my comments (Point 2), the added experiment with PSD95.FingR and GPN.FingR in cultured neurons (Fig. S5A-D) is a good addition; the in vivo data using FingRs in Figure S3 look less convincing however. In response to my Point 5, the authors have added a cell-free binding assay (Figure 5I). This is a useful addition, but to convincingly make the point of interaction between Gephyrin and EndoA1, more rigorous biophysical quantitation of binding is needed. The legend in Figure 5I states that 4 independent experiments were performed, but the graph only shows 3 dots. This needs to be corrected.

We sincerely appreciate your comments and apologize for any concerns raised. As suggested (Point 2), we made many efforts to visualize endogenous postsynaptic proteins using recombinant probes. However, due to much lower expression of GPN.FingR compared with PSD95.FingR in P21 brain slices following viral infection (Figure S3), we were unable to obtain better imaging results. To strengthen our data and conclusions, we additionally performed experiments with PSD95.FingR and GPN.FingR in cultured neurons (Fig. S5A-D) in the revised manuscript.

Regarding the biophysical quantification of gephyrin–endophilin A1 binding, we do not have the equipment for this type of experiment (surface plasmon resonance or isothermal titration calorimetry). Instead, we performed a pull-down assay as an alternative to confirm their interaction (Figure 5I). We also apologize for the error in the number of independent experiments stated in the figure legend and have corrected it in the revised manuscript.

eLife Assessment

This fundamental study characterizes the mechanics and stability of bolalipids from archaeal membranes using a minimalist, physics-based computational model. The authors present a robust mesoscale model of bolalipids-containing membranes, systematically evaluating it across diverse membrane configurations. The results are compelling, demonstrating that the incorporation of bolalipids and regular bilayer lipids in archaeal membranes significantly enhances membrane fluidity and structural stability.

Reviewer #2 (Public review):

Summary:

The authors aimed to understand the biophysical properties of archeal membranes made of bolalipids. Bacterial and eukaryotic membranes are made of lipids that self-assemble into bilayers. Archea, instead, use bolalipids, lipids that have two headgroups and can span the entire bilayer. The authors wanted to determine if the unique characteristics of archaea, which are often extremophiles, are in part due to the fact that their membranes contain bolalipids.

The authors develop a minimal computational model to compare the biophysics of bilayers made of lipids, bolalipids, and mixtures of the two. Their model enables them to determine essential parameters such as bilayer phase diagrams, mechanical moduli, and the bilayer behavior upon cargo inclusion and remodeling.

The author demonstrates that bolalipid bilayers behave as binary mixtures, containing bolalipids organized either in a straight conformation, spanning the entire bilayer, or in a u-shaped one, confined to a single leaflet. This dynamic mixture allows bolalipid bilayers to be very sturdy but also provides remodeling. However, remodeling is energetically more expensive than with standard lipids. The authors speculate that this might be why lipids were more abundant in the evolutionary process.

Strengths:

This is a wonderful paper, a very fine piece of scholarship. It is interesting from the point of view of biology, biophysics, and material science. The authors mastered the modeling and analysis of these complex systems. The evidence for their findings is really strong and complete. The paper is written superbly, the language is precise and the reading experience very pleasant. The plots are very well-thought.

Weaknesses:

None. The authors have addressed all the potential weaknesses that were raised by the reviewers.

Reviewer #3 (Public review):

Summary:

The authors have studied the mechanics of bolalipid and archaeal mixed-lipid membranes via comprehensive molecular dynamics simulations. The Cooke-Deserno 3-bead-per-lipid model is extended to bolalipids with 6 bead. Phase diagrams, bending rigidity, mechanical stability of curved membranes, and cargo uptake are studied. Effects such as formation of U-shaped bolalipids, pore formation in highly curved regions, and changes in membrane rigidity are studied and discussed. The main aim has been to show how the mixture of bolalipids and regular bilayer lipids in archaeal membrane models enhances the fluidity and stability of these membranes.

The authors have presented a wide range of simulation results for different membrane conditions and conformations. Analyses and findings are presented clearly and concisely. Figures, supplementary information and movies are of very high quality and very well present what has been studied. The manuscript is well written and is easy to follow.

The authors have provided detailed response to the points I raised on the first version and have revised their manuscript accordingly. Hence, I only mention what, in my opinion, still deserves to be noted.

Comments:

I previously raised an issue with respect to the resort to the Hamm-Kozlov model for fitting the power spectrum of membrane undulations. The authors provided very nice arguments against my concerns. For the sake of completeness, I include a simple scenario, which will better highlight the issue:

The tilt contribution to the Helfrich Hamiltonian can be written as a quadratic term 1/2 k_t |T|^2, where T is a tilt vector field. This field is written as the difference between the surface normal and the director field aligned with the lipid orientations. In the small deviation Monge description with z=h(x, y) as the height function, the surface normal has the form N=(-dh/dx, -dh/dy, 1). Now assume the director field, n = (b_x, b_y, 1) with small b_x and b_y components. The tilt contribution to the energy thus reads as 1/2 k_t (N - n)^2 ~= 1/2 k_t [|grad h|^2 + 2 b . grad h]. The first term, 1/2 k_t |grad h|^2, is indeed similar to a surface tension term, \sigma |grad h|^2 that you get from the (1 + 1/2 |grad h|^2) approximation to the area element. Therefore, if you only look at height fluctuations, while your membrane actually has some surface tension, it will make distinguishing the tilt contributions to the fluctuations in the linear Monge gauge impossible.

However, considering that the authors have made sure that the membrane is indeed tensionless, this argument is settled.

I had also raised an issue about the correct NpT sampling in the simulations, and I'm glad that the authors also set up more rigorously thermostatted/barostatted simulations to check the validity of their findings.

Also, from the SI, I previously noted that the authors had neglected the longest wavelength mode because it was not equilibrated. This was an important problem and the authors looked into it and ran more simulations that were better equilibrated.

The analysis of energy of U-shaped lipids with the linear model E=c_0 + c_1 * k_bola is indeed very interesting. I am glad that the authors have expanded this analysis and included mean energy measurements.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public review):

Summary:

Amaral et al. presents a study investigating the mesoscale modelling and dynamics of bolalipids.

Strengths:

The figures in this paper are exceptional. Both those to outline and introduce the lipid types, but also the quality and resolution of the plots. The data held within also appears to be outstanding and of significant (hopefully) general interest.

We thank the reviewer for their kind words and the appreciation of our work.

Weaknesses:

In the introduction, I would like to have read more specifics on the biological role of bolalipids. Archaea are mentioned, but this kingdom is huge - there must be specific species that can be discussed where bolalipids are integral to archaeal life. The authors should go beyond ’extremophiles’. In short, they should unpack why the general audience should be interested in these lipids, within a subset of organisms that are often forgotten about.

Following the reviewer’s advice we have revised the introduction of the manuscript, in which we now discuss specific species (Sulfolobus acidocaldarius and Thermococcus kodakarensis) and how in these species bolalipids are integral to archaeal life. We explain that the ratio between bilayer and bolalipids, and the number of cyclopentane rings contained within bolalipids can change to adapt to the environment. The revised parts of the introduction read (p.1 ):

“Like for bacteria and eukaryotes, archaea must keep their lipid membranes in a fluid state (homeoviscous adaptation). This is important even under extreme environmental conditions, such as hot and cold temperatures, or high and low pH values [7]. Because of this, many archaea adapt to changes in their environment by tuning the lipid composition of their membranes: altering the ratio between bola- and bilayer lipids in their membranes [8, 9] and/or by changing the number of cyclopentane rings in their lipid tails, which are believed to make lipid molecules more rigid [5]. For example, Thermococcus kodakarensis increases its tetraether bolalipid ratio from around 50% to over 80% when the temperature of the environment increases from 60 to 85 C [10]. Along the same lines, the cell membrane of Sulfolobus acidocaldarius, can contain over 90 % of bolalipids with up to 8 cyclopentane rings at 70 C and pH 2.5 [5, 11]. It is worth mentioning that in exceptional cases bacteria also synthesise bolalipids in response to high temperatures [12], highlighting that the study of bolalipid membranes is relevant not only for archaeal biology but also from a general membrane biophysics perspective.”

Reviewer #2 (Public review):

Summary:

The authors aimed to understand the biophysical properties of archeal membranes made of bolalipids. Bacterial and eukaryotic membranes are made of lipids that self-assemble into bilayers. Archea, instead, use bolalipids, lipids that have two headgroups and can span the entire bilayer. The authors wanted to determine if the unique characteristics of archaea, which are often extremophiles, are in part due to the fact that their membranes contain bolalipids.

The authors develop a minimal computational model to compare the biophysics of bilayers made of lipids, bolalipids, and mixtures of the two. Their model enables them to determine essential parameters such as bilayer phase diagrams, mechanical moduli, and the bilayer behaviour upon cargo inclusion and remodelling.

The author demonstrates that bolalipid bilayers behave as binary mixtures, containing bolalipids organized either in a straight conformation, spanning the entire bilayer, or in a u-shaped one, confined to a single leaflet. This dynamic mixture allows bolalipid bilayers to be very sturdy but also provides remodelling. However, remodelling is energetically more expensive than with standard lipids. The authors speculate that this might be why lipids were more abundant in the evolutionary process. Strengths:

This is a wonderful paper, a very fine piece of scholarship. It is interesting from the point of view of biology, biophysics, and material science. The authors mastered the modelling and analysis of these complex systems. The evidence for their findings is really strong and complete. The paper is written superbly, the language is precise and the reading experience is very pleasant. The plots are very well-thought-out.

Weaknesses:

I would not talk about weaknesses, because this is really a nice paper. If I really had to find one, I would have liked to see some clear predictions of the model expressed in such a way that experimentalists could design validation experiments.

We thank the reviewer for their very kind assessment. We incorporated their recommendations regarding experimental validation in the discussion section, as follows (p.14):

“Our model makes a number of predictions that could be tested by experiment either in cells or in vitro. First, it predicts that a small increase in the fraction of archaeal bilayer lipids should be sufficient to soften a bolalipid-rich membrane. While this could be tested in the future, so far only very few studies have yet reported experimental analysis of archaeal membrane mixtures [18, 50]. Second, we observed that membranes with moderate bolalipid molecular rigidity k_bola exhibit curvature-dependent bending rigidity. To experimentally verify this, one could extrude membrane tethers from cells while controlling for membrane tension. Finally, to get to the core mechanism underlying our findings, it will be important to develop experimental methods that will allow the fraction of U-shaped bolalipid conformers per leaflet to be imaged and measured.”

Reviewer #3 (Public review):

Summary:

The authors have studied the mechanics of bolalipid and archaeal mixed-lipid membranes via comprehensive molecular dynamics simulations. The Cooke-Deserno 3-bead-per-lipid model is extended to bolalipids with 6 beads. Phase diagrams, bending rigidity, mechanical stability of curved membranes, and cargo uptake are studied. Effects such as the formation of U-shaped bolalipids, pore formation in highly curved regions, and changes in membrane rigidity are studied and discussed. The main aim has been to show how the mixture of bolalipids and regular bilayer lipids in archaeal membrane models enhances the fluidity and stability of these membranes.

Strengths:

The authors have presented a wide range of simulation results for different membrane conditions and conformations. For the most part, the analyses and their results are presented clearly and concisely. Figures, supplementary information, and movies very well present what has been studied. The manuscript is well-written and is easy to follow.

We thank the reviewer for the detailed assessment of our work and their constructive feedback.

Major issues

R3.Q1: The Cooke-Deserno model, while very powerful for biophysical analysis of membranes at the mesoscale, is very much void of chemical information. It is parametrized such that it is good in producing fluid membranes and predicting values for bending rigidity, compressibility, and even thermalexpansioncoefficientfallingintheacceptedrangeofvaluesforbilayermembranes. But it still represents a generic membrane. Now, the authors have suggested a similar model for the archaeal bolalipids, which have chemically different lipids (the presence of cyclopentane rings for one), and there is no good justification for using the same pairwise interactions between their representative beads in the coarse-grained model. This does not necessarily diminish the worth of all the authors’ analyses. What is at risk here is the confusion between ”what we observe this model of bolalipidor mixed-membranes do” and ”how real bolalipid-containing archaeal membranes behave at these mechanical and thermal conditions.”.

As the reviewer correctly notes, Cooke and Deserno used a minimal model, devoid of chemical detail, to represent fluid lipid membranes composed of bilayer lipids. Indeed archaeal lipids are chemically different compared to non-archaeal lipids, but just like non-archaeal lipids, they can be very different from one another. Given the chemical diversity of bolalipids between each other, instead of representing their complexity in a complicated model with many experimentally unconstrained parameters, we here defined a minimal model for bolalipids. The power of this minimal model is to represent the key physical/geometrical characteristics of archaeal membranes, namely the fact that lipid heads on two sides of the membrane are often connected, that bolalipids can exhibit a conformational change, and that bolalipids mix with some percentage of bilayer molecules. We then ask a general question: how do these unique geometrical characteristics of archaeal membranes influence their mechanics and reshaping? The reviewer is however right in pointing out that a model, regardless of its level of details (atomistic, coarse-grained, minimal), is still a model.

Our approach of extending an established coarse-grained model for bilayer lipids to bolalipids is further supported by experimental observations, which report that archaeal bilayer lipids can form membranes of comparable bending rigidity to those of non-archaeal bilayer membranes [53]. Hence, different lipid linkages (archaeal vs. non-archaeal) give rise to fluid, deformable membranes of not too dissimilar rigidities, suggesting that both archaeal and non-archaeal bilayer lipids can be represented by a similar minimal coarse-grained model for the purpose of mesoscopic biophysical investigations. Since archaeal bolalipids have the same core chemical structure as two archaeal bilayer lipids joined by their tail ends, similarly we model a bolalipid by joining two bilayer lipids. Such an approach also efficiently enables us to compare bolalipid with bilayer membranes, and connect to the large body of knowledge on the physics of bilayer membranes.

To conclude, our coarse-grained model is indeed intended to capture the main physical properties of bolalipid membranes, and not their chemical diversity.

R3.Q2: Another more specific, major issue has to do with using the Hamm-Kozlov model for fitting the power spectrum of thermal undulations. The 1/q² term can very well be attributed to membrane tension. While a barostat is indeed used, have the authors made absolutely sure that the deviation from 1/q⁴ behaviour does not correspond to lateral tension?

To the casual observer, any 1/q² trend might point at membrane tension. However, the precise functional form is relevant as it determines whether the 1/q² dominates the 1/q⁴ trend for small or large values of the wave number q in the fitted power spectrum.

The first model (including lipid tilt) exhibits the functional form 1/(kq⁴) + 1/(kq²). In contrast, the second model (including membrane tension) exhibits the functional form 1/(kq⁴ + ∑q²). Importantly, the two models obey a different functional form. Here k and k_θ, are the bending and tilt moduli, which are assumed positive, and ∑ is the membrane tension, which can be either positive or negative. For the first model (with tilt), while for small q the amplitude is proportional to q^-4, for large q the amplitude is proportional to q^-2. In contrast, for the second model (with positive tension) while for small q the amplitude is proportional to q^-2, for large q the amplitude is proportional to q^-4. If membrane tension were to be negative in the second model, the slope would cross from negative infinity for small q to -4 for large q. The functional dependencies are summarized in Author response image 1A.

For rigid bolalipid membranes, it is clearly visible that the slope of the power spectrum plotted against the wave number q decreases with increasing q (Author response image 1B). While the slope initially assumes a value close to 4, it gradually approaches 2 for larger values of q. We conclude that only the model including lipid tilt can fit the power spectrum of membrane fluctuations appropriately (solid-dashed line), whereas the model with tension fails to fit the data (dashed line). We note that the combined model containing both lipid tilt and membrane tension does not give a better fit (dotted line).

To demonstrate that the tension model cannot fit the data, we included the best fits for both models for rigid bolalipid membranes in the new SI section 16 (p. S22) and show that only the tilt model leads to acceptable fits. We also measured the projected membrane tension - , where P_x,P_y are respectively the pressure in x and y direction and  L_z is the dimension of the simulation box in z axis. We found the projected membrane tension to give a negligible value similarly to the one that we indirectly measured by fitting a combined model with both tension and tilt, further confirming our conjecture.

Author response image 1.

(A) Schematic showing the decay of the power spectrum as a function of the wave number q in the tilt model (top), in the tension model with positive membrane tension (middle), and in the tension model with negative membrane tension (bottom). (B) Fitted power spectrum as a function of q for rigid bolalipid membranes (k_bola=5k_BT). The fit shows that while the model with tension (dashed line) cannot fit the data, the model with tilt nicely fits the spectrum (solid-dashed line). The combined model including both tension and tilt does not fit the spectrum any better (dotted line).

R3.Q3: I got more worried when I noticed in the SI that the simulations had been done with combined ”fix langevin” and ”fix nph” LAMMPS commands. This combination does not result in a proper isothermal-isobaric ensemble. The importance of tilt terms for bolalipids is indeed very interesting, but I believe more care is needed to establish that.

In what follows, we show that there is no reason to worry. First of all we want to clarify that the physical setup we simulate is that of a membrane contained in a heat bath under negligible tension with correct diffusional dynamics. To achieve this physical setup, for which we use a Langevin thermostat combined with pressure control via an overdamped barostat, which we implement in LAMMPS by combining ”fix langevin” and ”fix nph”.

In more detail: we simulated particles in an implicit solvent, for which we use a Langevin thermostat to get the right diffusional dynamics. To apply the theory of fitting fluctuation spectrums the simulation box length needs to be (near) constant. However, simulating membranes at a fixed box size results in an average non-zero membrane tension, making it hard to measure bending rigidity. The reason is that the effect of membrane tension is most influential on the largest wavelength modes, which are also most decisive when determining mechanical membrane properties like membrane rigidity. To minimize the effect of tension, we perform our simulation with an overdamped barostat (𝜏_baro = 10 𝜏 _langevin), which keeps the membrane near tensionless, as also done before [32]. In the revised manuscript, we have clarified the statement on the physical ensemble used (p.S2):

“For simulating flat membrane patches of bolalipids, we combined the previously used Langevin thermostat with relaxation time of 1𝜏 with a Nosé–Hoover barostat with relaxation time of 10𝜏. In LAMMPS this amounts to combining the commands ’fix langevin’ with ’fix nph’. We configured the barostat to set lateral pressure P_xy to zero by re-scaling the simulation box in the x-y plane. We compare this setup to a fixed box length setup, and an NPT ensemble setup, in SI section 17.”

To connect our results with statistical mechanics ensemble theory we tested alternative setups. Similar setups, including the formal isothermal-isobaric ensemble, where N,P,T are kept constant using Nose-Hoover style equations for thermostating and barostating with modern corrections [34], which the reviewer refers to, result in very similar fluctuation spectrums. Consequently, our measurements of bending and tilt modulus hold true regardless of the integration scheme. However, such a setup does not correctly capture implicit solvent and diffusional dynamics.

In even more detail: we tested our setup (implemented via ”fix langevin”+”fix nph”) versus a isothermal-isobaric ensemble (implemented via ”fix npt”). We measured volume mean and standard deviation, and found them matching for a reference LJ gas.

To be completely sure, and to please the reviewer, we have performed additional verifications in the new SI section 17, which we summarize in the following. We simulated three representative membranes with different integration schemes: ”fix npt”, ”fix langevin”+”fix nph”, and ”fix langevin” (Langevin dynamics with projected area fixed at the average value obtained from a ”langevin+nph”). We checked that the ”fix nph” barostat is merely equilibrating the membrane to a tensionless configuration, after which the projected membrane area (A_p = L_xL<sub<y</sub>) is practically constant. Consequently, the different schemes resulted in minor changes in the longest wavelength modes that we tracked down to small changes in the negligible tension. The resulting measurements of bending modulus change by less than 10%, and our main text conclusions do not change. Author response image 2 compares the fluctuation spectrums for the different integration schemes.

Author response image 2.

Height fluctuation spectrum, for a bilayer membrane at T_eff =1.1, simulated with Langevin dynamics (pink, ‘langevin‘), our setup (purple, ‘nph+langevin‘), and under an isothermal-isobaric ensemble (blue, ‘npt‘); fits are shown as dotted lines.

R3.Q4: This issue is reinforced when considering Figure 3B. These results suggest that increasing the fraction of regular lipids increases the tilt modulus, with the maximum value achieved for a normal Cooke-Deserno bilayer void of bolalipids. But this is contradictory. For these bilayers, we don’t need the tilt modulus in the first place.

We understand the concern why this might be counter-intuitive, and we thank the reviewer for pointing it out. We first want to stress that the tilt modulus can also be measured for bilayer membranes even if it is not needed to fit the fluctuation spectrum. If we measure the tilt modulus for a bilayer membrane, we obtain a value similar to the previously measured one [36]. Importantly, here we also report measurements for the tilt modulus for bolalipid membranes.

To understand the seemingly contradictory behaviour of the tilt modulus, it is insightful to rewrite the expression for the fluctuation spectrum as done in Eq. (1):

where is a characteristic length scale related to tilt, which we call the tilt persistence length. From the last equation it is easy to see that the tilt modulus 𝜅_𝜃 becomes relevant for the fluctuation spectrum if the tilt persistence length l_𝜃  is not negligible. In other words, this means that we have to consider the tilt modulus 𝜅_𝜃 as relevant, if it is sufficiently small compared to the bending rigidity 𝜅.

However, this is not only counter-intuitive, but also difficult to communicate graphically. Per the excellent reviewer’s suggestion, to make the interpretation more accessible, we converted in the main text and its figures the tilt modulus to the more directly interpretable tilt persistence length l_𝜃, as this is small when tilt is irrelevant (for bilayer lipids and flexible bolalipids) and large otherwise (for rigid bolalipids). This includes changes to the main text on p.6 and p.8 , and to the insets in Figs. 2C and 3B. We note that for completeness we also report the tilt modulus 𝜅_𝜃  in the SI.

R3.Q5: Also, from the SI, I gathered that the authors have neglected the longest wavelength mode because it is not equilibrated. If this is indeed the case, it is a dangerous thing to do, because with a small membrane patch, this mode can very well change the general trend of the power spectrum. As a lot of other analyses in the manuscript rely on these measurements, I believe more elaboration is in order.

We thank the reviewer for the careful examination of our supplementary material. For each fluctuation spectrum measurement, we ran multiple replicas. We observed that the largest wavelength modes were not fully equilibrated. In the simulations the first mode of the fluctuation spectrum is probed at different amplitudes and phases. We thus expected the potential systematic error would show up clearly when comparing spectrums of the different replicas. As we saw no correlation in these systematic offsets between replicas, we concluded that the simulations are sufficiently equilibrated and we could safely exclude the first mode of the fluctuation spectrum from our analysis.

To show without doubt that this procedure does not randomly bias our results, we also ran simulations for three representative membranes until all modes were equilibrated. On the modes previously equilibrated, the resulting spectrums agree with our previous shorter simulations. On the largest wavelength modes that were previously not fully equilibrated, we noticed a small deviation from theory, specifically for flexible membranes (small bending modulus). These small deviations can be explained by including a negligible negative tension. Importantly, however, the resulting bending modulus σ stays nearly the same. We note that the small negative tension disappears when we halve the timestep (see Author response image 3). This verification is shown in SI section 17.

R3.Q6: The authors have found that ”there is a strong dependency of the bending rigidity on the membrane mean curvature of stiffer bolalipids.” The effect is negative, with the membrane becoming less stiff at higher mean curvatures. Why is that? I would assume that with more flexible bolalipids, the possibility of reorganization into U-shaped chains should affect the bending rigidity more (as Figure 2E suggests). While for a stiff bolalipid, not much would change if you increase the mean curvature. This should be either a tilt effect, or have to do with asymmetry between the leaflets. But on the other hand, the tilt modulus is shown to decrease with increasing bolalipid rigidity. The authors get back to this issue only on page 10, when they consider U-shaped lipids in the inner and outer leaflets and write, ”this suggested that an additional membrane-curving mechanism must be involved.” But then again, in the Discussion, the authors write, ”It is striking that membranes made from stiffer bolalipids showed a curvature-dependent bending modulus, which is a clear signature that bolalipid membranes exhibit plastic behaviour during membrane reshaping,” adding to the confusion.

Author response image 3.

Height fluctuation spectrum, for a bilayer membrane at T_eff =1.1, as simulated in the main text (grey, for 60⇥10³τ), for longer duration (1_.44⇥10⁶τ) (pink), and with the longer duration and halved timestep =0.005_τ(purple); fits are shown as dotted lines (tension and tilt) or dash-dot lines (tilt only).

We thank the reviewer for asking this important question. Membrane bending rigidity in bolalipid membranes decreases dramatically once a small fraction of U-shapes is allowed to form, but then plateaus once this U-shape fraction reaches 20%. In a curved bolalipid membrane, U-shapes must accumulate in the outer leaflet to accommodate for area difference. Together, the bending rigidity non-linear dependence on U-shape fraction, and the promotion of U-shapes by curvature, explain why in a membrane made of moderately stiff bolalipids (k_bola = 1k_BT), which contain very few U-shapes in the flatstate, the bending rigidity of the membrane decreases as curvature increases. While in a membrane made of flexible bolalipid molecules (k_bola = 0), where many U-shapes are present in the flat membrane, the bending rigidity does not change with curvature.

Bending rigidity 𝜅 in flat membranes composed of bolalipids decreases dramatically once a small fraction of U-shapes is allowed to form, but plateaus once more than 20% of U-shaped bolalipids are present. In details, our data shows that with an increasing bolalipid molecular rigidity k_bola, both the number of U-shaped bolalipids decreases (Fig. 2B) and the membrane rigidity 𝜅 increases (Fig. 2C). Thus, the correlation suggests that U-shaped bolalipids soften the membrane, in a non-linear way where most of the change in membrane bending rigidity happens for U-shaped bolalipid fraction < 20% (Figure S11).

Separately, membrane curvature affects the area difference between curved membrane leaflets and thus drives U-shape accumulation. To be specific, a cylindrical membrane with area A, mean curvature H and thickness h has the outer leaflet with area A(1 + Hh) and the inner leaflet with smaller area A(1 Hh). This can be large, in our simulations up to an area change of Hh \= 25%. For pure bolalipid membranes, straight bolalipids occupy the same space in each leaflet. Area difference can then be achieved only by having a different amount of U-shaped bolalipids in each leaflet, which can result in a different U-shape fraction between leaflets and thus ’asymmetry between leaflets’. Figure S10 confirms U-shape head fraction asymmetry that increases with curvature, for both flexible (k_bola = 0) and moderately stiff bolalipids (k_bola = 1k_BT).

Together, these two effects result in membrane softening under curvature for the moderately stiff bolalipids, but constant rigidity for flexible bolalipids (Fig. 2F). In details: for membranes composed of moderately stiff bolalipid molecules (k_bola = 1k_BT), the U-shape bolalipid head fraction only increases in the outer leaflet, goingfrom10to20%(Figure S10). This is in the high sensitivity region where the bending rigidity is expected to change the most (Figure S11). We hypothesize that the molecular rigidity of a U-shaped bolalipid creates compression on the outer leaflet that stabilizes the membrane curvature and thus causes membrane softening. We suspect that for membranes composed of rigid bolalipids (k > 1k_BT), the effect is likely not present due to the absence of U-shape formation even under strong bending.

By contrast, for membranes composed of flexible bolalipids (k = 0), the U-shaped bolalipid head fraction changes relatively little from its value for flat membranes (from 50% to respectively 60 and 40% for the outer and inner leaflet, Figure S10). This is in the region where the membrane bending rigidity is expected to respond weakly to U-shape fraction (Figure S11). Additionally, the change is symmetric, so presumably the outer leaflet becomes softer as the inner leaflet becomes stiffer, thus creating opposing effects and only weakly affecting the membrane bending rigidity as a whole. We note that the distinction between the U-shape head fraction that we plot (Figure S10) and U-shape fraction (Figure S11) matters little for this analysis.

We have added this deduction and its plots to SI section 8, and revised the corresponding statement in the main text accordingly (p.7 ).

“Changing membrane curvature alters the area differently in the two membrane leaflets. To adapt to the area difference, we thus expect the fraction of U-shaped bolalipids to change as the membrane curvature changes. Moreover, the results of Fig. 2B and Fig. 2C showed that the U-shaped bolalipid fraction and the membrane bending rigidity are correlated. As a result, we predict that the fraction of straight versus U-shaped bolalipids in a membrane will change in response to membrane bending, in a way that makes the bending rigidity of a bolalipid membrane curvature dependent.”

R3.Q7: This issue is repeated when the authors study nanoparticle uptake. They write: ”to reconcile these seemingly conflicting observations we reason that the bending rigidity, similar to Figure 2F, is not constant but softens upon increasing membrane curvature, due to dynamic change in the ratio between bolalipids in straight and U-shaped conformation. Hence, bolalipid membranes show stroking plastic behaviour as they soften during reshaping.” But the softening effect that they refer to, as shown in Figure 4B, occurs for very stiff bolalipids, for which not much switching to U-shaped conformation should occur.

We thank the reviewer for locating a particularly dense sentence. We changed the text to explicitly refer to the range k 2 [0,2] k_BT for which there is significant change in U-shape fraction (p.8 ):

“To reconcile these seemingly conflicting observations we reason that the bending rigidity κ, similar to Fig. 2F, is not constant but softens in the range k 2 [0,2] k_BT, upon increasing membrane curvature. This is due to the dynamic change in the ratio between bolalipids in straight and U-shaped conformation.”

As for Fig. 4B, for k > 2k_BT, pores form thus explaining the plateau in adsorption energy.

R3.Q8: Another major issue is with what the authors refer to as the ”effective temperature”. While plotting phase diagrams for kT/eps value is absolutely valid, I’m not a fan of calling this effective temperature. It is a dimensionless quantity that scales linearly with temperature, but is not a temperature. It is usually called a ”reduced temperature”. Then the authors refer to their findings as studying the stability of archaeal membranes at high temperatures. I have to disagree because eps is not the only potential parameter in the simulations (there are at least space exclusion and angle-bending stiffnesses) so one cannot identify changing eps with changing the global simulation temperature. This only works when you have one potential parameter, like an LJ gas.

We indeed thought about this before and found that it makes little difference in our set-up. To thoroughly show that the distinction matters very little, per reviewer’s question, we computed our phase diagrams by scaling temperature T explicitly (and not lipid tail interactions T_eff = k_BT /ϵ_p). We added these results to the SI section 14 and found no significant difference when comparing scaling tail interactions (Figure S15A) with scaling temperature explicitly (Figure S15B).

We also computed Fig. 2A-C for scaling interactions (Figure S17A) and scaling temperature explicitly (Figure S17B). We found a slightly increased U-shaped bolalipid fraction for low k when comparing scaling interactions (Figure S17A) with temperature scaling (Figure S17B). The reason is that the U-shaped fraction depends on temperature, as with higher temperature bolalipids can easier transition into the U-shape. Most importantly, however, we found no qualitative changes on the liquid region or the mechanical membrane properties when we compared the different scaling variants.

The reason why both scaling variants match so well can be understood easily. All pair potentials, including volume exclusion interactions between head beads and other membrane beads, were also scaled in the same manner as tail-to-tail interactions, as described in the SI. In contrast, the energy scales for maintaining the lipid bonds, the bilayer lipid angles and the bolalipid angles are relatively large compared to the energy scales involved in tail-to-tail interactions. This separation of energy scales guarantees that there will be little effect when increasing global temperature. Regarding nomenclature, we take the reviewer’s advice and have added ’reduced temperature’ as an alias for T_eff in the main text.

In the revised version of the manuscript, we mention these observations in the SI section 14 and point towards these results in the main text (p.4 ):

“This interaction strength governs the membrane phase behaviour and can be interpreted as the effective temperature or reduced temperature T_eff = k_BT /ϵ_p. As the distinction between scaling interactions (T_eff) or temperature (T) is not important for our analysis (see Supplemental Information (SI) section 14), for simplicity we refer to T_eff as temperature in the following.”

Minor issues

R3.Q9: As the authors have noted, the fact that the membrane curvature can change the ratio of U-shaped to straight bolalipids would render the curvature elasticity non-linear (though the term ”plastic” should not be used, as this is still structurally reversible when the stress is removed. Technically, it is hypoelastic behaviour, possibly with hysteresis.) With this in mind, when the authors use essentially linear elastic models for fluctuation analysis, they should make a comparison of maximum curvatures occurring in simulations with a range that causes significant changes in bolalipid conformational ratios.

We thank the reviewer for their suggestion on calling the non-linear behaviour of the curvature elasticity hypoelastic. We have edited the main text accordingly (p.8 ):

“In an elastic material, the strain modulus holds constant and deformation is reversible. For bolalipid membranes at k = 1k_BT, however, the bending modulus decreases when deformation increases, rendering bolalipid membranes hypoelastic.”

Moreover, regarding the maximum curvatures occurring in the fluctuation simulations: We first note that the ensemble average of the mean curvature H from the fluctuation measurements is indicated as a vertical line in Fig. 2F. As the average value is nearly zero, the membrane can be considered as flat in good approximation. To investigate the question in more detail, we extended the SI with a careful analysis of the validity of the maximum membrane curvature and the validity of the Monge gauge approximation (SI section 15).

In short, we found that the involved membrane curvatures are small and therefore are unlikely to trigger any significant changes of the bending modulus. Moreover, since we are dealing with two bolalipid conformations, we also tested the homogeneity of the membrane. In our simulations of flat membrane patches we did not observe clustering or phase separation between the two bolalipid conformations beyond the [2,3]σ range. Furthermore, we get good agreement between our fluctuation measurement and the cylinder simulations in Fig. 2F. We now mention this verification in the revised version of the manuscript (p.8 ):

“Fortunately, this dependency on curvature does not invalidate our fluctuation results, where the curvature is small enough that its effect on the bending modulus is negligible (SI section 15).”

Last but least, simulating bending/unbending cycles of an arc-shaped membrane (frozen endpoints) shows agreement with cylinder membrane simulations, and no hysteresis at the rates of deformation employed (cf. M. Amaral’s thesis [54], soon to be out of the embargo period).

R3.Q10: The Introduction section of the manuscript is written with a biochemical approach, with very minor attention to the simulation works on this system. Some molecular dynamics works are only cited as existing previous work, without mentioning what has already been studied in archaeal membranes. While some information, like the binding of ESCRT proteins to archaeal membranes, though interesting, helps little to place the study within the discipline. The Introduction should be revised to show what has already been studied with simulations (as the authors mention in the Discussion) and how the presented research complements it.

The present research for the first time covers archaeal membranes with a single coarse-grained model capable of assuming both bolalipid in-membrane conformations and sweeps through temperature, membrane composition, and molecular rigidity. The work shows the first curvature dependent bending modulus for pure bolalipid membranes. It also investigates systematically bending modulus and Gaussian modulus, and tests the model in an all-encompassing budding simulation that incorporates topology changes. Existing atomistic or coarse-grained MD simulations (MARTINI or similar force fields) are limited to small patches of membrane, with no study of large-scale deformations or topology changes; plus, they rely on force fields that were parametrized for bilayer membranes.

To give a comprehensive overview of the field, we revised the introduction section of the manuscript, in which we now discuss previous computational work investigating membrane diffusivity, U-shaped lipid fraction, and bending rigidity (p.3 ):

“By contrast, only a few studies have investigated bolalipid membranes applying computational or theoretical tools [24, 25]. Specifically, the pore closure time in bolalipid membranes, and the role of cyclopentane rings for membrane properties has been investigated using all-atom simulations, showing decreased lateral mobility, reduced permeability to water, and increased lipid packing [26–28]. Moreover, using coarse-grained simulations, it was suggested that bolalipid membranes are thicker [29], exhibit a gel-to-liquid phase transition at higher temperature [30], and exhibit a reduced diffusivity [31]. However, little research has been devoted to investigating mechanics and reshaping of bolalipid membranes at the mesoscale despite the obvious importance of this question from evolutionary, biophysics, and biotechnological perspectives and although different membrane physics is expected to manifest.”

Following the reviewer’s advice and to keep the introduction concise and focused on bolalipid membranes, we have removed the paragraph on ESCRT-III proteins in the revised manuscript.

R3.Q11: The authors have been a bit loose with using the term ”stability”. I’d like to see the distinction in each case, as in ”chemical/thermal/mechanical/conformational stability”.

We have clarified when applicable the type of stability throughout the manuscript. In all other instances, if not clear from context, we mean simply that the membrane persists being a membrane. At our coarse-grained level, this means the membrane does not disassemble into a gas phase.

R3.Q12: In the original Cooke-Deserno model, a so-called ”poorman’s angle-bending term” is used, which is essentially a bond-stretching term between the first and third particle. However, I notice the authors using the full harmonic angle-bending potential. This should be mentioned.

This is made clear in the SI (Eq. (S3)). Cooke and Deserno mention the harmonic angle potential as a valid alternative in their original publication. We now also added this detail to the main text (p.3 ):

“The angle formed by the chain of three beads is kept near 180° via an angular potential with strength k₀, instead of the approximation by a bond between end beads of the original model [32].”

R3.Q13: The analysis of energy of U-shaped lipids with the linear model E \= c₀ + c₁k is indeed very interesting. I am curious, can this also be corroborated with mean energy measurements? The minor issue is calling the source of the favorability of U-shaped lipids ”entropic”, while clearly an energetic contribution is found. The two conformations, for example, might differ in the interactions with the neighbouring lipids.

We were also curious and thank the reviewer for the suggestion of mean energy measurements. We concluded that there must be either an entropic contribution to the free energy or an intermolecular interaction energy favouring U-shaped bolalipids. We have now included these measurements in SI section 6 (p.S5 ):

“By splitting the average potential energy between an internal contribution (bonds, angles and pair interactions between particles in the same molecule) and an external contribution (pair interactions between a molecule and its neighbours), we determined the transition energy from straight to U-shaped bolalipids in detail. We found that this transition lowers the internal potential energy of the bolalipid while increasing its interaction energy. In total, we obtained an energy barrier for the transition of ΔE_s→u = 0.79±0.01k_BT. Since the fit indicates, however, that the U-shaped bolalipid conformation is preferred over the straight conformation, we conclude that there must be either an entropic contribution to the free energy or an intermolecular interaction energy favouring U-shaped bolalipids.”

We refer to these measurements in the main text (p.6 ):

“For the fit it appears that c₀ < 0, which implies that bolalipids in U-shape conformation are slightly favoured over straight bolalipids at k = 0 (explored in SI section 6).”

R3.Q14: The authors write in the Discussion, ”In any case, our results indicate that membrane remodelling, such as membrane fission during membrane traffic, is much more difficult in bolalipid membranes [34].” Firstly, I’m not sure if studying the dependence of budding behaviour on adhesion energy with nanoparticles is enough to make claims about membrane fission. Secondly, why is the 2015 paper by Markus Deserno cited here?

We thank the reviewer for giving us the opportunity to clarify. We make an energetic argument on membrane fission based on the observed difference in the ratio of .

Splitting a spherical membrane vesicle into two spherical vesicles (fission) increases the bending energy by 8𝜋𝜅 and decreases the energy related to the Gaussian bending modulus by . The second part of the argument is given for example in the review by Markus Deserno (p.23, right column), that’s why we cite the paper here. Together, this gives an energy barrier, required for membrane fission in the considered geometry of ∆E_fission = . We found that is around 0.5 for bolalipid membranes and around 1 for bilayer membranes. Since 𝜅 was typically larger in bolalipid membranes we thus expect the energy barrier for fission ∆E_fission to be larger for bolalipid membranes. We therefore predict that membrane remodelling, such as membrane fission during membrane trafficking, is harder in bolalipid membranes. We explain our reasoning in the discussion of the revised manuscript (p.13 ):

“Membrane remodelling, such as the fission of one spherical vesicle into two, increases the bending energy by 8πκ but decreases the energy related to the Gaussian modulus by – [39], giving rise to a fission energy barrier of ∆E_fission = . Our results indicated that while in bolalipid membranes 𝜅 is larger, is smaller compared to bilayer membranes. Our results thus predict a larger energy barrier for membrane fission ∆E_fission in bolalipid membranes compared to bilayer membranes.”

R3.Q15: In the SI, where the measurement of the diffusion coefficient is discussed, the expression for D is missing the power 2 of displacement.

We thank the reviewer for spotting this oversight. We corrected it in the revised version of the SI (p.S5 ).

R3.Q16: Where cargo uptake is discussed, the term ”adsorption energy” is used. I think the more appropriate term would be ”adhesion energy”.

For the sake of simplicity, we changed the term to adhesion energy (caption of Fig. 4, and p.10). We do not have a strong opinion on this, but we believe that adsorption energy would be equally correct as we describe the adsorption of many lipid head beads to a nanoparticle.

R3.Q17: Typos:

Page 1, paragraph 2: Adaption → Adaptation. Page 10, paragraph 1: Stroking → Striking.

We thank the reviewer for spotting these typos which we have corrected in the revised version of the manuscript.

Recommendations for the authors

Reviewer #1 (Recommendations for the authors):

A few thoughts (likely out of the scope of this paper but possibly to consider upon revision):

R1.Q1: Do bolalipids always have the same headgroup? I don’t recall reading this in the introduction/discussion. R1 and R2 are in Figure 1, but I don’t know whether there are standard types. Could this be expanded upon? Is the model able to take these differences into account?

We thank the reviewer for raising this important question. Similar to bacteria and eukaryotes, in archaea there is a huge variety in terms of the different head groups that lipids can contain and thus also lipid variety. Most archaeal lipids have head groups that contain either phosphate groups or sugar residues. Typically, archaeal bolalipids are asymmetric and contain a phosphatidyl and a sugar moiety at the two ends of the lipid molecule. Within the membrane the lipid is oriented such that the phosphatidyl moiety points towards the interior of the cell whereas the sugar moiety points towards the outside of the cell as it occupies more space [5].

In our computational model, however, we consider symmetric bolalipids for the sake of simplicity and to decouple the role of ”connected geometry” from other effects. In principle, we could investigate the effect of lipid asymmetry by increasing the size of one of the lipid head beads. However, this investigation exceeds the scope of the present study and therefore requires future work.

In the revised version of the manuscript, we now clarify that bolalipids can have different headgroups (p.1 and the caption of Fig. 1):

“The hydrophilic heads can be composed of different functional groups with phosphatidyl and sugar being the most relevant moieties. For bolalipids the two head groups at either end of the molecule are typically distinct (Fig. 1A right) [5].”

“The hydrophilic head of a bolalipid can be composed of different functional groups represented by R1 and R2 (right).”

We also explicitly state that we neglect lipid head group asymmetry for the sake of simplicity (p.4 ):

“To decouple the effect of the connected geometry of the bolalipids from that of lipid asymmetry, we assume both head beads of a bolalipid to share the same properties.”

R1.Q2: Is it possible to compare the mesoscale models to either Coarse-grained or even all-atom lipid models? Have simulations previously been performed for bolalipids at those levels of description?

A few studies have investigated bolalipids membranes in simulations previously. These studies either used all-atom or coarse-grained simulations. However, none of these studies investigated how bolalipids respond to membrane deformations. Therefore, it is currently not possible to directly compare our results to studies in the literature. However, to recapitulate our predictions experimentally is certainly something that could and should be done in the future. As a reply to this reviewer and reviewer 3, we discuss the current state of modelling bolalipid membranes in simulations in the revised version of the manuscript (p.3 ):

“By contrast, only a few studies have investigated bolalipid membranes applying computational or theoretical tools [24, 25]. Specifically, the pore closure time in bolalipid membranes, and the role of cyclopentane rings for membrane properties has been investigated using all-atom simulations, showing decreased lateral mobility, reduced permeability to water, and increased lipid packing [26–28]. Moreover, using coarse-grained simulations, it was suggested that bolalipid membranes are thicker [29], exhibit a gel-to-liquid phase transition at higher temperature [30], and exhibit a reduced diffusivity [31]. However, little research has been devoted to investigating mechanics and reshaping of bolalipid membranes at the mesoscale despite the obvious importance of this question from evolutionary, biophysics, and biotechnological perspectives and although different membrane physics is expected to manifest.”

We want to mention, however, that we do compare membrane diffusivity, U-shaped lipid fraction, and bending rigidity to the behaviour and values that have been previously measured in simulations in the discussion section. In general, we find good agreement between our results and previously reported behaviour/values (p.13 ):

“While flexible bolalipid membranes are liquid under the same conditions as bilayer membranes, we found that stiff bolalipids form membranes that operate in the liquid regime at higher temperatures. These results agree well with previous molecular dynamics simulations that suggested that bolalipid membranes are more ordered and have a reduced diffusivity compared to bilayer membranes [24, 29]. In our simulations, this is due to the fact that completely flexible bolalipids molecules adopt both straight (transmembrane) as well as the U-shaped (loop) conformation with approximately the same frequency. In contrast, stiff bolalipids typically only take on the straight conformation when assembled in a membrane. These results agree with the previous coarse-grained molecular dynamics simulations using the MARTINI force field which showed that the ratio of straight to U-shaped bolalipids increased upon stiffening the linker between the lipid tails [29].

[...]

When we determined the bending rigidity of bolalipid membranes by measuring their response to thermal fluctuations, we found that membranes made from flexible bolalipids are only slightly more rigid than bilayer membranes. This result is consistent with previous atomistic simulations, which showed that the membrane rigidity was similar for membranes composed of bilayer lipids and flexible synthetic bolalipids [45].”

R1.Q3: How would membrane proteins alter the behaviour of bolalipids? Either those integral to the membrane or those binding peripherally?

The reviewer asks an important question. However, the question is difficult to answer due to its scope and the gaps in the current literature. Important examples of integral or peripheral membrane proteins that alter the behaviour of bolalipids and archaeal bolalipid membranes are involved in cell homeostasis, cell division, membrane trafficking, and lipid synthesis.

The cells of many archaeal species are enclosed in a paracrystalline protein layer called the Slayer, which is attached to the lipid membrane [4, 55]. The main function of the S-layer is to keep the cell’s shape and to protect it against osmotic stress. Due to the embedding of the S-layer in the membrane at specific locations, it is to be expected that the membrane properties are influenced by the S-layer. Furthermore, archaea execute cell division by locally reshaping the membrane using FtsZ and ESCRT-III proteins [56]. While Asgard archaeal genomes encode proteins with homology to those regulating aspects of eukaryotic membrane remodelling and trafficking [57], they have yet to be observed undergoing a process like endocytosis [58]. In addition, it has been speculated that the proteins that drive the synthesis of two diether lipids into a tetraether lipid are either membrane associated or integral membrane proteins [59].

However, to the best of our knowledge it is not known how membrane proteins specifically alter the behaviour of bolalipids. Future work will need to be executed to answer this question. Following the advice of reviewer 3 and to keep the introduction concise and focused on bolalipid membranes, we do not mention these observations in the revised manuscript.

R1.Q4: Is there a mechanism in cells to convert or switch bolalipids from a straight to a u-shaped description? Does this happen spontaneously or are there enzymes responsible for this?

We thank the reviewer for bringing up this important point. Despite the relevance of the question, little is currently known about the mechanism that make bolalipids transition between a straight and a U-shaped configuration mainly because there is to date no established experimental method.

Besides our own results, most of what we know comes from coarse-grained molecular dynamics simulations, which showed that bolalipids can spontaneously transition between the straight and U-shaped configuration [29]. In addition, by using comparative genomic analysis, it has been predicted that many archaeal species contain flippases, i.e., membrane proteins that are able, upon the consumption of energy, to transfer (flipflop) bilayer lipids between the two membrane leaflets [43]. Moreover, it has been shown that Halobacterium salinarum (an archaeon with a bilayer lipid membrane) [44] contains scramblases, which are membrane proteins that passively transfer bilayer lipids from one membrane leaflet to the other. It is therefore tempting to speculate that similar proteins might exist for bolalipids which could facilitate the straight to U-shaped transition.

In addition, it has been reported that vesicles composed of bolalipid membranes can undergo fusion with enveloped influenza viruses [17]. In this context, it has been suggested that the influenza fusion protein hemagglutinin may locally induce U-shaped bolalipids to facilitate membrane fusion. However, all these hints are by far no proof of a mechanism that can drive the straight to U-shaped bolalipid transition, and further work needs to be done to investigate this question in detail.

In the revised version of the manuscript, we now discuss what is known about potential mechanisms to facilitate the straight to U-shaped transition in the discussion section (p.13 ):

“While previous coarse-grained simulations predicted that bolalipids spontaneously transition between the straight and U-shaped conformations [29], how this happens in archaeal membranes and whether membrane proteins are involved in this conformational transition needs to be clarified in the future. Experimental studies suggest that archaeal membranes contain flippases and scramblases for the transitioning of bilayer lipids between membrane leaflets [43, 44], raising the possibility that similar proteins could also facilitate conformational transitions in bolalipids. In addition, it has been suggested that the viral fusion protein hemagglutinin could cause a transition from straight to U-shaped bolalipid conformation during the fusion of bolalipid vesicles with influenza viruses [17]. However, future investigation is required.”

R1.Q5: Ideally, coordinates and any parameter files required to run the molecular simulations should be included for reproducibility.

We absolutely share the reviewer’s concern with reproducibility and as such have included in the original submission as part of our data availability section a link to a code repository (available at: https://doi.org/10.5281/zenodo.13934991 [51]) that allows initializing and simulating flat membrane patches, with user control of the parameters explored in this paper (𝜔,T_eff,k_bola,f^bi).

Reviewer #2 (Recommendations for the authors):

This is a great paper and I congratulate the authors for writing such a fine piece of scholarship. The only nitty-gritty feedback that I have is summarized in the following three points:

R2.Q1: In the introduction the authors talk about archaea adapting their membrane to retain membrane fluidity. However, homeoviscous adaptation is also fundamental in bacteria and eukaryotes.

The reviewer is correct, like archaea the membranes of bacteria and eukaryotes must balance between flexibility and stability. Moreover, the cell membranes in all 3 domains of life need to maintain membrane fluidity and provide mobility to the embedded lipids and membrane proteins (homeoviscous adaptation). The general idea is that these organisms change the ratio of different lipids to change membrane properties and thereby optimally adapt to their environments [10]. Importantly, however, there are differences of how homeoviscous adaptation is maintained across the different domains of life. As a reply to this reviewer and reviewer 3, we now discuss the underlying mechanisms in the revised parts of the introduction (p.1 ):

“Like for bacteria and eukaryotes, archaea must keep their lipid membranes in a fluid state (homeoviscous adaptation). This is important even under extreme environmental conditions, such as hot and cold temperatures, or high and low pH values [7]. Because of this, many archaea adapt to changes in their environment by tuning the lipid composition of their membranes: altering the ratio between bola- and bilayer lipids in their membranes [8, 9] and/or by changing the number of cyclopentane rings in their lipid tails, which are believed to make lipid molecules more rigid [5]. For example, Thermococcus kodakarensis increases its tetraether bolalipid ratio from around 50% to over 80% when the temperature of the environment increases from 60 to 85 C [10]. Along the same lines, the cell membrane of Sulfolobus acidocaldarius, can contain over 90 % of bolalipids with up to 8 cyclopentane rings at 70 C and pH 2.5 [5, 11]. It is worth mentioning that in exceptional cases bacteria also synthesise bolalipids in response to high temperatures [12], highlighting that the study of bolalipid membranes is relevant not only for archaeal biology but also from a general membrane biophysics perspective.”

R2.Q2: Uncertainties in Gaussian rigidity modulus estimates are not properly reported.

The large uncertainties in the Gaussian rigidity modulus were due to the fact how they were calculated. In short, is determined in cap folding simulations [41] (SI section 9), by using the measured values of the dimensionless parameter 𝜉, related to the folding probability, the bending modulus 𝜅, the membrane line tension , and the cap radius R. In our case, the main source of uncertainty for determining comes from the uncertainty in the measurement of the bending rigidity 𝜅. To obtain 𝜅, previously, we fitted fluctuation spectra for different seeds and only then averaged the obtained values. In the revised version of the manuscript, we now first pool the fluctuation spectra of the different simulation seeds before we fit all spectra at the same time. This new approach results in smaller uncertainties for the bending rigidity 𝜅 and also the Gaussian rigidity modulus .

As a consistency check, in addition to the simulations that we previously performed at T_eff = 1.3, we have repeated the cap folding and line tension simulations at T_eff = 1.2, resulting in similar values for . In the revised version of the manuscript, we report the newly calculated values and uncertainties for at T_eff  = 1.2 in the main text (p.8 ):

“At T_eff  = 1.2, we obtained = 4.30±0.22kBT and thus a ratio of = 0.89±0.04 for bilayer membranes, similar to what has been reported previously [41]. For flexible bolalipid membranes, we got a slightly smaller value for = 5.04 ± 0.37kBT. Due to the larger bending modulus, however, flexible bolalipid membranes show a significantly smaller ratio = 0.64± 0.04 (k = 0). At larger temperature (Teff = 1.3), the ratio can be even smaller = 0.45 ± 0.07 (see SI section 9).”

In addition, we report the values at T_eff = 1.3 and T_eff = 1.2 in the SI (p.S15 , Tabl. S4):

We have also adapted the discussion of the Gaussian bending modulus accordingly (p.13 ):

“Another marked difference between bilayer and flexible bolalipid membranes is the ratio of the Gaussian rigidity to the bending modulus. Instead of being around 1 as for bilayer membranes [41], it is around 1/2 and therefore only half of that of bilayer lipids.”

Reviewer #3 (Recommendations for the authors):

While I think the bulk of the work presented is useful, some of the issues that I raised in my review are indeed major. Without properly addressing them, it is hard to accept the conclusions of the manuscript. I hope the authors can address them by revising their analysis.

We thank the reviewer for their constructive feedback, which helped us to improve the manuscript. We have addressed all points raised by the reviewer in our detailed point-by-point response to the reviewer (see above). We hope the reviewer will now find it easier to accept our conclusions.

(1) R. Phillips, J. Kondev, J. Theriot, and H. Garcia, Physical biology of the cell (Garland Science, New York, 2012).

(2) H. T. McMahon and J. L. Gallop, Membrane curvature and mechanisms of dynamic cell membrane remodelling, Nature 438, 590 (2005).

(3) S. B. Gould, Membranes and evolution, Curr. Biol. 28, R381 (2018).

(4) S.-V. Albers and B. H. Meyer, The archaeal cell envelope, Nat. Rev. Microbiol. 9, 414 (2011).

(5) P. M. Oger and A. Cario, Adaptation of the membrane in Archaea, Biophys. Chem. 183, 42 (2013).

(6) K. Rastädter, D. J. Wurm, O. Spadiut, and J. Quehenberger, The Cell Membrane of Sulfolobus spp.—Homeoviscous Adaption and Biotechnological Applications, International Journal of Molecular Sciences 21, 3935 (2020).

(7) P. L.-G. Chong, Archaebacterial bipolar tetraether lipids: Physico-chemical and membrane properties, Chem. Phys. Lipids 163, 253 (2010).

(8) M. Tourte, P. Schaeffer, V. Grossi, and P. M. Oger, Functionalized Membrane Domains: An Ancestral Feature of Archaea?, Front. Microbiol. 11, 526 (2020).

(9) Y. H. Kim, G. Leriche, K. Diraviyam, T. Koyanagi, K. Gao, D. Onofrei, J. Patterson, A. Guha, N. Gianneschi, G. P. Holland, M. K. Gilson, M. Mayer, D. Sept, and J. Yang, Entropic effects enable life at extreme temperatures, Sci. Adv. 5, eaaw4783 (2019).

(10) M. F. Siliakus, J. van der Oost, and S. W. M. Kengen, Adaptations of archaeal and bacterial membranes to variations in temperature, pH and pressure, Extremophiles 21, 651 (2017).

(11) D. W. Grogan, Phenotypic characterization of the archaebacterial genus sulfolobus: comparison of five wild-type strains, J. Bacteriol. 171, 6710 (1989).

(12) D. X. Sahonero-Canavesi, M. F. Siliakus, A. Abdala Asbun, M. Koenen, F. von Meijenfeldt, S. Boeren, N. J. Bale, J. C. Engelman, K. Fiege, L. Strack van Schijndel, J. S. Sinninghe Damsté, and L. Villanueva, Disentangling the lipid divide: Identification of key enzymes for the biosynthesis of membrane-spanning and ether lipids in Bacteria, Sci. Adv. 8, eabq8652 (2022).

(13) M. van Wolferen, A. A. Pulschen, B. Baum, S. Gribaldo, and S.-V. Albers, The cell biology of archaea, Nat. Microbiol. 10.1038/s41564-022-01215-8 (2022).

(14) U. Bakowsky, U. Rothe, E. Antonopoulos, T. Martini, L. Henkel, and H.-J. Freisleben, Monomolecular organization of the main tetraether lipid from Thermoplasma acidophilum at the water–air interface, Chem. Phys. Lipids 105, 31 (2000).

(15) C. Jeworrek, F. Evers, M. Erlkamp, S. Grobelny, M. Tolan, P. L.-G. Chong, and R. Winter, Structure and Phase Behavior of Archaeal Lipid Monolayers, Langmuir 27, 13113 (2011).

(16) D. P. Brownholland, G. S. Longo, A. V. Struts, M. J. Justice, I. Szleifer, H. I. Petrache, M. F. Brown, and D. H. Thompson, Phase Separation in Binary Mixtures of Bipolar and Monopolar Lipid Dispersions Revealed by 2H NMR Spectroscopy, Small Angle X-Ray Scattering, and Molecular Theory, Biophysical Journal 97, 2700 (2009).

(17) A. Bhattacharya, I. D. Falk, F. R. Moss, T. M. Weiss, K. N. Tran, N. Z. Burns, and S. G. Boxer, Structure–function relationships in pure archaeal bipolar tetraether lipids, Chem. Sci. 15, 14273 (2024).

(18) V. Vitkova, D. Mitkova, V. Yordanova, P. Pohl, U. Bakowsky, G. Staneva, and O. Batishchev, Elasticity and phase behaviour of biomimetic membrane systems containing tetraether archaeal lipids, Colloids Surf. A Physicochem. Eng. Asp. 601, 124974 (2020).

(19) E. Chang, Unusual thermal stability of liposomes made from bipolar tetraether lipids, Biochem. Biophys. Res. Commun. 202, 673 (1994).

(20) O. V. Batishchev, A. S. Alekseeva, D. S. Tretiakova, T. R. Galimzyanov, A. Y. Chernyadyev, N. R. Onishchenko, P. E. Volynsky, and I. A. Boldyrev, Cyclopentane rings in hydrophobic chains of a phospholipid enhance the bilayer stability to electric breakdown, Soft Matter 16, 3216 (2020).

(21) U. Seifert, Configurations of fluid membranes and vesicles, Adv. Phys. 46, 13 (1997).

(22) H. Noguchi, Membrane Simulation Models from Nanometer to Micrometer Scale, J. Phys. Soc. Jpn. 78, 041007 (2009).

(23) F. Frey and T. Idema, More than just a barrier: using physical models to couple membrane shape to cell function, Soft Matter 17, 3533 (2021).

(24) C. Huguet, S. Fietz, A. Rosell-Melé, X. Daura, and L. Costenaro, Molecular dynamics simulation study of the effect of glycerol dialkyl glycerol tetraether hydroxylation on membrane thermostability, Biochimica et Biophysica Acta (BBA) - Biomembranes 1859, 966 (2017).

(25) T. R. Galimzyanov, P. I. Kuzmin, P. Pohl, and S. A. Akimov, Elastic deformations of bolalipid membranes, Soft Matter 12, 2357 (2016).

(26) T. R. Galimzyanov, P. E. Volynsky, and O. V. Batishchev, Continuum elasticity and molecular dynamics of a pore in archaeal bolalipid membranes, Soft Matter 21, 687 (2025).

(27) A. O. Chugunov, P. E. Volynsky, N. A. Krylov, I. A. Boldyrev, and R. G. Efremov, Liquid but Durable: Molecular Dynamics Simulations Explain the Unique Properties of Archaeal-Like Membranes, Sci. Rep. 4, 7462 (2015).

(28) L. F. Pineda De Castro, M. Dopson, and R. Friedman, Biological Membranes in Extreme Conditions: Simulations of Anionic Archaeal, PLoS One 11, e0155287 (2016).

(29) M. Bulacu, X. Périole, and S. J. Marrink, In Silico Design of Robust Bolalipid Membranes, Biomacromolecules 13, 196 (2012).

(30) C. H. Davis, H. Nie, and N. V. Dokholyan, Insights into thermophilic archaebacterial membrane stability from simplified models of lipid membranes, Phys. Rev. E 75, 051921 (2007).

(31) S. Dey and J. Saha, Minimal Coarse-Grained Modeling toward Implicit Solvent Simulation of Generic Bolaamphiphiles, J. Phys. Chem. B 124, 2938 (2020).

(32) I. R. Cooke and M. Deserno, Solvent-free model for self-assembling fluid bilayer membranes: Stabilization of the fluid phase based on broad attractive tail potentials, J. Chem. Phys. 123, 224710 (2005).

(33) P. L.-G. Chong, U. Ayesa, V. Prakash Daswani, and E. C. Hur, On Physical Properties of Tetraether Lipid Membranes: Effects of Cyclopentane Rings, Archaea 2012, 1 (2012).

(34) A. P. Thompson, H. M. Aktulga, R. Berger, D. S. Bolintineanu, W. M. Brown, P. S. Crozier, P. J. in ’t Veld, A. Kohlmeyer, S. G. Moore, T. D. Nguyen, R. Shan, M. J. Stevens, J. Tranchida, C. Trott, and S. J. Plimpton, LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales, Comput. Phys. Commun. 271, 108171 (2022).

(35) A. Stukowski, Visualization and analysis of atomistic simulation data with ovito–the open visualization tool, Modelling and Simulation in Materials Science and Engineering 18, 015012 (2009).

(36) E. R. May, A. Narang, and D. I. Kopelevich, Role of molecular tilt in thermal fluctuations of lipid membranes, Physical Review E 76, 021913 (2007).

(37) W. Helfrich, Elastic Properties of Lipid Bilayers: Theory and Possible Experiments, Z. Naturforsch. C 28, 693 (1973).

(38) M. Hamm and M. Kozlov, Elastic energy of tilt and bending of fluid membranes, Eur. Phys. J. E 3, 323 (2000).

(39) M. Deserno, Fluid lipid membranes: From differential geometry to curvature stresses, Chemistry and Physics of Lipids 185, 11 (2015).

(40) V. A. Harmandaris and M. Deserno, A novel method for measuring the bending rigidity of model lipid membranes by simulating tethers, The Journal of Chemical Physics 125, 204905 (2006).

(41) M. Hu, J. J. Briguglio, and M. Deserno, Determining the Gaussian Curvature Modulus of Lipid Membranes in Simulations, Biophys. J. 102, 1403 (2012).

(42) M. Deserno, Elastic deformation of a fluid membrane upon colloid binding, Phys. Rev. E 69, 031903 (2004), arXiv: cond-mat/0303656.

(43) K. S. Makarova, M. Y. Galperin, and E. V. Koonin, Comparative genomic analysis of evolutionarily conserved but functionally uncharacterized membrane proteins in archaea: Prediction of novel components of secretion, membrane remodeling and glycosylation systems, Biochimie 118, 302 (2015).

(44) A. Verchère, W.-L. Ou, B. Ploier, T. Morizumi, M. A. Goren, P. Bütikofer, O. P. Ernst, G. Khelashvili, and A. K. Menon, Light-independent phospholipid scramblase activity of bacteriorhodopsin from Halobacterium salinarum, Sci. Rep. 7, 9522 (2017).

(45) T. B. H. Schroeder, G. Leriche, T. Koyanagi, M. A. Johnson, K. N. Haengel, O. M. Eggenberger, C. L. Wang, Y. H. Kim, K. Diraviyam, D. Sept, J. Yang, and M. Mayer, Effects of lipid tethering in extremophile-inspired membranes on H(+)/OH(-) flux at room temperature, Biophys. J. 110, 2430 (2016).

(46) R. Xu, A. Dehghan, A.-C. Shi, and J. Zhou, Elastic property of membranes self-assembled from diblock and triblock copolymers, Chem. Phys. Lipids 221, 83 (2019).

(47) Z. Dogic and S. Fraden, Ordered phases of filamentous viruses, Curr. Opin. Colloid Interface Sci. 11, 47 (2006).

(48) E. Barry and Z. Dogic, Entropy driven self-assembly of nonamphiphilic colloidal membranes, Proc. Natl. Acad. Sci. U.S.A. 107, 10348 (2010).

(49) A. J. Balchunas, R. A. Cabanas, M. J. Zakhary, T. Gibaud, S. Fraden, P. Sharma, M. F. Hagan, and Z. Dogic, Equation of state of colloidal membranes, Soft Matter 15, 6791 (2019).

(50) M. Saracco, P. Schaeffer, M. Tourte, S.-V. Albers, Y. Louis, J. Peters, B. Demé, S. Fontanay, and P. M. Oger, Bilayer-Forming Lipids Enhance Archaeal Monolayer Membrane Stability, Int. J. Mol. Sci. 26, 3045 (2025).

(51) M. Amaral, archaeal_membranes : code and examples (2024), available at https://doi.org/10.5281/zenodo. 13934991.

(52) M. F. Ergüder and M. Deserno, Identifying systematic errors in a power spectral analysis of simulated lipid membranes, The Journal of Chemical Physics 154, 214103 (2021).

(53) J. Genova, N. Ulrih, V. Kralj-Iglič, A. Iglič, and I. Bivas, Bending Elasticity Modulus of Giant Vesicles Composed of Aeropyrum Pernix K1 Archaeal Lipid, Life 5, 1101 (2015).

(54) M. Amaral, Archaeal Membranes: In Silico Modelling and Design, Ph.D. thesis, Institute of Science and Technology Austria (2024).

(55) M. Pohlschroder, F. Pfeiffer, S. Schulze, and M. F. A. Halim, Archaeal cell surface biogenesis, FEMS Microbiol. Rev. 42, 694 (2018).

(56) K. S. Makarova, N. Yutin, S. D. Bell, and E. V. Koonin, Evolution of diverse cell division and vesicle formation systems in Archaea, Nat. Rev. Microbiol. 8, 731 (2010).

(57) C. W. Stairs and T. J. Ettema, The Archaeal Roots of the Eukaryotic Dynamic Actin Cytoskeleton, Curr. Biol. 30, R521 (2020).

(58) B. Baum and D. A. Baum, The merger that made us, BMC Biol. 18, 72 (2020).

(59) Z. Zeng, H. Chen, H. Yang, Y. Chen, W. Yang, X. Feng, H. Pei, and P. V. Welander, Identification of a protein responsible for the synthesis of archaeal membrane-spanning GDGT lipids, Nat. Commun. 13, 1545 (2022).

eLife Assessment

This compelling work describes how the cell cycle-regulating phosphatase subunit, RepoMan, is regulated by the oxygen-dependent, metabolite-sensing hydroxylase PHD1. The characterisation of how proline hydroxylation alters signalling at the molecular and cellular level provides important evidence to enhance our understanding of how 2-oxoglutarate-dependent dioxygenases influence the cell cycle and mitosis.

Reviewer #1 (Public review):

Summary:

The study by Druker et al. shows that siRNA depletion of PHD1, but not PHD2, increases H3T3 phosphorylation in cells arrested in prometaphase. Additionally, the expression of wild-type RepoMan, but not the RepoMan P604A mutant, restored normal H3T3 phosphorylation localization in cells arrested in prometaphase. Furthermore, the study demonstrates that expression of the RepoMan P604A mutant leads to defects in chromosome alignment and segregation, resulting in increased cell death. These data support a role for PHD1-mediated prolyl hydroxylation in controlling progression through mitosis. This occurs, at least in part, by hydroxylating RepoMan at P604, which regulates its interaction with PP2A during chromosome alignment.

Strengths:

The data support most of the conclusions made. However, some issues need to be addressed.

Weaknesses:

(1) Although ectopically expressed PHD1 interacts with ectopically expressed RepoMan, there is no evidence that endogenous PHD1 binds to endogenous RepoMan or that PHD1 directly binds to RepoMan.

(2) There is no genetic evidence indicating that PHD1 controls progression through mitosis by catalyzing the hydroxylation of RepoMan.

(3) Data demonstrating the correlation between dynamic changes in RepoMan hydroxylation and H3T3 phosphorylation throughout the cell cycle are needed.

(4) The authors should provide biochemical evidence of the difference in binding ability between RepoMan WT/PP2A and RepoMan P604A/PP2A.

(5) PHD2 is the primary proline hydroxylase in cells. Why does PHD1, but not PHD2, affect RepoMan hydroxylation and subsequent control of mitotic progression? The authors should discuss this issue further.

Reviewer #2 (Public review):

Summary:

This is a concise and interesting article on the role of PHD1-mediated proline hydroxylation of proline residue 604 on RepoMan and its impact on RepoMan-PP1 interactions with phosphatase PP2A-B56 complex leading to dephosphorylation of H3T3 on chromosomes during mitosis. Through biochemical and imaging tools, the authors delineate a key mechanism in the regulation of the progression of the cell cycle. The experiments performed are conclusive with well-designed controls.

Strengths:

The authors have utilized cutting-edge imaging and colocalization detection technologies to infer the conclusions in the manuscript.

Weaknesses:

Lack of in vitro reconstitution and binding data.

Reviewer #3 (Public review):

Summary:

The manuscript is a comprehensive molecular and cell biological characterisation of the effects of P604 hydroxylation by PHD1 on RepoMan, a regulatory subunit of the PPIgamma complex. The identification and molecular characterisation of the hydroxylation site have been written up and deposited in BioRxiv in a separate manuscript. I reviewed the data and came to the conclusion that the hydroxylation site has been identified and characterised to a very high standard by LC-MS, in cells and in vitro reactions. I conclude that we should have no question about the validity of the PHD1-mediated hydroxylation.

In the context of the presented manuscript, the authors postulate that hydroxylation on P604 by PHD1 leads to the inactivation of the complex, resulting in the retention of pThr3 in H3.

Strengths:

Compelling data, characterisation of how P604 hydroxylation is likely to induce the interaction between RepoMan and a phosphatase complex, resulting in loading of RepoMan on Chromatin. Loss of the regulation of the hydroxylation site by PHD1 results in mitotic defects.

Weaknesses:

Reliance on a Proline-Alanine mutation in RepoMan to mimic an unhydroxylatable protein. The mutation will introduce structural alterations, and inhibition or knockdown of PHD1 would be necessary to strengthen the data on how hydroxylates regulate chromatin loading and interactions with B56/PP2A.

Author response:

Public Reviews:

Reviewer #1 (Public review):

We appreciate the reviewer’s agreement that our data, "support most of the conclusions made”.

With respect to Concerns raised by reviewer 1:

(1) Although ectopically expressed PHD1 interacts with ectopically expressed RepoMan, there is no evidence that endogenous PHD1 binds to endogenous RepoMan or that PHD1 directly binds to RepoMan.

We do not fully agree that this comment is accurate - the implication is that we only show interaction between two exogenously expressed proteins, i.e. both exogenous  PHD1 and RepoMan, when in fact we show that tagged PHD1 interacts with endogenous RepoMan. The major technical challenge here is the well known difficulty of detetcing endogenous PHD1 in such cell lines. We agree that co-IP studies do not prove that this interaction is direct and never claim to have shown this, though we do feel that a direct interaction is most likely, albeit not proven.

(2) There is no genetic evidence indicating that PHD1 controls progression through mitosis by catalyzing the hydroxylation of RepoMan.

We agree that our current study is primarily a biochemical and cell biological study, rather than a genetic study. Nonetheless, similar biochemical and cellular approaches have been widely used and validated in previous studies in mechanisms regulating cell cycle progression and we are confident in the conclusions drawn based on the data obtained so far.

(3) Data demonstrating the correlation between dynamic changes in RepoMan hydroxylation and H3T3 phosphorylation throughout the cell cycle are needed.

We agree that it will be very interesting to analyse in more detail the cell cycle dynamics of RepoMan hydroxylation and H3T3 phosphorylation - along with other cell cycle parameters. We view this as outside the scope of our present study and are actively engaged in raising the additional funding needed to pursue such future experiments.

(4) The authors should provide biochemical evidence of the difference in binding ability between RepoMan WT/PP2A and RepoMan P604A/PP2A.

Here again we agree that it will be very interesting to analyse in future the detailed binding interactions between wt and mutant RepoMan and other interacting proteins, including PP2A. We view this as outside the scope of our present study and are actively engaged in raising the additional funding needed to pursue such future experiments.

(5) PHD2 is the primary proline hydroxylase in cells. Why does PHD1, but not PHD2, affect RepoMan hydroxylation and subsequent control of mitotic progression? The authors should discuss this issue further.

We agree with the main point underlining this comment, i.e., that there are still many things to be learned concerning the specific roles and mechanisms of the different PHD enzymes in vivo. We look forward to addressing these questions in future studies.

Reviewer #2 (Public review):

We appreciate the reviewer’s comments that our manuscript uses biochemical and imaging tools to delineate a key mechanism in the regulation of the progression of the cell cycle and their appreciation that our experiments performed are, 'conclusive with well-designed controls.'

With respect to the specific Concern raised by reviewer 2:

Lack of in vitro reconstitution and binding data.

We agree that it will be very interesting to pursue in vitro reconstitution studies and detailed binding data. We view this as outside the scope of our present study and are actively engaged in raising the additional funding needed to pursue such future experiments.

Reviewer #3 (Public review):

We appreciate the reviewer’s comments that our study, “is a comprehensive molecular and cell biological characterisation of the effects of P604 hydroxylation by PHD1 on RepoMan, a regulatory subunit of the PPIgamma complex” and their conclusion that, “we should have no question about the validity of the PHD1-mediated hydroxylation”.

With respect to the specific Concern raised by reviewer 3:

Reliance on a Proline-Alanine mutation in RepoMan to mimic an unhydroxylatable protein. The mutation will introduce structural alterations, and inhibition or knockdown of PHD1 would be necessary to strengthen the data on how hydroxylates regulate chromatin loading and interactions with B56/PP2A.

We do not agree that we rely solely on analysis of the single site pro-ala mutatin in RepoMan for our conclusions, since we also present a raft of additional experimental evidence, including knock-down data and experiments using both fumarate and FG. We would also reference the data we present on RepoMan in the parallel study by Jiang et al, which has also been reviewed by eLife and is currently available on biorxiv (doi: https://doi.org/10.1101/2025.05.06.652400). Of course we agree with the reviewer that even although the muatnt RepoMan features only a single amino acid change, this could still result in undetermined structural effects on the RepoMan protein that could conceivably contribute, at least in part, to some of the phenotypic effects observed. Hopefully future studies will help to clarify this.

eLife Assessment

This manuscript presents solid experimental data using Fmr1 knockout mice to explore the fundamental role of Fmr1 in sleep regulation. The study supports the hypothesis that scheduled feeding can improve circadian rhythm and behavior in a mouse model of Fragile X syndrome. These findings may offer new insights into neurodevelopmental disorders and their potential treatment strategies.

Reviewer #1 (Public review):

The authors conducted a comprehensive investigation into sleep and circadian rhythm disturbances in Fmr1 knockout (KO) mice, a model for Fragile X Syndrome (FXS). They began by monitoring daily home cage behaviors to identify disruptions in sleep and circadian patterns, then assessed the mice's adaptability to altered light conditions through photic suppression and skeleton photoperiod experiments. To uncover potential mechanisms, they examined the connectivity between the retina and the suprachiasmatic nucleus. The study also included an analysis of social behavior deficits in the mutant mice and tested whether scheduled feeding could alleviate these issues. Notably, scheduled feeding not only improved sleep, circadian, and social behaviors but also normalized plasma cytokine levels. The manuscript is strengthened by its focus on a significant and underexplored area-sleep deficits in an FXS model-and by its robust experimental design, which integrates a variety of methodological approaches to provide a thorough understanding of the observed phenomena and potential therapeutic avenues.

Reviewer #2 (Public review):

Summary:

In the present study, the authors, using a mouse model of Fragile X syndrome, explore the intriguing hypothesis that restricting food access over the daily schedule will improve sleep patterns and subsequently enhance behavioral capacities. By restricting food access from 12h to 6h over the nocturnal period (the active period for mice), they show, in these KO mice, an improvement in the sleep pattern accompanied by reduced systemic levels of inflammatory markers and improved behavior. These data, using a classical mouse model of neurodevelopmental disorder (NDD), suggest that modifying eating patterns might improve sleep quality, leading to reduced inflammation and enhanced cognitive/behavioral capacities in children with NDD.

Overall, the paper is well-written and easy to follow. The rationale of the study is generally well introduced. Data are globally sound. The interpretation is overall supported by the provided data.

Author response:

The following is the authors’ response to the previous reviews

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

Thank you for the extensive response to my comments and questions.

Reviewer #2 (Recommendations for the authors):

(1) The Fmr1/Fxr2 double KO mice are not well described in the Introduction.

We have changed the sentence in the introduction to clarify that in Zhang et al ., 2008 they used a mouse lacking both the Fmr1 gene and its paralog Fxr2.

(3) The Authors decided not to discuss the potential translation of the present study to human patients, despite their final conclusion statement.

The paragraph below has been added to the end of the discussion:

“Translational Implications”

The present findings support the view that circadian disruption is not merely a downstream consequence of disease processes but actively contributes to symptom expression. Hence, the possibility that interventions designed to reinforce circadian rhythms can hold therapeutic value for individuals with FXS and related neurodevelopmental conditions. Given that sleep and circadian dysfunction are detectable early in development and are predictive of more severe clinical phenotypes, circadian-based interventions may be particularly beneficial if applied during periods of heightened neural plasticity. Importantly, time-restricted feeding represents a relatively low-cost, non-invasive strategy that could be feasibly implemented in realworld settings. Further translational work is needed to evaluate whether the mechanistic links identified here—between circadian misalignment, immune dysregulation, and behavioral impairments—are conserved in humans, and similar approaches can be implemented for clinical use.

eLife Assessment

This study presents an important finding on the signaling mechanisms underlying Treg cell homeostasis by identifying the simultaneous requirement of diacylglycerol (DAG) kinases (DGK) alpha and zeta for Foxp3+ Treg cell function and follicular responses, with implications for the pathogenesis of some autoimmune diseases. Whereas data based on the characterization of double knock-out mice (for DGK alpha and zeta) is solid, showing the emergence of autoimmune manifestations, the study has gaps in its experimental approaches since it is not clear what can be attributed to the simultaneous DKGα and ζ deficiency, versus the individual deficiency of either one. Experiments on the pathogenic potential of the DKO Tregs in the absence of other T-cells were not presented and results on the role of CD25 downregulation and CD28-independent activation of Treg cells were not properly discussed. Nonetheless, the reported data would be of interest to immunologists working on T-cell intracellular signaling and autoimmunity.

Reviewer #1 (Public review):

Summary:

The manuscript by Li and colleagues describes the impact of deficiency on the DKGα and ζ on Treg cells and follicular responses. The experimental approach is based on the characterization of double KO mice that show the emergence of autoimmune manifestations that include the production of autoantibodies. Additionally, there is an increase in Tfh cells, but also Tfr cells in these mice deficient in both DKGα and ζ. Although the observations are interesting, the interpretation of the observations is difficult in the absence of data related to single mutations. While a supplementary figure shows that the autoimmune manifestations are more severe in the DKGα and ζ deficient mice, prior observations show that a single DKGα deficiency has an impact on Treg homeostasis. As such, the contribution of the two chains to the overall phenotype is hard to establish.

Strengths:

Well-conducted experiments with informative mouse models with defined genetic defects.

Weaknesses:

The major weakness is the lack of clarity concerning what can be attributed to simultaneous DKGα and ζ deficiency versus deficiency on DKGα or ζ alone. Technical concerns related to a number of figures were raised in the initial report and not adequately addressed by the authors in the revised manuscript.

In conclusion, the claims in the manuscript are not convincingly supported by the data,

Reviewer #2 (Public review):

Summary:

In this manuscript, Li et al investigates the combined role of diacylglycerol (DAG) kinases (DGK) a and z in Foxp3+ Treg cells function that prevent autoimmunity. The authors generated DGK a and z Treg-specific double knock out mice (DKO) by crossing Dgkalpha-/- mice to DgKzf/f and Foxp3YFPCre/+ mice. The resulting "DKO" mice thus lack DGK a in all cells and DGK and z in Foxp3+Treg cells. The authors show that the DKO mice spontaneously develop autoimmunity, characterized by multiorgan inflammatory infiltration and elevated anti double strand DNA (dsDNA), -single strand DNA (ssDNA), and -nuclear autoantibodies. The authors attribute the DKO mice phenotype to Foxp3+Treg dysfunction, including accelerated conversion into "exTreg" cells with pathogenic activity. Interestingly, the combined deficiency of DGK a and z seems to release Treg cell dependence on CD28-mediated costimulatory signals, which the authors show by crossing their DKO mice to CD28-/- mice (TKO mice), which also develop autoimmunity.

Strengths:

The phenotypes of the mutant mice described in the manuscript are striking, and the authors provide a comprehensive analysis of the functional processes alters by the lack of DGKs.

Weaknesses:

One aspect that could be better explored is the direct role of "ex-Tregs" in causing pathogenesis in the models utilized.

But overall, this is an important report that makes a significant addition to the understanding of DAG kinases to Treg cells biology.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public review):

Summary:

The manuscript by Li and colleagues describes the impact of deficiency on the DKGα and ζ on Treg cells and follicular responses. The experimental approach is based on the characterization of double KO mice that show the emergence of autoimmune manifestations that include the production of autoantibodies. Additionally, there is an increase in Tfh cells, but also Tfr cells in these mice deficient in both DKGα and ζ. Although the observations are interesting, the interpretation of the observations is difficult in the absence of data related to single mutations. While a supplementary figure shows that the autoimmune manifestations are more severe in the DKGα and ζ deficient mice, prior observations show that a single DKGα deficiency has an impact on Treg homeostasis. As such, the contribution of the two chains to the overall phenotype is hard to establish.

Strengths:

Well-conducted experiments with informative mouse models with defined genetic defects.

Weaknesses:

The major weakness is the lack of clarity concerning what can be attributed to simultaneous DKGα and ζ deficiency versus deficiency on DKGα or ζ alone.

Some interpretations are also not conclusively supported by data.

We appreciate the reviewer 1’s positive comments about our manuscript and for the suggestion to include DGKα‑ or DGKζ‑single‑knockout (SKO) Tregs for the mechanistical studies. Unfortunately, performing this sound simple but truly extensive experiment would exceed our current budget and personnel capacity. Importantly, it is well known that DGKα and DGKζ act redundantly or synergistically in T cells, with single loss producing minimal or partial phenotypes compared with the double knockout. The comprehensive mechanistic data already presented for DGKαζ‑DKO Tregs therefore capture the combined functional and mechanistical deficit that is most relevant to DGK functions in Treg biology, and they support the conclusions drawn in this manuscript. The reviewer also pointed out some interpretation issues such as CD25 down regulation in Tfr cells and some minor issues. We appreciate the reviewer’s expertise and have revised the text and discussion accordingly.

Reviewer #2 (Public review):

Summary:

In this manuscript, Li et al investigate the combined role of diacylglycerol (DAG) kinases (DGK) α and ζ in Foxp3+ Treg cells function that prevent autoimmunity. The authors generated DGK α and ζ Treg-specific double knockout mice (DKO) by crossing Dgkalpha-/- mice to DgKzf and Foxp3YFPCre/+ mice. The resulting "DKO" mice thus lack DGK α in all cells and DGK ζ in Foxp3+Treg cells. The authors show that the DKO mice spontaneously develop autoimmunity, characterized by multiorgan inflammatory infiltration and elevated anti-double-strand DNA (dsDNA), -single-strand DNA (ssDNA), and -nuclear autoantibodies. The authors attribute the DKO mice phenotype to Foxp3+Treg dysfunction, including accelerated conversion into "exTreg" cells with pathogenic activity. Interestingly, the combined deficiency of DGK α and ζ seems to release Treg cell dependence on CD28-mediated costimulatory signals, which the authors show by crossing their DKO mice to CD28-/- mice (TKO mice), which also develop autoimmunity.

Strengths:

The phenotypes of the mutant mice described in the manuscript are striking, and the authors provide a comprehensive analysis of the functional processes altered by the lack of DGKs.

Weaknesses:

One aspect that could be better explored is the direct role of "ex-Tregs" in causing pathogenesis in the models utilized.

However, overall, this is an important report that makes a significant addition to the understanding of DAG kinases in Treg cell biology.

We greatly appreciate reviewer 2’s positive comments about the manuscript. The data we presented in the manuscript show that DGKαζDKO Tregs but not WT Tregs are able to trigger autoimmunity in T cell deficient mice in the presence of WT CD4 T cells support that DGKαζDKO Tregs are pathogenic. Reviewer 2 suggested to test the direct role of DGKαζDKO Treg/ex-Tregs in the pathogenesis of autoimmune diseases in the absence of conventional T cells. This is really an interesting idea that we will test it in the future should recourse for executing the experiment become available.

eLife Assessment

This important study decoded target-associated information in prefrontal and sensory cortex during the preparatory period of a visual search task, suggesting a memory component of human subjects performing such visual attention task. The evidence supporting this claim is compelling, based on multivariate pattern analyses of fMRI data. The results will be of interest to psychologists and cognitive neuroscientists.

Reviewer #1 (Public review):

When you search for something, you need to maintain some representation (a "template") of that target in your mind/brain. Otherwise, how would you know what you were looking for? If your phone is in a shocking pink case, you can guide your attention to pink things based on a target template that includes the attribute 'pink'. That guidance should get you to the phone pretty effectively, if it is in view. Most real-world searches are more complicated. If you are looking for the toaster, you will make use of your knowledge of where toasters can be. Thus, if you are asked to find a toaster, you might first activate a template of a kitchen or a kitchen counter. You might worry about pulling up the toaster template only after you are reasonably sure you have restricted your attention to a sensible part of the scene.

Zhou and Geng are looking for evidence of this early stage of guidance by information about the surrounding scene in a search task. They train Os to associate four faces with four places. Then, with Os in the scanner, they show one face - the target for a subsequent search. After an 8 sec delay, they show a search display where the face is placed on the associated scene 75% of the time. Thus, attending to the associated scene is a good idea. The questions of interest are "When can the experimenters decode which face Os saw from fMRI recording?" "When can the experimenters decode the associated scene?" and "Where in the brain can the experimenters see evidence of this decoding? The answer is that the face but not the scene can be read out during the face's initial presentation. The key finding is that the scene can be read out (imperfectly but above chance) during the subsequent delay when Os are looking at just a fixation point. Apparently, seeing the face conjures up the scene in the mind's eye.

This is a solid and believable result. The only issue, for me, is whether it is telling us anything specifically about search. Suppose you trained Os on the face-scene pairing but never did anything connected to search. If you presented the face, would you not see evidence of recall of the associated scene? Maybe you would see the activation of the scene in different areas and you could identify some areas as search specific. I don't think anything like that was discussed here.

You might also expect this result to be asymmetric. The idea is that the big scene gives the search information about the little face. The face should activate the larger useful scene more than the scene should activate the more incidental face, if the task was reversed. That might be true if finding is related to search where the scene context is presumed to be the useful attention guiding stimulus. You might not expect an asymmetry if Os were just learning an association.

It is clear in this study that the face and the scene have been associated and that this can be seen in the fMRI data. It is also clear that a valid scene background speeds the behavioral response in the search task. The linkage between these two results is not entirely clear but perhaps future research will shed more light.

It is also possible that I missed the clear evidence of the search-specific nature of the activation by the scene during the delay period. If so, I apologize and suggest that the point be underlined for readers like me.

Comments on revised version:

I am satisfied with the revision.

Reviewer #2 (Public review):

Summary:

This work is one of the best instances of a well-controlled experiment and theoretically impactful findings within the literature on templates guiding attentional selection. I am a fan of the work that comes out of this lab and this particular manuscript is an excellent example as to why that is the case. Here, the authors use fMRI (employing MVPA) to test whether during the preparatory search period, a search template is invoked within the corresponding sensory regions, in the absence of physical stimulation. By associating faces with scenes, a strong association was created between two types of stimuli that recruit very specific neural processing regions - FFA for faces and PPA for scenes. The critical results showed that scene information that was associated with a particular cue could be decoded from PPA during the delay period. This result strongly supports invoking of a very specific attentional template.

Strengths:

There is so much to be impressed with in this report. The writing of the manuscript is incredibly clear. The experimental design is clever and innovative. The analysis is sophisticated and also innovative. The results are solid and convincing.

Weaknesses:

I only have a few weaknesses to point out.<br /> This point is not so much of a weakness, but a further test of the hypothesis put forward by the authors. The delay period was long - 8 seconds. It would be interesting to split the delay period into the first 4seconds and the last 4seconds and run the same decoding analyses. The hypothesis here is that semantic associations take time to evolve, and it would be great to show that decoding gets stronger in the second delay period as opposed to the period right after the cue. I think it would be a stronger test of the template hypothesis.

Typo in the abstract "curing" vs "during."

It is hard to know what to do with significant results in ROIs that are not motivated by specific hypotheses. However, for Figure 3, what are explanations for ROIs that show significant differences above and beyond the direct hypotheses set out by the authors?

Following the revision, I have no further comments or concerns.

Reviewer #3 (Public review):

The manuscript contains a carefully designed fMRI study, using MVPA patter analysis to investigate which high-level associate cortices contain target-related information to guide visual search. A special focus is hereby on so-called 'target-associated' information, that has previously been shown to help in guiding attention during visual search. For this purpose the author trained their participants and made them learn specific target-associations, in order to then test which brain regions may contain neural representations of those learnt associations. They found that at least some of the associations tested were encoded in prefrontal cortex during the cue and delay period.

The manuscript is very carefully prepared. As far as I can see, the statistical analyses are all sound and the results integrate well with previous findings.

I have no strong objections against the presented results and their interpretation.

The authors have addressed all my previous comments and questions in their revision of the text.

Author response:

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

Reviewer #1 (Public review):

When you search for something, you need to maintain some representation (a "template") of that target in your mind/brain. Otherwise, how would you know what you were looking for? If your phone is in a shocking pink case, you can guide your attention to pink things based on a target template that includes the attribute 'pink'. That guidance should get you to the phone pretty effectively if it is in view. Most real-world searches are more complicated. If you are looking for the toaster, you will make use of your knowledge of where toasters can be. Thus, if you are asked to find a toaster, you might first activate a template of a kitchen or a kitchen counter. You might worry about pulling up the toaster template only after you are reasonably sure you have restricted your attention to a sensible part of the scene.

Zhou and Geng are looking for evidence of this early stage of guidance by information about the surrounding scene in a search task. They train Os to associate four faces with four places. Then, with Os in the scanner, they show one face - the target for a subsequent search. After an 8 sec delay, they show a search display where the face is placed on the associated scene 75% of the time. Thus, attending to the associated scene is a good idea. The questions of interest are "When can the experimenters decode which face Os saw from fMRI recording?" "When can the experimenters decode the associated scene?" and "Where in the brain can the experimenters see evidence of this decoding? The answer is that the face but not the scene can be read out during the face's initial presentation. The key finding is that the scene can be read out (imperfectly but above chance) during the subsequent delay when Os are looking at just a fixation point. Apparently, seeing the face conjures up the scene in the mind's eye.

This is a solid and believable result. The only issue, for me, is whether it is telling us anything specifically about search. Suppose you trained Os on the face-scene pairing but never did anything connected to the search. If you presented the face, would you not see evidence of recall of the associated scene? Maybe you would see the activation of the scene in different areas and you could identify some areas as search specific. I don't think anything like that was discussed here.

You might also expect this result to be asymmetric. The idea is that the big scene gives the search information about the little face. The face should activate the larger useful scene more than the scene should activate the more incidental face, if the task was reversed. That might be true if the finding is related to a search where the scene context is presumed to be the useful attention guiding stimulus. You might not expect an asymmetry if Os were just learning an association.

It is clear in this study that the face and the scene have been associated and that this can be seen in the fMRI data. It is also clear that a valid scene background speeds the behavioral response in the search task. The linkage between these two results is not entirely clear but perhaps future research will shed more light.

It is also possible that I missed the clear evidence of the search-specific nature of the activation by the scene during the delay period. If so, I apologize and suggest that the point be underlined for readers like me.

We have added text related to this issue, particularly in the discussion (page 19, line 6), and have also added citations of studies in humans and non-human primates showing a causal relationship between preparatory activity in prefrontal and visual cortex and visual search performance (page 6, line 16).

Reviewer #2 (Public review):

Summary:

This work is one of the best instances of a well-controlled experiment and theoretically impactful findings within the literature on templates guiding attentional selection. I am a fan of the work that comes out of this lab and this particular manuscript is an excellent example as to why that is the case. Here, the authors use fMRI (employing MVPA) to test whether during the preparatory search period, a search template is invoked within the corresponding sensory regions, in the absence of physical stimulation. By associating faces with scenes, a strong association was created between two types of stimuli that recruit very specific neural processing regions - FFA for faces and PPA for scenes. The critical results showed that scene information that was associated with a particular cue could be decoded from PPA during the delay period. This result strongly supports the invoking of a very specific attentional template.

Strengths:

There is so much to be impressed with in this report. The writing of the manuscript is incredibly clear. The experimental design is clever and innovative. The analysis is sophisticated and also innovative. The results are solid and convincing.

Weaknesses:

I only have a few weaknesses to point out.<br /> This point is not so much of a weakness, but a further test of the hypothesis put forward by the authors. The delay period was long - 8 seconds. It would be interesting to split the delay period into the first 4seconds and the last 4seconds and run the same decoding analyses. The hypothesis here is that semantic associations take time to evolve, and it would be great to show that decoding gets stronger in the second delay period as opposed to the period right after the cue. I don't think this is necessary for publication, but I think it would be a stronger test of the template hypothesis.

We conducted the suggested analysis, and we did not find clear evidence of differences in decoding scene information between the earlier and later portions of the delay period. This may be due to insufficient power when the data are divided, individual differences in when preparatory activation is the strongest, or truly no difference in activation over the delay period. More details of this analysis can be found in the supplementary materials (page 12, line 16; Figure S1).

Type in the abstract "curing" vs "during."

Fixed.

It is hard to know what to do with significant results in ROIs that are not motivated by specific hypotheses. However, for Figure 3, what are the explanations for ROIs that show significant differences above and beyond the direct hypotheses set out by the authors?

We added reasoning for the other a priori ROIs in the introduction (page 4, line 26). There is substantial evidence suggesting that frontoparietal areas are involved in cognitive control, attentional control, and working memory. The ROIs we selected from frontal and parietal cortex are based on parcels within resting state networks defined by the s17-network atlases (Schaefer et al., 2018). The IFJ was defined by the HCP-MMP1 (Glasser et al., 2016). These regions are commonly used in studies of attention and cognitive control, and the exact ROIs selected are described in the section on “Regions of interest (ROI) definition”. While we have the strongest hypothesis for IFJ based on relatively recent work from the Desimone lab, the other ROIs in lateral frontal cortex and parietal cortex, are also well documented in similar studies, although the exact computation being done by these regions during tasks can be hard to differentiate with fMRI.\

Reviewer #3 (Public review):

The manuscript contains a carefully designed fMRI study, using MVPA pattern analysis to investigate which high-level associate cortices contain target-related information to guide visual search. A special focus is hereby on so-called 'target-associated' information, that has previously been shown to help in guiding attention during visual search. For this purpose the author trained their participants and made them learn specific target-associations, in order to then test which brain regions may contain neural representations of those learnt associations. They found that at least some of the associations tested were encoded in prefrontal cortex during the cue and delay period.

The manuscript is very carefully prepared. As far as I can see, the statistical analyses are all sound and the results integrate well with previous findings.

I have no strong objections against the presented results and their interpretation.

Reviewer #1 (Recommendations for the authors):

One bit of trivia. In the abstract, you should define IFJ on its first appearance in the text. You get to that a bit later.

Fixed.

Reviewer #2 (Recommendations for the authors):

I really don't have much to suggest, as I thought that this was a clearly written report that offered a clever paradigm and data that supported the conclusions. My only suggestion would be to split the delay period activity and test whether the strength of the template evolves over time. Even though fMRI is not the best tool for this, still you would predict stronger decoding in the second half of the delay period

Please see above for our response to the same comment.

Reviewer #3 (Recommendations for the authors):

I would just like to point out some minor aspects that might be worth improving before publishing this work.

Abstract: While in general, the writing is clear and concise, I felt that the abstract of the manuscript was particularly hard to follow, probably because the authors at some point re-arranged individual sentences. For example, they write in line 12 about 'the preparatory period', but explain only in the following sentence that the preparatory period ensues 'before search begins'. This made it a bit hard to follow the overall logic and I think could easily be fixed. 

We have addressed this comment and updated the abstract.

Also in the abstract: 'The CONTENTS of the template typically CONTAIN...' sounds weird, no? Also, 'information is used to modulate sensory processing in preparation for guiding attention during search' sounds like a very over-complicated description of attentional facilitation. I'm not convinced either whether the sequence is correct here. Is the information really used to (first) modulate sensory processing (which is a sort of definition of attention in itself) to (then) prepare the guidance of attention in visual search?

We have addressed this comment and updated the abstract.

The sentence in line 7, 'However, many behavioral studies have shown that target-associated information is used to guide attention,...' (and the following sentence) assumes that the reader is somewhat familiar with the term 'target-associations'. I'm afraid that, for a naive reader, this term may only become fully understandable once the idea is introduced a bit later when mentioning that participants of the study were trained on face-scene pairings. I think it could help to give some very short explanation of 'target-associations' already when it is first mentioned. The term 'statistically co-occurring object pairs', for example, could be of great help here.

Thank you for the suggestion. We have added it to the abstract.

page 2, line 22: 'prefrotnal'

Fixed.

page 2, line 24/25: 'information ... can SUPPLANT (?) ... information'. (That's also a somewhat unfortunate repetition of 'information')

Fixed.

page 4, line 23-25: 'Working memory representations in lateral prefrontal and parietal regions are engaged in cognitive control computations that ARE (?) task non-specific but essential to their functioning'

Fixed.

page 7, line 1: maybe a comma before 'suggesting'?

Fixed.

page 7, line 14-16: Something seems wrong with this sentence: 'The distractor face was a race-gender match, which we previously FOUND MADE (?) target discrimination difficult enough to make the scene useful for guiding attention'

We have addressed this comment and rewritten this part (now on page 7, line 18).

Results / Discussion sections:

In several figures, like in Fig3A, the three different IFJ regions, are grouped separately from the other frontal areas, which makes sense given the special role IFJ plays for representing task-related templates. However, IFJ is still part of PFC. I think it would be more correct to group the other frontal areas (like FEF vLPFC etc.) as 'Other Frontal' or even 'Other PFC'.

We have made the changes based on the reviewer’s suggestion.

In some of the Figures, e.g. Fig 3 and 5, I had the impression that the activation patterns of some conditions in vLPFC were rather close to the location of IFJ, which is just a bit posterior. I think I remember that functional localisers of IFJ can actually vary quite a bit in localisation (see e.g. in the Baldauf/Desimone paper). Also, I think it has been shown in the context of other regions, like the human FEF that its position when defined by localisation tasks is not always nicely and fully congruent with the respective labels in an atlas like the Glasser atlas. It might help to take this in consideration when discussing the results, particularly since the term vLPFC is a rather vague collection of several brain parcels and not a parcel name in the Glasser atlas. Some people might even argue that vLPFC in the broad sense contains IFJ, similar to how 'Frontal' contains IFJ (see above). How strong of a point do the authors want to make about activation in IFJ versus in vlPFC?

We have now added text discussing the inability to truly differentiate between subregions of IFJ and other parts of vLPFC in the methods section on ROIs (page 25, line 13) and in the discussion (page 18, line 25). However, one might think that it is even more surprising given the likely imprecision of ROI boundaries that we see distinct patterns between the subregions of IFG defined by Glasser HCP-MMP1 and the other vLPFC regions defined by the 17-network atlases. We do not wish to overstate the precision of IFJ regions, but note the ROI results within the context of the larger literature. We are sure that our findings will have to be reinterpreted when newer methods allow for better localization of functional subregions of the vLPFC in individuals.

Given that the authors nicely explain in the introduction how important templates are in visual search, and given that FEF has such an important role in serially guiding saccades through visual search templates, I think it would be worth discussing the finding that FEF did not hold representation of these targets. Of course, this could be in part due to the specific task at hand, but it may still be interesting to note in the Discussion section that here FEF, although important for some top-down attention signals, did not keep representations of the 'search' templates. Is it because there is no spatial component to the task at hand (like proposed in Bedini 2021)?

We have now added text directly addressing this point and citing the Bedini et al. paper in the discussion (page 18, line 18). Besides our current findings, the relationship between IFJ and FEF is really interesting and will hopefully be investigated more in the future.

Page 18, line 5: 'we the(N) associated...'

Fixed.

eLife Assessment

This manuscript by Li, Lu et al., presents important findings on the role of cDC1 in atherosclerosis and their influence on the adaptive immune system. Using Xcr1Cre-Gfp Rosa26LSL-DTA ApoE-/- mouse models, these data convincingly reveal an unexpected, non-redundant role of the XCL1-XCR1 axis in mediating cDC1 contributions to atherosclerosis.

Reviewer #1 (Public review):

Summary:

In this study by Li et al., the authors re-investigated the role of cDC1 for atherosclerosis progression using the ApoE model. First, the authors confirmed the accumulation of cDC1 in atherosclerotic lesions in mice and humans. Then in order to examine the functional relevance of this cell type, the authors developed a new mouse model to selectively target cDC1. Specifically, they inserted the Cre recombinase directly after the start codon of endogenous XCR1 gene, thereby avoiding off-target activity. Following validation of this model, the authors crossed it with ApoE-deficient mice and found a striking reduction of aortic lesions (numbers and size) following high fat diet. The authors further characterized the impact of cDC1 depletion on lesional T cells and their activation state. Also, they provide in-depth transcriptomic analyses of lesional in comparison to splenic and nodal cDC1. These results imply cellular interactions between lesion T cells and cDC1. Finally, the authors show that the chemokine XCL1, which is produced by activated CD8 T cells (and NK cells) plays a key role for the interaction with XCR1-expressing cDC1 and particularly for the atherosclerotic disease progression.

Strengths:

The surprising results on XCL1 represent a very important gain in knowledge. The role of cDC1 is clarified with a new genetic mouse model.

Comments on revised version:

The authors have addressed my concerns in the revised version of this manuscript.

Reviewer #2 (Public review):

This study investigates the role of cDC1 in atherosclerosis progression using Xcr1Cre-Gfp Rosa26LSL-DTA ApoE-/- mice. The authors demonstrate that selective depletion of cDC1 reduces atherosclerotic lesions in hyperlipidemic mice. While cDC1 depletion did not alter macrophage populations, it suppressed T cell activation (both CD4+ and CD8+ subsets) within aortic plaques. Further, targeting the chemokine Xcl1 (ligand of Xcr1) effectively inhibits atherosclerosis. The manuscript is well-written, and data are clearly presented. The data provided in the article can well support the author's conclusion.

Comments on revised version:

The authors have addressed all previous concerns and made appropriate revisions to the data. I have no further questions.

Author response:

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

Reviewer #1 (Public review):

Summary:

In this study by Li et al., the authors re-investigated the role of cDC1 for atherosclerosis progression using the ApoE model. First, the authors confirmed the accumulation of cDC1 in atherosclerotic lesions in mice and humans. Then, in order to examine the functional relevance of this cell type, the authors developed a new mouse model to selectively target cDC1. Specifically, they inserted the Cre recombinase directly after the start codon of the endogenous XCR1 gene, thereby avoiding off-target activity. Following validation of this model, the authors crossed it with ApoE-deficient mice and found a striking reduction of aortic lesions (numbers and size) following a high-fat diet. The authors further characterized the impact of cDC1 depletion on lesional T cells and their activation state. Also, they provide in-depth transcriptomic analyses of lesional in comparison to splenic and nodal cDC1. These results imply cellular interactions between lesion T cells and cDC1. Finally, the authors show that the chemokine XCL1, which is produced by activated CD8 T cells (and NK cells), plays a key role in the interaction with XCR1-expressing cDC1 and particularly in the atherosclerotic disease progression.<br /> Strengths:

The surprising results on XCL1 represent a very important gain in knowledge. The role of cDC1 is clarified with a new genetic mouse model.

Thank you

Weaknesses:

My criticism is limited to the analysis of the scRNAseq data of the cDC1. I think it would be important to match these data with published data sets on cDC1. In particular, the data set by Sophie Janssen's group on splenic cDC1 might be helpful here (PMID: 37172103; https://www.single-cell.be/spleen_cDC_homeostatic_maturation/datasets/cdc1). It would be good to assign a cluster based on the categories used there (early/late, immature/mature, at least for splenic DC).

Thank you very much for your help. Using the scRNA seq data of Xcr1⁺ cDC1 sorted from ApoE^–/– mice, we re-annotated the populations, following the methodology proposed by Sophie Janssen's group. These results are presented in Figure S9 and Figure S10 and described in detail in the Results and Discussion section.

Please refer to the Results section from line 264 to 284: “Using the scRNA seq data of Xcr1⁺ cDC1 sorted from hyperlipidemic mice, we annotated the 10 populations as shown in Figure S9A, following the methodology from a previous study [41]. Ccr7⁺ mature cDC1s (Cluster 3, 7 and 9) and Ccr7- immature cDC1s (remaining clusters) were identified across cDC1 cells sorted from aorta, spleen and lymph nodes (Figure S9B). Further stratification based on marker genes reveals that Cluster 10 is the pre-cDC1, with high expression level of CD62L (Sell) and low expression level of CD8a (Figure S9C). Cluster 6 and 8 are the proliferating cDC1s, which express high level of cell cycling genes Stmn1 and Top2a (Figure S9D). Cluster 1 and 4 are early immature cDC1s, and cluster 2 and 5 are late immature cDC1s, according to the expression pattern of Itgae, Nr4a2 (Figure S9E). Cluster 9 cells are early mature cDC1s, with elevated expression of Cxcl9 and Cxcl10 (Figure S9F). Cluster 3 and 7 as late mature cDC1s, characterized by the expression of Cd63 and Fscn1 (Figure S9G). As shown in Figure 5C and Figure S9, the 10 populations displayed a major difference of aortic cDC1 cells that lack in pre-cDC1s (cluster 10) and mature cells (cluster 3, 7 and 9). Interestingly, in hyperlipidemic mice splenic cDC1 possess only Cluster 3 as the late mature cells while the lymph node cDC1 cells have two late mature populations namely Cluster 3 and Cluster 7. In further analysis, we also compared splenic cDC1 cells from HFD mice to those from ND mice. As shown in Figure S10, HFD appears to impact early immature cDC1-1 cells (Cluster 1) and increases the abundance of late immature cDC1 cells (Cluster 2 and 5), regardless of the fact that all 10 populations are present in two origins of samples. We also found that Tnfaip3 and Serinc3 are among the most upregulated genes, while Apol7c and Tifab are downregulated in splenic cDC1 cells sorted from HFD mice”.  

Please refer to the Discussion section from line 380 to 385: “Based on the maturation analysis of the cDC1 scRNA seq data [41], our findings suggest that the aortic cDC1 cells display a major difference from those of spleen and lymph nodes by lacking the mature clusters, whereas lymph node cDC1 cells contain an additional Fabp5⁺ S100a4⁺ late mature Cluster. Our results also suggest that hyperlipidemia contributes to alteration in early immature cDC1 and in the abundance of late immature cDC1 cells, which was associated with dramatic change in gene expression of Tnfaip3, Serinc3, Apol7c and Tifab”.

Reviewer #2 (Public review):

This study investigates the role of cDC1 in atherosclerosis progression using Xcr1Cre-Gfp Rosa26LSL-DTA ApoE-/- mice. The authors demonstrate that selective depletion of cDC1 reduces atherosclerotic lesions in hyperlipidemic mice. While cDC1 depletion did not alter macrophage populations, it suppressed T cell activation (both CD4+ and CD8+ subsets) within aortic plaques. Further, targeting the chemokine Xcl1 (ligand of Xcr1) effectively inhibits atherosclerosis. The manuscript is well-written, and the data are clearly presented. However, several points require clarification:

(1) In Figure 1C (upper plot), it is not clear what the Xcr1 single-positive region in the aortic root represents, or whether this is caused by unspecific staining. So I wonder whether Xcr1 single-positive staining can reliably represent cDC1. For accurate cDC1 gating in Figure 1E, Xcr1+CD11c+ co-staining should be used instead.

The observed false-positive signal in the wavy structures within immunofluorescence Figure 1C (upper panel) results from the strong autofluorescence of elastic fibers, a major vascular wall component (alongside collagen). This intrinsic property of elastic fibers is a well-documented confounder in immunofluorescence studies [A, B].

In contrast, immunohistochemistry (IHC) employs an enzymatic chromogenic reaction (HRP with DAB substrate) that generates a brown precipitate exclusively at antigen-antibody binding sites. Importantly, vascular elastic fibers lack endogenous enzymatic activity capable of catalyzing the DAB reaction, thereby preventing this source of false positivity in IHC.

Given that Xcr1 is exclusively expressed on conventional type 1 dendritic cells [C], and considering that IHC lacks the multiplexing capability inherent to immunofluorescence for antigen co-localization, single-positive Xcr1 staining reliably identifies cDC1s in IHC results.

[A] König, K et al. “Multiphoton autofluorescence imaging of intratissue elastic fibers.” Biomaterials vol. 26,5 (2005): 495-500. doi:10.1016/j.biomaterials.2004.02.059

[B] Andreasson, Anne-Christine et al. “Confocal scanning laser microscopy measurements of atherosclerotic lesions in mice aorta. A fast evaluation method for volume determinations.” Atherosclerosis vol. 179,1 (2005): 35-42. doi:10.1016/j.atherosclerosis.2004.10.040

[C] Dorner, Brigitte G et al. “Selective expression of the chemokine receptor XCR1 on cross-presenting dendritic cells determines cooperation with CD8+ T cells.” Immunity vol. 31,5 (2009): 823-33. doi:10.1016/j.immuni.2009.08.027

(2) Figure 4D suggests that cDC1 depletion does not affect CD4+/CD8+ T cells. However, only the proportion of these subsets within total T cells is shown. To fully interpret effects, the authors should provide:

(a) Absolute numbers of total T cells in aortas.

(b) Absolute counts of CD4+ and CD8+ T cells.

Thanks for your suggestions. We agree that assessing both proportions and absolute numbers in Figure 4 provides a more complete picture of the effects of cDC1 depletion on T cell populations. Furthermore, we also add the absolute count of cDC1 cells and total T cells, and CD44 MFI (mean fluorescence intensity) in CD4⁺ and CD8⁺ T cells in Figure 4, and supplemented corresponding textual descriptions in the revised manuscript.

Please refer to the Results section from line 183 to 187: “Subsequently, we assessed T cell phenotype in the two groups of mice. While neither the frequencies nor absolute counts of aortic CD4⁺ and CD8⁺ T cells differed significantly between two groups of mice (Figure 4D-F), CD69 frequency and CD44 MFI (Mean Fluorescence Intensity), the T cell activation markers, were significantly reduced in both CD4⁺ and CD8⁺ T cells from Xcr1⁺ cDC1 depleted mice compared to controls (Figure 4G and H)”.

(3) How does T cell activation mechanistically influence atherosclerosis progression? Why was CD69 selected as the sole activation marker? Were other markers (e.g., KLRG1, ICOS, CD44) examined to confirm activation status?

We sincerely appreciate these insightful comments. As extensively documented in the literature, activated effector T cells (both CD4+ and CD8+) critically promote plaque inflammation and instability through their production of pro-inflammatory cytokines (particularly IFN-γ and TNF-α), which drive endothelial activation, exacerbate macrophage inflammatory responses, and impair smooth muscle cell function [A].

In our study, we specifically investigated the role of cDC1 cells in atherosclerosis progression. Our key findings demonstrate that cDC1 depletion attenuates T cell activation (as shown by reduced CD69/CD44 expression) and that this reduction in activation is functionally linked to the observed decrease in atherosclerosis burden in our model. 

Regarding CD44 as an activation marker, we performed quantitative analyses of CD44 mean fluorescence intensity (MFI) in aortic T cells (Figure 4). Importantly, the MFI of CD44 was significantly lower on both CD4+ and CD8+ T cells from Xcr1^Cre-Gfp Rosa26^LSL-DTA ApoE^–/– mice compared to the control ApoE^–/– mice (data shown below), which is consistent with the result of CD69 in Figure 4. We added the related description in the Result section.

Please refer to the Results section from line 185 to 187 “CD69 frequency and CD44 MFI (Mean Fluorescence Intensity), the T cell activation markers, were significantly reduced in both CD4+ and CD8+ T cells from Xcr1+ cDC1 depleted mice compared to controls (Figure 4G and H)”.

Similarly, MFI of CD44 was significantly lower on both CD4⁺ and CD8⁺ T cells from Xcl1^–/– ApoE^–/– mice compared to the control ApoE^–/– mice (data shown below), which is consistent with the result of CD69 in Figure 7. We also added the related description in the Result section.

Please refer to the Results section from line 308 to 309 “Crucially, CD69⁺ frequency and CD44 MFI remained comparable in both aortic CD4⁺ and CD8⁺ T cells between two groups (Figure 7D-F).”

[A] Hansson, Göran K, and Andreas Hermansson. “The immune system in atherosclerosis.” Nature immunology vol. 12,3 (2011): 204-12. doi:10.1038/ni.2001

(4) Figure 7B: Beyond cDC1/2 proportions within cDCs, please report absolute counts of: Total cDCs, cDC1, and cDC2 subsets. Figure 7D: In addition to CD4+/CD8+ T cell proportions, the following should be included:

(a) Total T cell numbers in aortas

(b) Absolute counts of CD4+ and CD8+ T cells.

Thanks for your suggestions. We have now included in Figure 7 the absolute counts of cDC, cDC1, and cDC2 cells, along with CD4⁺ and CD8⁺ T cells in aortic tissues. Additionally, we provide the corresponding CD44 mean fluorescence intensity (MFI) measurements for both CD4⁺ and CD8⁺ T cell populations. We added the related description in the Result section.

Please refer to the Results section from line 303 to 311: “The flow cytometric results illustrated that both frequencies and absolute counts of Xcr1⁺ cDC1 cells in the aorta were significantly reduced, but cDCs and cDC2 cells from Xcl1^–/– ApoE^–/– were comparable with that from ApoE^–/– (Figure 7A-C). Moreover, in both lymph node and spleen, the absolute numbers of pDC, cDC1 and cDC2 from Xcl1^–/– ApoE^–/– were comparable with that from ApoE^–/– (Figure S11). Crucially, CD69⁺ frequency and CD44 MFI remained comparable in both aortic CD4⁺ and CD8⁺ T cells between two groups (Figure 7D-F). However, aortic CD8⁺ T cells exhibited reduced frequency and absolute count, while CD4⁺ T cells showed increased frequency but unchanged counts in Xcl1^–/– ApoE^–/– mouse versus controls (Figure 7G and H).”

(5) cDC1 depletion reduced CD69+CD4+ and CD69+CD8+ T cells, whereas Xcl1 depletion decreased Xcr1+ cDC1 cells without altering activated T cells. How do the authors explain these different results? This discrepancy needs explanation.

We sincerely appreciate your professional and insightful comments regarding the mechanistic relationship between cDC1 depletion and T cell activation. Direct cDC1 depletion in the Xcr1^Cre-Gfp Rosa26^LSL-DTA ApoE^–/– micmodel removes both recruited and tissue-resident cDC1s, eliminating their multifunctional roles in antigen presentation, co-stimulation and cytokine secretion essential for T cell activation. In contrast, Xcl1 depletion reduces, but does not eliminate cDC1 migration into plaques. Furthermore, alternative chemokine axes (e.g., CCL5/CCR5, CXCL9/CXCR3, BCL9/BCL9L) may partially rescue cDC1 recruitment [13, 68, 69], and non-cDC1 APCs (e.g., monocytes, cDC2s) may compensate for T cell activation [55, 70]. We emphasize that Xcl1 depletion specifically failed to alter T cell activation in hyperlipidemic ApoE^–/– mice. However, its impact may differ in other pathophysiological contexts due to compensatory mechanisms. We thank you again for highlighting this nuance, which strengthens our mechanistic interpretation. We have added these points to the discussion section and included new references.

Please refer to the Discussion section from line 407 to 413: “Notably, while complete ablation of Xcr1⁺ cDC1s impaired T cell activation, reduction of Xcr1⁺ cDC1 recruitment via Xcl1 deletion did not significantly compromise this process. This discrepancy may arise through compensatory mechanisms: alternative chemokine axes (e.g., CCL5/CCR5, CXCL9/CXCR3, BCL9/BCL9L) may partially rescue Xcr1⁺ cDC1 homing [13, 68, 69], while non-cDC1 antigen-presenting cells (e.g., monocytes, cDC2s) may sustain T cell activation [55, 70]. Furthermore, tissue-specific microenvironment factors could potentially modulate its role in other diseases.”. [13] Eisenbarth, S C. “Dendritic cell subsets in T cell programming: location dictates function.” Nature reviews. Immunology vol. 19,2 (2019): 89-103. doi:10.1038/s41577-018-0088-1 [55] Brewitz, Anna et al. “CD8+ T Cells Orchestrate pDC-XCR1+ Dendritic Cell Spatial and Functional Cooperativity to Optimize Priming.” Immunity vol. 46,2 (2017): 205-219. doi:10.1016/j.immuni.2017.01.003 [68] de Oliveira, Carine Ervolino et al. “CCR5-Dependent Homing of T Regulatory Cells to the Tumor Microenvironment Contributes to Skin Squamous Cell Carcinoma Development.” Molecular cancer therapeutics vol. 16,12 (2017): 2871-2880. doi:10.1158/1535-7163.MCT-17-0341.[69] He F, Wu Z, Liu C, Zhu Y, Zhou Y, Tian E, et al. Targeting BCL9/BCL9L enhances antigen presentation by promoting conventional type 1 dendritic cell (cDC1) activation and tumor infiltration. Signal Transduct Target Ther. 2024;9(1):139. Epub 2024/05/30. doi: 10.1038/s41392-024-01838-9. PubMed PMID: 38811552; PubMed Central PMCID: PMCPMC11137111.[70] Böttcher, Jan P et al. “Functional classification of memory CD8(+) T cells by CX3CR1 expression.” Nature communications vol. 6 8306. 25 Sep. 2015, doi:10.1038/ncomms9306.

Reviewer #1 (Recommendations for the authors):

(1) Line 32 - The authors might want to add that the mouse model leads to a "constitutive" depletion of cDC1.

Thanks for your advice, we have revised the sentence as follows.

Please refer to the Results section from line 31 to 33: “we established Xcr1^Cre-Gfp Rosa26^LSL-DTA ApoE^–/– mice, a novel and complex genetic model, in which cDC1 was constitutively depleted in vivo during atherosclerosis development”.

(2) Line 187-188: The authors claim that T cell activation was "inhibited" if cDC1 was depleted. The data shows that the T cells were less activated, but there is no indication of any kind of inhibition; this should be corrected.

Thanks for your advice, we have revised the sentence as follows.

Please refer to the Results section from line 183 to 187: “Subsequently, we assessed T cell phenotype in the two groups of mice. While neither the frequencies nor absolute counts of aortic CD4⁺ and CD8⁺ T cells differed significantly between two groups of mice (Figure 4D-F), CD69 frequency and CD44 MFI (Mean Fluorescence Intensity), the T cell activation markers, were significantly reduced in both CD4⁺ and CD8⁺ T cells from Xcr1⁺ cDC1 depleted mice compared to controls (Figure 4G and H)”.

(3) Why are some splenic DC clusters absent in LNs and vice versa? This is not obvious to this reviewer and should at least be discussed.

We appreciate the insightful question regarding the absence of certain splenic DC clusters in LNs. This phenomenon in Figure 5 aligns with the 'division of labor' paradigm in dendritic cell biology: tissue microenvironments evolve specialized DC subsets to address local immunological challenges. The absence of universal clusters reflects functional adaptation, not technical artifacts. We acknowledge that this tissue-specific heterogeneity warrants further discussion and have expanded our analysis to address this point in the discussion part of our manuscript.

Please refer to the Discussion section from line 375 to 385: “This pronounced tissue-specific compartmentalization of Xcr1⁺ cDC1 subsets may related to multiple mechanisms including developmental imprinting that instructs precursor differentiation into transcriptionally distinct subpopulations [62], and microenvironmental filtering through organ-specific chemokine axes (e.g., CCL2/CCR2 in spleen) selectively recruits receptor-matched subsets [63, 64]. This spatial specialization optimizes pathogen surveillance for local immunological challenges. Based on the maturation analysis of the cDC1 scRNA seq data [41], our findings suggest that the aortic cDC1 cells display a major difference from those of spleen and lymph nodes by lacking the mature clusters, whereas lymph node cDC1 cells contain an additional Fabp5⁺ S100a4⁺ late mature Cluster. Our results also suggest that hyperlipidemia contributes to alteration in early immature cDC1 and in the abundance of late immature cDC1 cells, which was associated with dramatic change in gene expression of Tnfaip3, Serinc3, Apol7c and Tifab”.

[62]. Liu Z, Gu Y, Chakarov S, Bleriot C, Kwok I, Chen X, et al. Fate Mapping via Ms4a3-Expression History Traces Monocyte-Derived Cells. Cell. 2019;178(6):1509-25 e19. Epub 2019/09/07. doi: 10.1016/j.cell.2019.08.009. PubMed PMID: 31491389.

[63]. Bosmans LA, van Tiel CM, Aarts S, Willemsen L, Baardman J, van Os BW, et al. Myeloid CD40 deficiency reduces atherosclerosis by impairing macrophages' transition into a pro-inflammatory state. Cardiovasc Res. 2023;119(5):1146-60. Epub 2022/05/20. doi: 10.1093/cvr/cvac084. PubMed PMID: 35587037; PubMed Central PMCID: PMCPMC10202633.

[64]. Mildner A, Schonheit J, Giladi A, David E, Lara-Astiaso D, Lorenzo-Vivas E, et al. Genomic Characterization of Murine Monocytes Reveals C/EBPbeta Transcription Factor Dependence of Ly6C(-) Cells. Immunity. 2017;46(5):849-62 e7. Epub 2017/05/18. doi: 10.1016/j.immuni.2017.04.018. PubMed PMID: 28514690.

[41]. Bosteels V, Marechal S, De Nolf C, Rennen S, Maelfait J, Tavernier SJ, et al. LXR signaling controls homeostatic dendritic cell maturation. Sci Immunol. 2023;8(83):eadd3955. Epub 2023/05/12. doi: 10.1126/sciimmunol.add3955. PubMed PMID: 37172103.

(4) The authors should discuss how XCL1 could impact lesional cDC1 and T cell abundance. Notably, preDCs do not express XCR1, and T cells express XCL1 following TCR activation. Is there a recruitment or local proliferation defect of cDC1 in the absence of XCL1? Could there also be a role for NK cells as a potential source of XCL1?

We appreciate your insightful questions regarding the differential effects of Xcl1 on cDC1s and T cells. Xcl1 primarily mediates the recruitment of mature cDC1s. Our data demonstrate that Xcl1 deletion significantly reduces aortic cDC1 abundance, which correlates with a concomitant decrease in CD8⁺ T cell numbers within the aorta. These findings strongly suggest that the Xcl1-Xcr1 axis plays a regulatory role in T cell accumulation in aortic plaques.

Consistent with prior studies [A, B], cDC1 recruitment can occur in the absence of Xcl1 which echoes our findings that cDC1 cells were still found in Xcl1 knockout aortic plaque but in lower abundance. It is very true that further studies are required to address how the Xcl1 dependent and independent cDC1 cells activate T cells and if they possess capability of proliferation in tissue differentially. We have added these points in discussion section.

Please refer to the Discussion section from line 407 to 415: “Notably, while complete ablation of Xcr1⁺ cDC1s impaired T cell activation, reduction of Xcr1⁺ cDC1 recruitment via Xcl1 deletion did not significantly compromise this process. This discrepancy may arise through compensatory mechanisms: alternative chemokine axes (e.g., CCL5/CCR5, CXCL9/CXCR3, BCL9/BCL9L) may partially rescue Xcr1⁺ cDC1 homing [13, 68, 69], while non-cDC1 antigen-presenting cells (e.g., monocytes, cDC2s) may sustain T cell activation [55, 70]. Furthermore, tissue-specific microenvironment factors could potentially modulate its role in other diseases. In summary, our findings identify Xcl1 as a potential therapeutic target for atherosclerosis therapy, though its cellular origins and regulation of lesional Xcr1⁺ cDC1 and T cells dynamics require further studies”.

In literatures, Xcl1 are expressed in NK cells and subsects of T cells, and NK cells can be a potential source of Xcl1 during atherosclerosis which deserve further investigations [A, C, D].

[A] Böttcher, Jan P et al. “NK Cells Stimulate Recruitment of cDC1 into the Tumor Microenvironment Promoting Cancer Immune Control.” Cell vol. 172,5 (2018): 1022-1037.e14. doi:10.1016/j.cell.2018.01.004

[B] He, Fenglian et al. “Targeting BCL9/BCL9L enhances antigen presentation by promoting conventional type 1 dendritic cell (cDC1) activation and tumor infiltration.” Signal transduction and targeted therapy vol. 9,1 139. 29 May. 2024, doi:10.1038/s41392-024-01838-9

[C] Woo, Yeon Duk et al. “The invariant natural killer T cell-mediated chemokine X-C motif chemokine ligand 1-X-C motif chemokine receptor 1 axis promotes allergic airway hyperresponsiveness by recruiting CD103+ dendritic cells.” The Journal of allergy and clinical immunology vol. 142,6 (2018): 1781-1792.e12. doi:10.1016/j.jaci.2017.12.1005

[D] Winkels, Holger et al. “Atlas of the Immune Cell Repertoire in Mouse Atherosclerosis Defined by Single-Cell RNA-Sequencing and Mass Cytometry.” Circulation research vol. 122,12 (2018): 1675-1688. doi:10.1161/CIRCRESAHA.117.312513

Reviewer #2 (Recommendations for the authors):

There is a logical error in line 298. I suggest revising to: "Collectively, these data suggest that Xcl1 promotes atherosclerosis by recruiting Xcr1+ cDC1 cells, which subsequently drive T cell activation in lesions."

Thanks for your advice. Since Xcl1 deficiency reduced both the frequencies and absolute counts of Xcr1+ cDC1 and CD8+ T cells in lesions without affecting T cell activation, we revised the sentence as you suggested.

Please refer to the Results section from line 314 to 315: “Collectively, these data suggest that Xcl1 promotes atherosclerosis by recruiting Xcr1⁺ cDC1 cells, and facilitating CD8⁺ T cell accumulation in lesions”.

eLife Assessment

This important study elucidates the molecular function of the SARS-CoV-2 helicase NSP13, which inhibits the transcriptional activity of the YAP/TEAD complex in vitro and in vivo. The evidence supporting the authors' claims is compelling, based on cell biological assays and multi-omic studies. This work contributes to the understanding of the new regulatory mechanism of YAP/TEAD after SARS-CoV-2 infection and will be of interest to researchers investigating COVID-19 infection and the Hippo-YAP signaling pathway.

Reviewer #1 (Public review):

In the revised manuscript, Meng et al. report that SARS-CoV-2 infection suppresses YAP target gene transcription in both patient lung samples and iPSC-derived cardiomyocytes. Among the tested viral proteins, the helicase nonstructural protein 13 (NSP13) was identified as a key factor that impairs YAP/TEAD transcriptional activity. Through mutagenesis and protein-protein interaction studies, the authors propose a mechanism where NSP13 binds YAP/TEAD complex, remodels chromatin structure, and recruits transcriptional repressors to inhibit YAP/TEAD's transcriptional activity.

Overall, this study uncovers a novel regulation of Hippo signaling by SARS-CoV-2 through NSP13, suggesting a potential role of this growth-related pathway in host innate immune response to viral infection. While these findings are intriguing, future studies are needed to validate the involvement of YAP/TEAD in patient tissues and to assess their potential as therapeutic targets against SARS-CoV-2.

Reviewer #2 (Public review):

Summary:

The manuscript by Meng et al. describes a role for the coronavirus helicase NSP13 in the regulation of YAP-TEAD-mediated transcription. The authors present data that NSP13 expression in cells reduces YAP-induced TEAD luciferase reporter activity and that NSP13 transduction in cardiomyocytes blocks hyperactive YAP-mutant phenotypes in vivo. Mechanisms by which viral proteins (particularly those from coronaviruses) intersect with cellular signaling events is an important research topic, and the intersection of NSP13 with YAP-TEAD transcriptional activity (independent of upstream Hippo pathway mediated signals) offers new knowledge that is of interest to a broad range of researchers.

Strengths:

The manuscript presents convincing data mapping the effects of NSP13 on YAP-TEAD reporter activity to the helicase domain. Moreover, the in vivo data demonstrating that NSP13 expression in YAP5SA mouse cardiomyocytes increased survival animal rates, and restored cardiac function is striking and is supportive of the model presented.

Weaknesses:

While there are some hints at the mechanisms by which NSP13 regulates YAP-TEAD activity through the identification of NSP13-associated proteins by mass spec, the relationships and functions of these factors in the context of YAP-TEAD regulation requires further study in the future.

Author response:

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

Reviewer #1 (Public Review):

Major points

(1) The authors discovered a novel regulation of the Hippo-YAP pathway by SARS-CoV-2 infection but did not address the pathological significance of this finding. It remains unclear why YAP downstream gene transcription needs to be inhibited in response to SARS-CoV-2 infection. Is this inhibition crucial for the innate immune response to SARS-CoV-2? The authors should re-analyze their snRNA-seq and bulk RNA-seq data described in Figure 1 to determine whether any of the affected YAP downstream genes are involved in this process.

We appreciate the reviewer’s suggestion to clarify the pathological significance of YAP pathway inhibition in SARS-CoV-2 infection. To address this, we re-analyzed our snRNA-seq and bulk RNA-seq datasets to determine whether YAP target genes overlap with known mediators of the innate immune response. As described in Fig. 1C, bulk RNA-seq revealed decreased expression of multiple YAP downstream targets linked to innate immune regulation (e.g., Thbs1, Ccl2, Axl, and Csf1) in SARS-CoV-2–infected cells in vitro.

snRNA-seq of alveolar type I (AT1) cells from COVID-19 patients revealed a more complex landscape: While we observed reduced YAP activity overall (Fig. 1G), multiple YAP target genes involved in innate immunity and cytokine signaling were paradoxically elevated (Supplemental Fig. 1E). Several factors likelt explain these conflicting observations: 1. In the lung, AT1 cells (which are critical for gas exchange) may cell specifically respond to virus infection by upregulating genes related to immune response by other signaling pathway(s); 2. In vivo, SARS-CoV-2 infection triggers a surge in cytokines, chemokines, and other local factors that can differentially modulate YAP binding sites and thus affect its downstream targets, a complexity not fully captured in vitro; 3. YAP is highly sensitive to mechanical signals and tissue architecture. The 3D structure of altered cell–cell junctions in infected lung tissue, and fluid shear stress in the alveolar space could shape YAP target gene transcription differently from simplified monolayer cell cultures.

We have expanded the results section of the new version to include the above points. We also acknowledge that ongoing and future work is needed to delineate the exact molecular and tissue-specific pathways through which YAP inhibition confers a potential advantage in combating SARS-CoV-2.

(2) The authors concluded that helicase activity is required for NSP13-induced inhibition of YAP transcriptional activity based on mutation studies (Figure 3B). This finding is somewhat confusing, as K131, K345/K347, and R567 are all essential residues for NSP13 helicase activity while mutating K131 did not affect NSP13's ability to inhibit YAP (Figure 3B). Additionally, there are no data showing exactly how NSP13 inhibits the YAP/TEAD complex through its helicase function. This point was also not reflected in their proposed working model (Figure 4H).

We appreciate the reviewer’s concerns regarding the helicase‐dependent inhibition of YAP by NSP13, particularly the roles of K131, K345/K347, and R567. Based on published structural and biochemical studies, each of these residues uniquely supports helicase function (1): K131 is crucial for stabilizing the NSP13 stalk region by interacting with S424. Substituting K131 with alanine (K131A) reduces helicase efficiency but does not completely abolish it; K345/K347 are key DNA‐binding residues, and mutating both (K345A/K347A) largely prevents NSP13 from binding DNA, thus eliminating unwinding. R567 is critical for ATP hydrolysis, and the R567A mutant retains DNA binding capacity but fails to unwind it. In Fig. 3B, K131A suppresses YAP transactivation to nearly the same extent as wild‐type NSP13, suggesting that partial helicase activity is sufficient for complete YAP/TEAD inhibition. Conversely, the K345A/K347A and R567A mutants show markedly diminished repression, underscoring the importance of DNA binding and ATP hydrolysis.

As the new Fig. 4J illustrates, NSP13 must bind DNA and hydrolyze ATP to unwind nucleic acids. This helicase‐dependent process likely enables NSP13 to remodel chromatin structure by binding TEAD and properly organize YAP repressors at YAP/TEAD complex to prevent YAP/TEAD transactivation. In support of this mechanism, the K345A/K347A mutant, unable to anchor to DNA, fails to repress YAP and slightly increases YAP‐driven transcription (Fig. 3B), presumably by mislocalizing YAP repressors. Likewise, the ATPase‐dead R567A can bind DNA but does not unwind and remodel chromatin to recruit YAP repressors, resulting in a loss of YAP suppression (Fig. 3B and 3F). Our revised model demonstrates that both DNA binding and ATP‐dependent unwinding are essential for NSP13 to suppress YAP transcriptional activity. We have updated the results, discussion, and model accordingly.

(3) The proposed model that NSP13 binds TEAD4 to recruit repressor proteins and inhibits YAP/TEAD downstream gene transcription (Figure 4H) needs further characterization. Second, NSP13 is a DNA-binding protein, and its nucleic acid-binding mutant K345A/K347A failed to inhibit YAP transcriptional activity (Figure 3B). The authors should investigate whether NSP13 could bind to the TEAD binding sequence or the nearby sequence on the genome to modulate TEAD's DNA binding ability. Third, regarding the identified nuclear repressors, the authors should validate the interaction of NSP13 with the ones whose loss activates YAP transcriptional activity (Figure 4G). Lastly, why can't NSP13 bind TEAD4 in the cytoplasmic fractionation if both NSP13 and TEAD4 are detected there (Figure 3B)? This finding indicates their interaction is not a direct protein-protein interaction but is mediated by something in the nucleus, such as genomic DNA.

(1) Low TEAD expression in HEK293T cells: Our IP-MS experiments were performed in HEK293T cells, which, according to the Human Protein Atlas, express TEAD1–4 at comparatively low levels (TEAD1: 16.5, TEAD2: 16.4, TEAD3: 4.9, TEAD4: 38.7 nTPM). In contrast, HeLa cells, where we successfully validated NSP13-mediated YAP suppression (Fig. 4H, Supplementary Fig.5B-D), show higher expression of these TEAD isoforms (TEAD1: 97.1, TEAD2: 27.3, TEAD3: 12.2, TEAD4: 48.1 nTPM). Therefore, insufficient TEAD abundance in HEK293T cells may limit the sensitivity needed to detect TEAD–NSP13 interactions in our proteomic screens.

(2) Transience and potential DNA dependence: Our co-immunoprecipitation (co-IP) experiments (Fig. 4B, Supplementary Fig.4C-E) indicated that NSP13–TEAD4 binding is low-affinity. Under standard IP-MS conditions (which typically do not include chemical cross-linkers or nucleic acids to stabilize transient complexes), weak or short-lived interactions can be lost during washes or sample processing.

(3). Additional supporting evidence: We carefully checked our IP-MS data and found that the well-known TEAD binding proteins, including CTBP1/2 and GATA4, were pulled down, suggesting TEAD’s absence does not rule out an NSP13–TEAD association.

(3a) We acknowledge that our NSP13 immunoprecipitation–mass spectrometry (IP-MS) did not identify any TEAD proteins (Fig. 4G and IP-MS tables). Several factors likely contributed to this outcome:

(3b) We sincerely appreciate the reviewer’s insightful suggestion. While we agree that mapping NSP13 occupancy at individual TEAD-binding motifs is valuable, we respectfully consider this to be beyond the scope of the current study. Biochemical and structural work on coronavirus NSP13 shows that it recognizes nucleic‑acid substrates primarily through their 5′ single‑stranded overhang and duplex architecture, not through a defined base sequence(2, 3). Accordingly, our data (Fig. 3B and 3F) indicate that DNA binding ability, rather than recognition of a specific motif, enables NSP13 to perform its helicase activity in proximity to TEAD and recruit repressors. Moreover, the DNA‑binding mutant K345A/K347A and the ATPase‑dead mutant R567A both fail to suppress YAP/TEAD transcription despite retaining the ability to interact with TEAD (Fig. 3B). These loss‑of‑function phenotypes demonstrate that NSP13’s chromatin engagement and unwinding activity, rather than sequence‑restricted targeting, are essential for repression. For these reasons, motif‑specific binding assays were not pursued in this revision, but we clarified in the discussion that NSP13’s DNA engagement is likely structural or TEAD-dependent, rather than sequence‑directed. We also highlighted this as an important avenue for future investigation.

(3c) To validate the NSP13 interacting proteins from our IP-MS data, we generated plasmids expressing several candidates (CCT3, SMARCD1, EIF4A1, LMNA, TTF2, and YY2) and performed co-IP assays. As predicted, we confirmed the robust interaction between NSP13 and TEAD (Supplemental Fig. 5E). However, these putative nuclear repressors exhibited weak binding to NSP13 compared with TEAD4, suggesting that NSP13 associates with them indirectly, possibly as part of a larger multiprotein complex or depending on the chromatin structure, rather than via direct protein–protein interaction (Fig. 4J).

(3d) We appreciate the reviewer’s question. To investigate whether their association might be DNA‐dependent, we performed co‐IP experiments using nuclear lysates in the presence or absence of various nucleases: Universal Nuclease (which degrades all forms of DNA and RNA), DNase I (which cleaves both single‐ and double‐stranded DNA), and RNase H (which selectively cleaves the RNA strand in RNA/DNA hybrids). Our findings revealed that nucleic acid removal did not disrupt the NSP13/TEAD4 interaction (Supplemental Fig.4E), indicating that their binding is not solely mediated by DNA or RNA.

Reviewer #2 (Public Review):

Specific comments and suggestions for improvement of the manuscript:

(1) NSP13 has been reported to block, in a helicase-dependent manner, episomal DNA transcription (PMID: 37347173), raising questions about the effects observed on the data shown from the HOP-Flash and 8xGTIIC assays. It would be valuable to demonstrate the specificity of the proposed effect of NSP13 on TEAD activation by YAP (versus broad effects on reporter assays) and also to show that NSP13 reduces the function of endogenous YAP-TEAD transcriptional activity (i.e., does ectopic NSP13 expression reduce the expression of YAP induced TEAD target genes in cells).

We appreciate the reviewer’s comments and have carefully revisited the conclusions from the published paper(4) (PMID: 37347173), which reported that NSP13 suppresses episomal DNA transcription, as evidenced by reduced Renilla luciferase (driven by the herpes simplex virus thymidine kinase promoter) and GFP expression upon co‐expression with NSP13. For our experiments, we used a dual‐luciferase assay with Renilla luciferase (under the same promoter) as an internal control. After re-examining our raw Renilla luciferase data (now provided in the supplemental Excel file “Supporting data value”), we found that while 100 ng of NSP13 did not affect Renilla luciferase levels, 400 ng of NSP13 reduced them by approximately 50% relative to the YAP5SA‐only group (Supplemental Fig.2B, Fig.3C-D). We observed a similar reduction with NSP13 truncation mutants—an outcome not fully consistent with the published study (Supplemental Fig.3D, PMID: 37347173). However, unlike their finding of robust episomal DNA suppression, our data indicate that the K345A/K347A mutant of NSP13, which lacks DNA‐binding ability, completely lost its suppressive effect (Fig.3B).

We performed additional Notch reporter assays to address the concern that NSP13 might nonspecifically inhibit episomal DNA transcription (including the HOP‑Flash and 8×GTIIC reporters). These experiments revealed that co‑expression of NSP13 with NICD (Notch intracellular domain) does not suppress Notch signaling (Supplemental Fig. 2C), indicating that NSP13 does not globally block all reporter systems. To evaluate whether NSP13 reduces endogenous YAP‑TEAD activity, we transiently overexpressed NSP13 WT and its R567A mutant in HeLa cells. However, bulk RNA‑seq and qPCR analyses did not reveal a clear decrease in YAP target genes, possibly due to the low transfection efficiency (< 50%, Supplemental Fig.4D). Interestingly, we observed that YAP5SA was predominantly retained in the nucleus upon NSP13 or R567A co‑expression, suggesting that NSP13 (or together with its interacting partners) restricts YAP5SA cytoplasmic shuttling. Future studies will involve stable cell lines expressing NSP13 WT or R567A to better characterize the mechanisms driving YAP5SA nuclear retention and clarify how NSP13 specifically suppresses YAP activity.

(2) While the IP-MS experiment may have revealed new regulators of TEAD activity, the data presented are preliminary and inconclusive. No interactions are validated and beyond slight changes in TEAD reporter activity following knockdown, no direct links to YAP-TEAD are demonstrated, and no link to NPS13 was shown. Also, no details are provided about the methods used for the IP-MS experiment, raising some concerns about potential false positive associations within the data.

We appreciate the reviewer’s feedback regarding our IP-MS findings and acknowledge that additional validation is required to establish definitive links between the identified putative regulators, YAP-TEAD, and NSP13. We have taken the following steps (and plan further experiments) to address these concerns:

(2a) Co-IP validation: Same with the answer for Reviewer #1 (3c), we generated plasmids expressing several top candidate interactors from the IP-MS data (CCT3, SMARCD1, EIF4A1, LMNA, TTF2, and YY2) and performed direct co-IP assays in a more controlled setting. The results indicated that these putative NSP13 interactors had weaker binding compared to TEAD4, implying that NSP13 may associate with them as part of a larger complex or depending on the chromatin structure rather than through a direct protein–protein interaction (Fig. 4J).

(2b) qPCR validation: Beyond reporter assays for evaluating YAP transactivation after the candidate YAP suppressor knockdown (Fig. 4H and Supplemental Fig. 5C), we performed qPCR to detect YAP activation on endogenous YAP-TEAD target genes (e.g., CTGF CYR61, and AMOTL2) after CCT3 knockdown. Expression of CTGF and CYR61 was higher compared to control (Supplemental Fig. 5D), strengthening the case for an interaction relevant to YAP-TEAD signaling.

(2c) To investigate how NSP13‐interacting proteins link to the YAP/TEAD complex, we examined the IP‑MS dataset and identified several well‐known YAP and TEAD binding partners, including CTBP1/2 (TEAD‐binding), GATA4 (TEAD‐binding), and multiple 14‐3‐3 isoforms (YWHAZ/YWHAB/YWHAH/YWHAQ, YAP binding). These findings suggest that NSP13 may form a larger nuclear complex with YAP/TEAD and associated cofactors. In the future, we will determine whether these putative TEAD regulators also interact with NSP13 under various conditions (e.g., in the presence or absence of DNA) and whether co‐expression of NSP13 influences their association with YAP or TEAD. This approach will clarify how NSP13 might leverage these factors to regulate YAP‐TEAD function.

(2e) For the mass spectrometry experiments, HEK293T cells were transfected with Flag‐YAP1, HA‐NSP13, or Flag‐YAP1 + HA‐NSP13 according to the manufacturer’s standard protocols. After nuclear extraction and lysis, the supernatant was incubated with HA magnetic beads to immunoprecipitate (IP) NSP13. The IP samples were subsequently analyzed by mass spectrometry to identify NSP13‐associated proteins (Fig. 4F). Each experimental condition was performed in duplicate to ensure reproducibility. We included an appropriate negative control (Flag‐YAP1) and stringent data‐filtering criteria to minimize false positives. We apologize for not including these details in our original Methods section; in this revised manuscript, we have fully described the number of replicates, the controls used, and our data analysis pipelines.

eLife Assessment

The study presents valuable theoretical insights by attempting to classify pattern-forming gene subnetworks and exploring their potential mechanisms. However, the results are incomplete, as they rely on oversimplified models, limited classifications, and assumptions that may not hold in more complex or realistic scenarios.

Reviewer #1 (Public review):

Summary:

The authors tackle a long-standing question in developmental theory: given a gene-regulatory network that includes extracellular signalling, which topologies are even capable of transforming an initial spatial profile into a genuinely new pattern? Building on the classical reaction-diffusion framework in one dimension, but imposing biologically motivated constraints, they prove that every one-signal sub-network must be either Hierarchical (H), self-activating (L+), or self-inhibiting (L-). They further demonstrate that only three composite classes of full networks - pure H, a coupled L+ L- "Turing" pair, and an L- module fed by an intracellular positive loop ("noise-amplifying")-can create non-trivial spatial transformations. Analytical criteria and illustrative simulations are provided, together providing a closed taxonomy, which is supposed to be relevant for real systems.

Strengths:

(1) Useful classification framework. Reducing a vast number of possible gene circuits to three canonical pattern-forming motifs is a valuable organising insight for both theorists and experimentalists.

(2) Logical completeness. All required cases are addressed, and the proofs elevate previous computational observations to formal statements.

(3) Practical interpretability. Given a reaction network diagram, one can now decide (assuming the model applies to the real systems) whether spatial patterning is even possible, saving experimental effort on in-silico screens that could never succeed.

Weaknesses:

(1) The Results section is difficult to follow. Key logical steps and network configurations are described shortly in prose, which constantly require the reader to address either SI or other parts of the text (see numerous links on the requirements R1-R5 listed at the beginning of the paper) to gain minimal understanding. As a result, a scientifically literate but non-specialist reader may struggle to grasp the argument with a reasonable time invested.

(2) A central step in the model formulation is the linearisation of the reaction term around a homogeneous steady state; higher-order kinetics, including ubiquitous bimolecular sinks such as A + B → AB, are simply collapsed into the Jacobian without any stated amplitude bound on the perturbations. Because the manuscript never analyses how far this assumption can be relaxed, the robustness of the three-class taxonomy under realistic nonlinear reactions or large spike amplitudes remains uncertain.

(3) All modelling is confined to one spatial dimension, and the very definition of a "non-trivial" transformation is framed in terms of peak positions along a line, which clearly must be reformulated for higher dimensions. It's well-known that diffusions in 1, 2, and 3 dimensions are also dramatically different, so the relevance of the three-class taxonomy to real multicellular tissues remains unclear, or at least should be explained in more detail.

Discussion:

As stated above, there are several uncertainties about the relevance of the presented framework for real systems. However, if the results hold, researchers could look at a gene-network diagram and quickly judge whether it can make spatial patterns and, if so, which of the three known mechanisms it will use. That shortcut would save experimental and computational time. In the case that the results don't hold for the real systems, the authors' proof tools at least give theorists a solid base they can extend to more complex cases.

Reviewer #2 (Public review):

Summary:

This study explores how gene regulatory networks that include intra- and extracellular signaling can give rise to spatial patterns of gene expression in cells. The authors investigate this question in a simplified theoretical framework, where all cells are assumed to respond identically to signals, and spatial details such as cell boundaries and extensions are abstracted away. Within this setting, they identify three distinct signaling topologies, referred to as L and H types, and combine them into three minimal subnetworks capable of generating patterns. The study analyzes possible combinations of these topologies and examines how each subnetwork behaves under three different initial conditions. Combining the analyses with mathematical proofs and heuristic arguments, the authors define necessary conditions under which such networks can produce non-trivial spatial patterns.

Strengths:

The authors break down larger gene regulatory networks into smaller subnetworks, which allows for a more tractable analysis of pattern formation. These minimal subnetworks are examined under different initial conditions, providing a range of examples for how patterns can emerge in simplified settings. The study also proposes necessary conditions for pattern formation, which may be useful for identifying relevant network structures. In addition, the manuscript offers heuristic explanations for the emergence of patterns in each subnetwork, which help to interpret the simulation results and analytical criteria.

Weaknesses:

(1) We have serious concerns regarding the validity of the simulation results presented in the manuscript. Rather than simulating the full nonlinear system described by Equation (1), the authors base their results on a truncated expansion (Equation S.8.2) that captures only the time evolution of small deviations around a spatially homogeneous steady state. However, it remains unclear how this reduced system is derived from the full equations - specifically, which terms are retained or neglected and why - and how the expansion of the nonlinear function can be steady-state independent, as claimed. Additionally, in simulations involving the spike plus homogeneous initial condition, it is not evident - or, where equations are provided, it is not correct - that the assumed global homogeneous background actually corresponds to a steady state of the full dynamics. We elaborate on these concerns in the following:

It is assumed that the homogeneous steady states are given by g_i=0 and g_i=c_i, where 1/c_i = \mu_i or \hat{\mu}_i​, independently of the specific network structure. However, the basis for this assumption is unclear, especially since some of the functions do not satisfy this condition - for example, f5​ as defined below Eq. S8.10.5. Moreover, if g_i=c_i does not correspond to a true steady state, then the time evolution of deviations from this state is not correctly described by Eq. S8.2, as the zeroth-order terms do not vanish in that case.

Additionally, the equations used contain only linear terms and a cubic degradation term for each species g_i, while neglecting all quadratic terms and cubic terms involving cross-species interactions (i≠j). An explanation for this selective truncation is not provided, and without knowledge of the full equation (f), it is impossible to assess whether this expansion is mathematically justified. If, as suggested in the Supplementary Information, the linear and cubic terms are derived from f, then at the very least, the Jacobian matrix should depend on the background steady-state concentration. However, the equations for the small deviation around a steady state (including the Jacobian matrix) used in the simulations appear to be independent of the particular steady state concentration.

This is why we believe that the differences observed between the spike-only initial condition and the spike superimposed on a homogeneous background are not due to the initial conditions themselves, but rather result from a modified reaction scheme introduced through a questionable cutoff.

"In simulations with spike initial patterns, the reference value g≡0 represents an actual concentration of 0 and therefore, we must add to (S8.2) a Heaviside function Φ acting of f (i.e., Φ(f(g))=f(g) if f(g)>0 , Φ(f(g))=0 if f(g){less than or equal to}0 ) to prevent the existence of negative concentrations for any gene product (i.e., g_i<0 for some i )." (SI chapter S8).

This cutoff alters the dynamics (no inhibition) and introduces a different reaction scheme between the two simulations. The need for this correction may itself reflect either a problem in the original equations (which should fulfill the necessary conditions and prevent negative concentrations (R4 in main text)) or the inappropriateness of using an expanded approximation which assumes independence on the steady state concentration. It is already questionable if the linearized equations with a cubic degradation term are valid for the spike initial conditions (with different background concentration values), as the amplitude of this perturbation seems rather large.

Lastly, we note that under the current simulation scheme, it is not possible to meaningfully assess criteria RH2a and RH2b, as they rely on nonlinear interactions that are absent from the implemented dynamics.

(2) Most of the proofs presented in the Supplementary Information rely on linearized versions of the governing equations, and it remains unclear how these results extend to the fully nonlinear system. We are concerned that the generality of the conclusions drawn from the linear analysis may be overstated in the main text. For example, in Section S3, the authors introduce the concept of dynamic equivalence of transitive chains (Proposition S3.1) and intracellular transitive M-branching (Proposition S3.2), which pertains to the system's steady-state behavior. However, the proof is based solely on the linearized equations, without additional justification for why the result should hold in the presence of nonlinearities. Moreover, the linearized system is used to analyze the response to a "spike initial pattern of arbitrary height C" (SI Chapter S5.1), yet it is not clear how conclusions derived from the linear regime can be valid for large perturbations, where nonlinear effects are expected to play a significant role. We encourage the authors to clarify the assumptions under which the linearized analysis remains valid and to discuss the potential limitations of applying these results to the nonlinear regime.

(3) Several statements in the main text are presented without accompanying proof or sufficient explanation, which makes it difficult to assess their validity. In some cases, the lack of justification raises serious doubts about whether the claims are generally true. Examples are:

"For the purpose of clarity we will explain our results as if these cells have a simple arrangement in space (e.g., a 1D line or a 2D square lattice) but, as we will discuss, our results shall apply with the same logic to any distribution of cells in space." (Main text l.145-l.148).

"For any non-trivial pattern transformation (as long as it is symmetric around the initial spike), there exists an H gene network capable of producing it from a spike initial pattern." (Main text l.366f).

"In 2D there are no peaks but concentric rings of high gene product concentration centered around the spike, while in 3D there are concentric spherical shells." (Main text l. 447ff).

(4) The study identifies one-signal networks and examines how combinations of these structures can give rise to minimal pattern-forming subnetworks. However, the analysis of the combinations of these minimal pattern-forming subnetworks remains relatively brief, and the manuscript does not explore how the results might change if the subnetworks were combined in upstream and downstream configurations. In our view, it is not evident that all possible gene regulatory networks can be fully characterized by these categories, nor that the resulting patterns can be reliably predicted. Rather, the approach appears more suited to identifying which known subnetworks are present within a larger network, without necessarily capturing the full dynamics of more complex configurations.

(5) The definition of non-trivial pattern formation is provided only in the Supplementary Information, despite its central importance for interpreting the main results. It would significantly improve clarity if this definition were included and explained in the main text. Additionally, it remains unclear how the definition is consistently applied across the different initial conditions. In particular, the authors should clarify how slope-based measures are determined for both the random noise and sharp peak/step function initial states. Furthermore, the authors do not specify how the sign function is evaluated at zero. If the standard mathematical definition sgn(0)=0 is used, then even a simple widening of a peak could fulfill the criterion for non-trivial pattern transformation.

(6) The manuscript lacks a clear and detailed explanation of the underlying model and its assumptions. In particular, it is not well-defined what constitutes a "cell" in the context of the model, nor is it justified why spatial features of cells - such as their size or boundaries - can be neglected. Furthermore, the concept of the extracellular space in the one-dimensional model remains ambiguous, making it unclear which gene products are assumed to diffuse.

Reviewer #3 (Public review):

Pattern formation is responsible for generating the spatial organization of cells, tissues, and organs during embryogenesis. It operates within a multifactorial system including initial conditions, gene regulatory networks, extracellular signals, mechanical forces, stochastic noise, and environmental inputs. Finally, it ensures the functional anatomy of an organism.

This study focuses on the one central aspect in pattern formation: how spatial heterogeneity arises from an initial condition and evolves into a more complex or distinct spatial pattern (non-trivial pattern formation, as they termed). The authors made efforts to explore and characterize all possible ways to achieve the pattern formation. They do this by discussing how extracellular signals spread, how individual cells respond to those signals, and how those responses, in turn, modulate signal propagation.

Finally, their comprehensive analysis summarizes that there are three classes of interactions between extracellular signals and intracellular responses, corresponding to previously known mechanisms that can generate spatial patterns: difference in morphogen concentrations in space, noise-amplification, and Turing pattern.

Author response:

We thank the reviewers for their thorough evaluation and constructive feedback on our manuscript.

We think that their valuable suggestions will strengthen the manuscript and help us clarify several important points.

All reviewers acknowledged the importance of our theoretical results and network classification in making pattern formation analysis a more tractable problem. At the same time, they have also raised a number of important concerns that we shall carefully consider.

A. A major clarification that the reviewers found important concerns the definition of non-trivial pattern transformations and its generalization to higher dimensions. In this regard, the reviewers’ comments are:

Reviewer #1:

(on non-trivial pattern transformations):

(3) All modelling is confined to one spatial dimension, and the very definition of a "non-trivial" transformation is framed in terms of peak positions along a line, which clearly must be reformulated for higher dimensions. It's well-known that diffusions in 1, 2, and 3 dimensions are also dramatically different, so the relevance of the three-class taxonomy to real multicellular tissues remains unclear, or at least should be explained in more detail. Reviewer #2 (on non-trivial pattern transformations):

(5) The definition of non-trivial pattern formation is provided only in the Supplementary Information, despite its central importance for interpreting the main results. It would significantly improve clarity if this definition were included and explained in the main text. Additionally, it remains unclear how the definition is consistently applied across the different initial conditions. In particular, the authors should clarify how slope-based measures are determined for both the random noise and sharp peak/step function initial states. Furthermore, the authors do not specify how the sign function is evaluated at zero. If the standard mathematical definition sgn(0)=0 is used, then even a simple widening of a peak could fulfill the criterion for nontrivial pattern transformation.

We agree with Reviewer #2 that including a more detailed definition of non-trivial pattern transformation in the main text would enhance the clarity of the paper. The one-dimensional (1D) definition currently provided in the Supplementary Information was chosen because all computations presented therein involve exclusively one-dimensional patterns. However, we acknowledge that this definition, as it was, did not have a totally unambiguous generalization  to higher dimensions. Therefore, in a revised version of the manuscript, we will incorporate an expanded definition applicable to higher-dimensional cases.

This general definition of a non-trivial pattern transformation should make no reference to the sign of spatial derivatives of either the initial or resulting patterns. Specifically, a pattern transformation is considered non-trivial if it satisfies the following criteria:

- It is heterogeneous: The resulting pattern is heterogeneous in space.

- It is rearranging: The arrangement of critical points (i.e. peaks, valleys and saddle points in a gene product concentration) along the domain in the resulting pattern of a gene product is different to the arrangement of critical points in its initial pattern. This includes the emergence of new critical points, the disappearance of existing ones, or the spatial displacement of critical points from one location to another.

- It is non-replicating: The spatial arrangement of critical points in the pattern of one gene product must differ from that of any other upstream gene product.

Nonetheless, our two initial patterns are spatially discontinuous functions: in homogeneous initial patterns, the white noise is discontinuous by definition; and for the spike and spike+homogeneous initial patterns, we use sharp spikes defined by the rectangular function, which is discontinuous at the spike boundaries. Therefore, the aforementioned definition should be supplemented with the following two ad hoc assumptions:

- Homogeneous initial patterns do not comprise any critical point. White noise in this type of initial patterns represents small thermodynamic fluctuations around the steady state and, for the purpose of pattern transformation, this is equivalent to a constant concentration along the domain.

- Spike and spike+homogeneous initial patterns each contain a single critical point located at the center of the spike. The sharp spikes, modeled using the rectangular function, serve as a theoretical idealization to facilitate mathematical analysis. Once diffusion begins to act, these sharp boundaries are smoothed into differentiable gradients, maintaining a unique critical point at the center of the initial spike, which is the most relevant information for pattern transformation.

Finally, it is worth recalling that our gene network classification is fundamentally based on an analysis of the dispersion relation associated with the gene network, and the construction of this dispersion relation is independent of the spatial dimensionality of the domain (i.e. it does not require assuming any specific number of dimensions). The fact that the description of this dispersion relation was in the SI may have been non-ideal for the understandability of the article and will, consequently, be moved to the main text in an upcoming version of the article. Thus, the gene networks that can lead to pattern transformation are the same in 1D, 2D or 3D. As for the resulting patterns, the broad description we provide also applies to any number of dimensions; these would be periodic, non periodic as in the amplified noise patterns or non periodic as in the hierarchic networks. For the latter notice that, except for boundary effects that we later discuss, the spike initial condition is radially symmetric and thus, the patterns resulting from it will also be radially symmetric. We will make this point more explicit in a revised version of the article, especially since, as suggested, this important portion of the Supplementary Information will be incorporated into the main text.

Reviewer 2 suggests that with our definition of non-trivial pattern transformation, the simple widening of a concentration peak would constitute a non-trivial pattern transformation. This is not the case, as already shown in the figures as a example, since in a widening there is no change in the position of the critical point. A different situation applies if a wide and completely flat concentration peak (i.e. a plateau) forms. As we will explain in the coming version this is not possible because of requirement R5.

We think that this clarification of the definition of non-trivial pattern transformation will also help clarify the next point (B below) since it would make it clearer that this article does not intend to explain which specific resulting pattern would arise from any given gene network.

B. The main concern among these relates to the validity of our linearization of the model equations and the extension of the results obtained for the linear system to the fully nonlinear system. In this regard, the reviewers’ comments are:

Reviewer #1:

(on linearization):

(2) A central step in the model formulation is the linearisation of the reaction term around a homogeneous steady state; higher-order kinetics, including ubiquitous bimolecular sinks such as A + B → AB, are simply collapsed into the Jacobian without any stated amplitude bound on the perturbations. Because the manuscript never analyses how far this assumption can be relaxed, the robustness of the three-class taxonomy under realistic nonlinear reactions or large spike amplitudes remains uncertain.

Reviewer #2:

(on linearization):

(2) Most of the proofs presented in the Supplementary Information rely on linearized versions of the governing equations, and it remains unclear how these results extend to the fully nonlinear system. We are concerned that the generality of the conclusions drawn from the linear analysis may be overstated in the main text. For example, in Section S3, the authors introduce the concept of dynamic equivalence of transitive chains (Proposition S3.1) and intracellular transitive M-branching (Proposition S3.2), which pertains to the system's steady-state behavior. However, the proof is based solely on the linearized equations, without additional justification for why the result should hold in the presence of nonlinearities. Moreover, the linearized system is used to analyze the response to a "spike initial pattern of arbitrary height C" (SI Chapter S5.1), yet it is not clear how conclusions derived from the linear regime can be valid for large perturbations, where nonlinear effects are expected to play a significant role. We encourage the authors to clarify the assumptions under which the linearized analysis remains valid and to discuss the potential limitations of applying these results to the nonlinear regime.

In this article, we address two main questions: first, which gene network topologies can give rise to non-trivial pattern transformations; and second, which broad types of resulting patterns can these gene network topologies give rise to resulting pattern. Thus, we are not intending to explain which exact resulting patterns would arise from any given gene network (i.e. a gene network topology with specific functions and interaction strengths or weights), a question for which non-linearities do indeed matter.

For most known gene regulatory networks, available empirical information is typically limited to the nature of gene product regulations -indicating whether they act as activators or inhibitors- while details about the specific functional form of these regulations are rare. For instance, given two gene products, i and j, the network may indicate that i acts as an activator of j, implying that the concentration of j increases with that of i. However, this increase could follow a variety of functional forms: it may be quadratic (e.g., ), cubic (e.g., ), or any other function f j(gi). As we explain in the description of our model, we restrict our study to functions with a monotonicity constraint: higher concentrations of i lead to increased production of j (i.e., ).  In other words, a given gene interaction is always inhibitory or activatory, it does not change of sign. This monotonicity constraint corresponds to requirement (R5) in our main text. This requirement it is based on the biologically plausible idea that the complexity of gene regulation in development stems more from the topology of gene networks than from the complexity of the regulation by which a gene product may regulate another (i.e. we use simple monotonic functions).

Question 1: A critical part to understand question 1 is in the dispersion relation that was explained in SI. From the reviewers’ comments it is clear that having this crucial part in the main text of an upcoming version of the article would improve understandability, specially for question 1.

In brief, any pattern transformation requires the initial pattern to change. The trigger of such change is a change in the concentration of some gene product, either conceptualized as a noise fluctuation (in the homogeneous initial pattern) or a regulated change in a specific point (in the spike initial pattern). Mathematically, both can be conceptualized as perturbations and, for pattern transformation to be possible, such perturbation should grow so that the initial pattern becomes unstable and can change to another resulting pattern.

If the perturbation is small, one can use the standard linear perturbation analysis in S6.2 of our Supplementary Information. In other words, the linear analysis is enough to ascertain if a small perturbation would grow or not. A gene network in which this will not happen would be unable to lead to pattern transformation, whichever the nonlinear part of f(g). In that sense, the linear approximation provides a necessary condition that any gene network needs to fulfill to lead to pattern transformation.

However, the linear analysis would not ascertain whether a specific gene network will actually lead to pattern transformation (i.e., the condition is not sufficient). This, as well as the shape of the specific resulting pattern, may actually depend on the non-linear parts too. As we discuss, based on the dispersion relation, and other complementing arguments along the article, we can also get some insights on the possible patterns from the linear approximation alone (question 2). This arguments hold thanks to the imposition of requirements (R1-R5) on function f(g), which prevent strange behaviors stemming from the nonlinear part of the equation.

The amplitude bound of perturbations mentioned by Reviewer #1 is addressed by requirements (R2) and (R4). Although the solution to the linear system predicts unbounded growth of unstable eigenmodes, the assume functions f(g) on which the nonlinear terms  eventually halt this growth, thereby ensuring the boundedness of solutions as imposed by (R4). This assumption on the nonlinear part is literally requirement R2 on f(g) in the main text.

The transitive chains and branchings in section S3 of the Supplementary Information mentioned by the Reviewer #2 are topological properties of gene networks and therefore they influence only the linear part of the reaction-diffusion equations. This is why the proofs in that section are based on the linearized equations. We agree that clarifying this point in the text, as suggested by the reviewer, would improve the reader’s understanding of the section.

Regarding Reviewer #2’s concerns about large perturbations, we acknowledge that the phrasing using “arbitrary height” may be confusing. For the homogeneous initial conditions these perturbations are assumed to be small because they are actually molecular noise (otherwise the initial condition could not be considered homogenous in the classical sense of developmental biology models). In the spike initial conditions in hierarchic networks the perturbation is not necessarily small. For the analysis provided in the SI we indeed assume that the perturbations are small enough for the linear approximation to be possible. Notice, however, that since these networks require an intracellular self-activating loop upstream of the first extracellular signal, the effective perturbation would rapidly grow to a value determined by such loop.

In general the height of the initial spike does not affect the fact that hierarchic networks can lead to non-trivial pattern transformation. By definition these networks require the secretion of an extracellular signal from the cells in the spike (otherwise no change in gene product concentrations can occur over space). By definition this signal is not produced by any other cells and, thus, its concentration is governed by diffusion from the spike and its production in the cells in the spike. Thus, whichever the initial height of the spike and whichever the non-linearities in f(g), the signal’s concentration would decrease with the distance from the spike. As explained in the main text, this would lead to non-trivial pattern transformations if other general conditions are met. In general, the height of the initial perturbation can affect which specific pattern transformation would arise from a specific gene network but not which gene network topologies can lead to pattern transformation. This will be more clearly stated in an upcoming version of the article. C. In the following, we respond to the remaining concerns raised by the reviewers:

Reviewer #1:

(1) The Results section is difficult to follow. Key logical steps and network configurations are described shortly in prose, which constantly require the reader to address either SI or other parts of the text (see numerous links on the requirements R1-R5 listed at the beginning of the paper) to gain minimal understanding. As a result, a scientifically literate but non-specialist reader may struggle to grasp the argument with a reasonable time invested.

We acknowledge that the current version of the main text may not be as clear as we intended. Initially, we believed that placing the more technical mathematical passages in the Supplementary Information would make the main text more accessible to readers. However, we agree with the reviewer that including some of these computations in the main text could improve clarity. We also believe that adding a summary table outlining all the model’s requirements would further contribute to that goal.

Reviewer #2:

(1) We have serious concerns regarding the validity of the simulation results presented in the manuscript. Rather than simulating the full nonlinear system described by Equation (1), the authors base their results on a truncated expansion (Equation S.8.2) that captures only the time evolution of small deviations around a spatially homogeneous steady state. However, it remains unclear how this reduced system is derived from the full equations specifically, which terms are retained or neglected and why- and how the expansion of the nonlinear function can be steady-state independent, as claimed. Additionally, in simulations involving the spike plus homogeneous initial condition, it is not evident -or, where equations are provided, it is not correct- that the assumed global homogeneous background actually corresponds to a steady state of the full dynamics. We elaborate on these concerns in the following:

We believe there has been a misunderstanding regarding the presentation of the model equations (S8.2) used throughout our simulations. Accordingly, we agree that this relevant section of the Supplementary Information should be rewritten in a revised version of the manuscript to clarify this issue. Below, we address all the concerns raised by the reviewer.

Equation (S8.2) represents the full nonlinear system described in Equation (1). While we recognize that the model may oversimplify real biological processes, its purpose is to illustrate our general statements about pattern formation rather than to capture any specific or detailed mechanism. In this context, model (S8.2) offers three key advantages for our goals: it allows rapid manipulation of gene network topology simply by modifying the matrix J, making it ideal for illustrating pattern formation across different network classes; it accommodates gene networks of arbitrary size -unlike other models, such as the classical Gierer-Meinhardt model, which are limited to two-element Turing or noise-amplifying networks-; and, due to the simplicity of its nonlinear terms, this model involves relatively few free parameters, facilitating the fine-tuning needed to identify parameter regions where non-trivial pattern transformations occur.

Indeed, we find that the ability of model (S8.2) to illustrate our results despite having such simple nonlinear terms -bearing in mind that at least some nonlinearity is always necessary for selforganization- strongly supports the claim that the capacity of a gene network to produce pattern transformations is fully determined by the linear part of Equation (1). In this sense, nonlinear terms primarily influence the precise parameter values at which these transformations occur and contribute to shaping specific features of the resulting patterns.

Model (S8.2) has been successfully employed in pattern formation studies elsewhere in the literature; accordingly, we provide relevant bibliographic references to support its widespread use.

We believe the misunderstanding arises from our explanation of the biological interpretation of the model. As noted in the accompanying bibliography, the model is based on a general reactiondiffusion mechanism assuming the existence of a steady state. However, this conceptual reactiondiffusion framework is not the same as our Equation (1); rather, it was introduced by the original proponents of the model in the seminal paper cited in our text. In this context, Equation (S8.2) describes small concentration perturbations around that steady state, where the variables represent deviations in concentration relative to the general steady state.

The aforementioned general steady state corresponds to the trivial equilibrium point g≡0 in equations (S8.2). Consequently, all our simulations based on model (S8.2) start from this steady state, to which we add white noise to generate homogeneous initial patterns or a sharp spike for the two types of spike initial patterns.

It is also worth noting that Equations (S8.2) represent a non-dimensional model.

It is assumed that the homogeneous steady states are given by g_i=0 and g_i=c_i, where 1/c_i = \mu_i or \hat{\mu}_i, independently of the specific network structure. However, the basis for this assumption is unclear, especially since some of the functions do not satisfy this condition -for example, f5 as defined below Eq. S8.10.5. Moreover, if g_i=c_i does not correspond to a true steady state, then the time evolution of deviations from this state is not correctly described by Eq. S8.2, as the zeroth-order terms do not vanish in that case.

From the explanations above, it is important to distinguish two scales in the process: the scale of small perturbations, where equations (S8.2) apply; and the global scale, where the conceptual general reaction-diffusion system operates. Since the specific form of this general system does not affect equations (S8.2), we assume that it follows any of the models cited in the text, which yield a non-zero steady state at .

In this sense, Equation (S8.2) represent a small concentration deviation of such global system and g(t ,x) is a relative concentration where g≡0 represents the steady-state at are concentrations above , and g<0 are concentrations below .

As previously mentioned, simulations are performed using Equations (S8.2) on the basis of the equilibrium point g≡0. The result of these simulations is then superimposed on the non-zero steady state and presented in the figures along the article.

Using the full model instead of the simplified Equations (S8.2) may result in slightly different resulting patterns, but it does not affect the gene network’s ability to produce pattern transformations, nor does it alter the main structural properties of the patterns—for example, the periodic nature of patterns generated by Turing networks.

Additionally, the equations used contain only linear terms and a cubic degradation term for each species g_i, while neglecting all quadratic terms and cubic terms involving cross-species interactions (i≠j). An explanation for this selective truncation is not provided, and without knowledge of the full equation (f), it is impossible to assess whether this expansion is mathematically justified. If, as suggested in the Supplementary Information, the linear and cubic terms are derived from f, then at the very least, the Jacobian matrix should depend on the background steady-state concentration. However, the equations for the small deviation around a steady state (including the Jacobian matrix) used in the simulations appear to be independent of the particular steady state concentration.

The Jacobian of Equation (S8.2) is independent of g because g represents a small perturbation around a steady state of a general reaction-diffusion system. Consequently, the matrix J corresponds to the Jacobian of the general system evaluated at that steady state. Evaluating the Jacobian of equations (S8.2) at the equilibrium point g≡0 -which represents the general steady state- recovers the matrix J.

This is why we believe that the differences observed between the spike-only initial condition and the spike superimposed on a homogeneous background are not due to the initial conditions themselves, but rather result from a modified reaction scheme introduced through a questionable cutoff.

"In simulations with spike initial patterns, the reference value g≡0 represents an actual concentration of 0 and therefore, we must add to (S8.2) a Heaviside function Φ acting of f (i.e., Φ(f(g))=f(g) if f(g)>0 , Φ(f(g))=0 if f(g){less than or equal to}0 ) to prevent the existence of negative concentrations for any gene product (i.e., g_i<0 for some i )." (SI chapter S8).

This cutoff alters the dynamics (no inhibition) and introduces a different reaction scheme between the two simulations. The need for this correction may itself reflect either a problem in the original equations (which should fulfill the necessary conditions and prevent negative concentrations (R4 in main text)) or the inappropriateness of using an expanded approximation which assumes independence on the steady state concentration. It is already questionable if the linearized equations with a cubic degradation term are valid for the spike initial conditions (with different background concentration values), as the amplitude of this perturbation seems rather large.

For homogeneous and spike+homogeneous initial conditions, we interpret equations (S8.2) as small perturbations around a non-zero steady state of a general reaction-diffusion system. For spike-only initial conditions, that steady state is zero. As we mention before, g≡0 will then represent such steady-state of zero concentration, g>0 are positive concentrations of the general system, and g<0 would represent unfeasible negative concentrations of the general system. Therefore, the use of a cutoff function to handle such initial conditions is justified. Moreover, this cutoff function is the same as the one employed in the reference general system cited in our paper.

We acknowledge that the cutoff influences the simulations and accounts for the differences observed between spike and spike+homogeneous initial conditions. However, this distinction reflects what occurs in real biological systems, which is precisely why we differentiate these two types of initial states. For instance, the emergence of a periodic pattern in a noise-amplifying network depends critically on the formation of regions with concentrations below the steady state near the initial spike. Such regions can form in spike-plus-homogeneous initial patterns but not in spike-only initial patterns, where concentrations below the steady state would correspond to biologically unfeasible negative values.

Lastly, we note that under the current simulation scheme, it is not possible to meaningfully assess criteria RH2a and RH2b, as they rely on nonlinear interactions that are absent from the implemented dynamics.

It is explicitly stated in the relevant subsections of Section S7 in the Supplementary Information that, for the simulations involving RH2a and RH2b, the function f(g) in equation (S8.2) is modified by adding an ad hoc quadratic term to enable the assessment of these criteria.

(3) Several statements in the main text are presented without accompanying proof or sufficient explanation, which makes it difficult to assess their validity. In some cases, the lack of justification raises serious doubts about whether the claims are generally true. Examples are:

"For the purpose of clarity we will explain our results as if these cells have a simple arrangement in space (e.g., a 1D line or a 2D square lattice) but, as we will discuss, our results shall apply with the same logic to any distribution of cells in space." (Main text l.145-l.148).

We believe that the confusion in this statement arises from the ambiguous use of the phrase “our results”. We will revise the text to provide a more precise description. Specifically, by “our results,” we refer to the conclusion that it is possible to determine whether a gene network leads to nontrivial pattern transformations based solely on its topology. This conclusion is independent of the dimensionality of space, as none of our arguments rely on assumptions specific to spatial dimensions. While one-dimensional examples are used for clarity and illustration, the underlying reasoning applies generally. In an improved version of the article, we will clarify this point explicitly and move relevant arguments from the Supplementary Information into the main text.

Critically, our classification of gene networks is ultimately based on an argument concerning the dispersion relation associated with the network, and the construction of this dispersion relation is independent of the spatial dimensionality of the domain. In this sense, the networks identified in the text as capable of producing pattern transformations will be able to generate non-trivial pattern transformations in any spatial domain and in any number of dimensions. While the specific parameter values that permit such transformations may vary depending on the geometry and dimensionality of the domain, the existence of at least one such parameter set remains unaffected.

The geometry of the domain can influence the specific form of the resulting patterns, but it does not alter the broader class of patterns (e.g., periodic patterns, peaks emerging around a spike, etc.) that a given gene network topology can produce. One such geometric influence, commonly observed in simulations, involves boundary effects. For example, structures such as peaks or rings forming near the boundaries may appear higher, broader, or spatially shifted compared to those arising in the central regions of the domain. However, we think a pattern consisting of a periodic train of peaks where only those near the boundary are slightly different can still be classified as a periodic pattern.

"For any non-trivial pattern transformation (as long as it is symmetric around the initial spike), there exists an H gene network capable of producing it from a spike initial pattern." (Main text l.366f).

A justification for this statement is provided shortly after the claim, although we acknowledge that the current explanation is somewhat cumbersome and would benefit from a clearer presentation in a revised version of the main text.

A more detailed justification is provided in the Supplementary Information, based on three key ideas. First, any pattern (provided it is symmetric with respect to the initial spike) can be described as an arrangement of peaks with varying heights and spatial positions along a one-dimensional domain. Second, there exists a simple gene network—the diamond network—that, through parameter tuning, can produce two peaks of arbitrary height and symmetric position relative to the initial spike. Third, by placing multiple diamond networks positively upstream of a common gene product, that gene product can express peaks at each location where the upstream diamond networks induce them. Under mild additional conditions, this mechanism allows the formation of essentially any symmetric pattern. These mild conditions, along with a detailed analysis of the diamond network’s ability to generate peaks with controllable height and position, are discussed in the Supplementary Information.

"In 2D there are no peaks but concentric rings of high gene product concentration centered around the spike, while in 3D there are concentric spherical shells." (Main text l. 447ff).

This result pertains specifically to pattern transformations arising from spike initial patterns. As defined in the text, spike initial patterns are radially symmetric. Since diffusion preserves radial symmetry, pattern transformations from spike initial patterns in two or three dimensions reduce to effectively one-dimensional transformations along each radial direction. In this framework, each pair of concentration peaks symmetric with respect to the spike in one dimension corresponds to a ring surrounding the spike in two dimensions, and each ring in two dimensions becomes a hollow spherical shell around the spike in three dimensions.

We agree that including a brief section in the Supplementary Information to clarify these subtleties would be helpful for readers to better understand the generalization of certain patterns to higher dimensions.

(4) The study identifies one-signal networks and examines how combinations of these structures can give rise to minimal pattern-forming subnetworks. However, the analysis of the combinations of these minimal pattern-forming subnetworks remains relatively brief, and the manuscript does not explore how the results might change if the subnetworks were combined in upstream and downstream configurations. In our view, it is not evident that all possible gene regulatory networks can be fully characterized by these categories, nor that the resulting patterns can be reliably predicted. Rather, the approach appears more suited to identifying which known subnetworks are present within a larger network, without necessarily capturing the full dynamics of more complex configurations.

We acknowledge that our explanation regarding the combination of sub-networks was relatively brief, and we intend to address this in a revised version. Our argument that combining sub-networks does not produce qualitatively new types of pattern transformations -beyond those already described- is based on the dispersion relation. Although this relation was only detailed in the Supplementary Information, it is central to our argument and will therefore be moved to the main text. Below, we provide an outline of this argument:

Our study identifies two distinct behaviors of the principal branch of the dispersion relation at large wavenumbers. Based on this, gene networks capable of pattern formation can be classified into two categories: networks of the first kind, where the real part of the principal branch diverges to infinity as the wavenumber increases; and networks of the second kind, where the real part of the principal branch converges to a positive finite value for large wavenumbers. Naturally this argument applies to any gene network irrespectively of which, or how many, sub-networks are used to built it.

Any gene regulatory network capable of pattern formation falls into one of these two categories. We identified that networks of the first kind contain at least one Turing sub-network, whereas networks of the second kind include either an H sub-network or a noise-amplifying sub-network. In this way, the primary objective of our study -namely, achieving a topological classification of gene regulatory networks capable of pattern formation- is fulfilled. It is important to note that while the dispersion relation provides broad information about the possible resulting patterns a gene network topology can produce (e.g., periodic versus noisy), it does not specify the exact patterns that emerge for each particular set of parameter values.

Finally, regarding the shape of the resulting patterns, Figure S10 in the Supplementary Information exemplifies the notion that the behavior of combined networks can be understood as a combination of the individual behaviors of each constituent sub-network (note that the contribution of each type of sub-network in the resulting pattern is readily distinguishable). Consequently, we focus our detailed analysis on the patterning properties of the fundamental classes.

(6) The manuscript lacks a clear and detailed explanation of the underlying model and its assumptions. In particular, it is not well-defined what constitutes a "cell" in the context of the model, nor is it justified why spatial features of cells -such as their size or boundaries- can be neglected. Furthermore, the concept of the extracellular space in the one-dimensional model remains ambiguous, making it unclear which gene products are assumed to diffuse.

The size of cells is ignored in our model because we assume that they are small enough with respect to the total size of the domain that the space continuous reaction-diffusion equation (equation (1) in the main text) holds. Conceptually, one could understand cells in our model each of the pieces in an even partition of the domain into small subdomains surrounding each position x. This is anyway the standard procedure in most models of pattern formation by reaction-diffusion in embryonic development.

For extracellular signals, we assume that g(t ,x) corresponds to the concentration of the signal in the extracellular space surrounding the cell located at position x. The extracellular space is any fluid medium for which Fick Laws apply and, therfore, the Fickian diffusion term in equation (1) is valid.

For intracellular gene products, we assume that g(t ,x) corresponds to the concentration of such gene product within the cell at position x (if the gene product in hand is a transcription factor, for example), or on its surface (if it is a membrane-bound receptor). When collapsed in the continuous equations there is not such difference between being strictly within the cell or on its boundary. The only important fact is that these gene products cannot diffuse.

Regarding cell boundaries, let us consider an extracellular signal s that regulates a transcriptor factor i within cells (in our model, i is an intracellular gene product). Such regulation shall be mediated by a membrane-bound receptor, which corresponds to intracellular gene product j. In terms of the gene regulatory network this is s→ j→i. Cell boundary effects mentioned by the reviewer should be encapsulated in the specific functional form of the regulation function f(g), but they have no effect in the actual topology of the network. Consequently, they are out of the scope of this study: as we mentioned before, considering different non-linear terms for f(g) will affect the parameter range for which a gene network is capable of producing non-trivial pattern transformations, but not their overall ability to produce non-trivial pattern transformations (i.e., the existence of at least one choice of model parameters for which such transformations take place).

Finally, we would like to once again express our sincere gratitude to all reviewers for their insightful and constructive feedback. We are confident that the thorough peer review process will significantly enhance both the clarity and depth of our work. We greatly value the detailed comments provided and will carefully incorporate them in the preparation of a revised manuscript, which we intend to submit in the coming months.

eLife Assessment

This study presents a sequence-based method for predicting drug-interacting residues in intrinsically disordered proteins (IDPs), addressing an important challenge in understanding small-molecule:IDP interactions. The findings have solid support in illustrative examples that underscore the role of aromatic interactions. While predicted binding sites remain coarse, validation was done on a total of 10 IDPs, four of which thoroughly and six others less so. The method builds on previous work from the authors, with necessarily ad hoc modifications, and offers a starting point for further exploration in this emerging field.

Reviewer #1 (Public review):

Summary:

The authors developed a sequence-based method to predict drug-interacting residues in IDP, based on their recent work, to predict the transverse relaxation rates (R2) of IDP trained on 45 IDP sequences and their corresponding R2 values. The discovery is that the IDPs interact with drugs mostly using aromatic residues that are easy to understand, as most drugs contain aromatic rings. They validated the method using several case studies, and the predictions are in accordance with chemical shift perturbations and MD simulations. The location of the predicted residues serves as a starting point for ligand optimization.

Strengths:

This work provides the first sequence-based prediction method to identify potential drug-interacting residues in IDP. The validity of the method is supported by case studies. It is easy to use, and no time-consuming MD simulations and NMR studies are needed.

Weaknesses:

The method does not depend on the information of binding compounds, which may give general features of IDP-drug binding. However, due to the size and chemical structures of the compounds (for example, how many aromatic rings), the number of interacting residues varies, which is not considered in this work. Lacking specific information may restrict its application in compound optimization, aiming to derive specific and potent binding compounds.

Reviewer #2 (Public review):

Summary:

In this work, the authors introduce DIRseq, a fast, sequence-based method that predicts drug-interacting residues (DIRs) in IDPs without requiring structural or drug information. DIRseq builds on the authors' prior work looking at NMR relaxation rates, and presumes that those residues that show enhanced R2 values are the residues that will interact with drugs, allowing these residues to be nominated from the sequence directly. By making small modifications to their prior tool, DIRseq enables the prediction of residues seen to interact with small molecules in vivo.

Strengths:

The preprint is well written and easy to follow

Weaknesses:

(1) The DIRseq method is based on SeqDYN, which itself is a simple (which I do not mean as a negative - simple is good!) statistical predictor for R2 relaxation rates. The challenge here is that R2 rates cover a range of timescales, so the physical intuition as to what exactly elevated R2 values mean is not necessarily consistent with "drug interacting". Presumably, the authors are not using the helix boost component of SeqDYN here (it would be good to explicitly state this). This is not necessarily a weakness, but I think it would behove the authors to compare a few alternative models before settling on the DIRseq method, given the somewhat ad hoc modifications to SeqDYN to get DIRseq.

Specifically, the authors previously showed good correlation between the stickiness parameter of Tesei et al and the inferred "q" parameter for SeqDYN; as such, I am left wondering if comparable accuracy would be obtained simply by taking the stickiness parameters directly and using these to predict "drug interacting residues", at which point I'd argue we're not really predicting "drug interacting residues" as much as we're predicting "sticky" residues, using the stickiness parameters. It would, I think, be worth the authors comparing the predictive power obtained from DIRseq with the predictive power obtained by using the lambda coefficients from Tesei et al in the model, local density of aromatic residues, local hydrophobicity (note that Tesei at al have tabulated a large set of hydrophobicity scores!) and the raw SeqDYN predictions. In the absence of lots of data to compare against, this is another way to convince readers that DIRseq offers reasonable predictive power.

(2) Second, the DIRseq is essentially SeqDYN with some changes to it, but those changes appear somewhat ad hoc. I recognize that there is very limited data, but the tweaking of parameters based on physical intuition feels a bit stochastic in developing a method; presumably (while not explicitly spelt out) those tweaks were chosen to give better agreement with the very limited experimental data (otherwise why make the changes?), which does raise the question of if the DIRseq implementation of SeqDYN is rather over-parameterized to the (very limited) data available now? I want to be clear, the authors should not be critiqued for attempting to develop a model despite a paucity of data, and I'm not necessarily saying this is a problem, but I think it would be really important for the authors to acknowledge to the reader the fact that with such limited data it's possible the model is over-fit to specific sequences studied previously, and generalization will be seen as more data are collected.

(3) Third, perhaps my biggest concern here is that - implicit in the author's assumptions - is that all "drugs" interact with IDPs in the same way and all drugs are "small" (motivating the change in correlation length). Prescribing a specific lengthscale and chemistry to all drugs seems broadly inconsistent with a world in which we presume drugs offer some degree of specificity. While it is perhaps not unexpected that aromatic-rich small molecules tend to interact with aromatic residues, the logical conclusion from this work, if one assumes DIRseq has utility, is that all IDRs bind drugs with similar chemical biases. This, at the very least, deserves some discussion.

(4) Fourth, the authors make some general claims in the introduction regarding the state of the art, which appear to lack sufficient data to be made. I don't necessarily disagree with the author's points, but I'm not sure the claims (as stated) can be made absent strong data to support them. For example, the authors state: "Although an IDP can be locked into a specific conformation by a drug molecule in rare cases, the prevailing scenario is that the protein remains disordered upon drug binding." But is this true? The authors should provide evidence to support this assertion, both examples in which this happens, and evidence to support the idea that it's the "prevailing view" and specific examples where these types of interactions have been biophysically characterized.

Similarly, they go on to say:

"Consequently, the IDP-drug complex typically samples a vast conformational space, and the drug molecule only exhibits preferences, rather than exclusiveness, for interacting with subsets of residues." But again, where is the data to support this assertion? I don't necessarily disagree, but we need specific empirical studies to justify declarative claims like this; otherwise, we propagate lore into the scientific literature. The use of "typically" here is a strong claim, implying most IDP complexes behave in a certain way, yet how can the authors make such a claim?

Finally, they continue to claim:

"Such drug interacting residues (DIRs), akin to binding pockets in structured proteins, are key to optimizing compounds and elucidating the mechanism of action." But again, is this a fact or a hypothesis? If the latter, it must be stated as such; if the former, we need data and evidence to support the claim.

Author response:

Reviewer #1 (Public review):

Summary:

The authors developed a sequence-based method to predict drug-interacting residues in IDP, based on their recent work, to predict the transverse relaxation rates (R2) of IDP trained on 45 IDP sequences and their corresponding R2 values. The discovery is that the IDPs interact with drugs mostly using aromatic residues that are easy to understand, as most drugs contain aromatic rings. They validated the method using several case studies, and the predictions are in accordance with chemical shift perturbations and MD simulations. The location of the predicted residues serves as a starting point for ligand optimization.

Strengths:

This work provides the first sequence-based prediction method to identify potential drug-interacting residues in IDP. The validity of the method is supported by case studies. It is easy to use, and no time-consuming MD simulations and NMR studies are needed.

Weaknesses:

The method does not depend on the information of binding compounds, which may give general features of IDP-drug binding. However, due to the size and chemical structures of the compounds (for example, how many aromatic rings), the number of interacting residues varies, which is not considered in this work. Lacking specific information may restrict its application in compound optimization, aiming to derive specific and potent binding compounds.

We fully recognize that different compounds may have different interaction propensity profiles along the IDP sequence. In future studies, we will investigate compound-specific parameter values. The limiting factor is training data, but such data are beginning to be available.

Reviewer #2 (Public review):

Summary:

In this work, the authors introduce DIRseq, a fast, sequence-based method that predicts drug-interacting residues (DIRs) in IDPs without requiring structural or drug information. DIRseq builds on the authors' prior work looking at NMR relaxation rates, and presumes that those residues that show enhanced R2 values are the residues that will interact with drugs, allowing these residues to be nominated from the sequence directly. By making small modifications to their prior tool, DIRseq enables the prediction of residues seen to interact with small molecules in vivo.

Strengths:

The preprint is well written and easy to follow

Weaknesses:

(1) The DIRseq method is based on SeqDYN, which itself is a simple (which I do not mean as a negative - simple is good!) statistical predictor for R2 relaxation rates. The challenge here is that R2 rates cover a range of timescales, so the physical intuition as to what exactly elevated R2 values mean is not necessarily consistent with "drug interacting". Presumably, the authors are not using the helix boost component of SeqDYN here (it would be good to explicitly state this). This is not necessarily a weakness, but I think it would behove the authors to compare a few alternative models before settling on the DIRseq method, given the somewhat ad hoc modifications to SeqDYN to get DIRseq.

Actually, the factors that elevate R2 are well-established. These are local interactions and residual secondary structures (if any). The basic assumption of our method is that intra-IDP interactions that elevate R2 convert to IDP-drug interactions. This assumption was supported by our initial observation that the drug interaction propensity profiles predicted using the original SeqDYN parameters already showed good agreement with CSP profiles. We only made relatively small adjustments to the parameters to improve the agreement. Indeed we did not apply the helix boost portion of SeqDYN to DIRseq, and will state as such. We will also compare DIRseq with several alternative models.

Specifically, the authors previously showed good correlation between the stickiness parameter of Tesei et al and the inferred "q" parameter for SeqDYN; as such, I am left wondering if comparable accuracy would be obtained simply by taking the stickiness parameters directly and using these to predict "drug interacting residues", at which point I'd argue we're not really predicting "drug interacting residues" as much as we're predicting "sticky" residues, using the stickiness parameters. It would, I think, be worth the authors comparing the predictive power obtained from DIRseq with the predictive power obtained by using the lambda coefficients from Tesei et al in the model, local density of aromatic residues, local hydrophobicity (note that Tesei at al have tabulated a large set of hydrophobicity scores!) and the raw SeqDYN predictions. In the absence of lots of data to compare against, this is another way to convince readers that DIRseq offers reasonable predictive power.

We will compare predictions of these various parameter sets, and summarize the results in a table.

(2) Second, the DIRseq is essentially SeqDYN with some changes to it, but those changes appear somewhat ad hoc. I recognize that there is very limited data, but the tweaking of parameters based on physical intuition feels a bit stochastic in developing a method; presumably (while not explicitly spelt out) those tweaks were chosen to give better agreement with the very limited experimental data (otherwise why make the changes?), which does raise the question of if the DIRseq implementation of SeqDYN is rather over-parameterized to the (very limited) data available now? I want to be clear, the authors should not be critiqued for attempting to develop a model despite a paucity of data, and I'm not necessarily saying this is a problem, but I think it would be really important for the authors to acknowledge to the reader the fact that with such limited data it's possible the model is over-fit to specific sequences studied previously, and generalization will be seen as more data are collected.

We have explained the rationale for the parameter tweaks, which were limited to q values for four amino-acid types, i.e., to deemphasize hydrophobic interactions and slightly enhance electrostatic interactions (p. 4-5). We will add that these tweaks were motivated by observations from MD simulations of drug interactions with a-syn (ref 20). As already noted in the response to the preceding comment, we will also present results for the original parameter values as well as for when the four q values are changed one at a time.

(3) Third, perhaps my biggest concern here is that - implicit in the author's assumptions - is that all "drugs" interact with IDPs in the same way and all drugs are "small" (motivating the change in correlation length). Prescribing a specific lengthscale and chemistry to all drugs seems broadly inconsistent with a world in which we presume drugs offer some degree of specificity. While it is perhaps not unexpected that aromatic-rich small molecules tend to interact with aromatic residues, the logical conclusion from this work, if one assumes DIRseq has utility, is that all IDRs bind drugs with similar chemical biases. This, at the very least, deserves some discussion.

The reviewer raises a very important point. In Discussion, we will add that it is important to further develop DIRseq to include drug-specific parameters when data for training become available.

(4) Fourth, the authors make some general claims in the introduction regarding the state of the art, which appear to lack sufficient data to be made. I don't necessarily disagree with the author's points, but I'm not sure the claims (as stated) can be made absent strong data to support them. For example, the authors state: "Although an IDP can be locked into a specific conformation by a drug molecule in rare cases, the prevailing scenario is that the protein remains disordered upon drug binding." But is this true? The authors should provide evidence to support this assertion, both examples in which this happens, and evidence to support the idea that it's the "prevailing view" and specific examples where these types of interactions have been biophysically characterized.

We will cite several studies showing that IDPs remain disordered upon drug binding.

Similarly, they go on to say:

"Consequently, the IDP-drug complex typically samples a vast conformational space, and the drug molecule only exhibits preferences, rather than exclusiveness, for interacting with subsets of residues." But again, where is the data to support this assertion? I don't necessarily disagree, but we need specific empirical studies to justify declarative claims like this; otherwise, we propagate lore into the scientific literature. The use of "typically" here is a strong claim, implying most IDP complexes behave in a certain way, yet how can the authors make such a claim? 

Here again we will add citations to support the statement.

Finally, they continue to claim:

"Such drug interacting residues (DIRs), akin to binding pockets in structured proteins, are key to optimizing compounds and elucidating the mechanism of action." But again, is this a fact or a hypothesis? If the latter, it must be stated as such; if the former, we need data and evidence to support the claim. 

We will add citations to both compound optimization and mechanism of action.

eLife Assessment

This work presents valuable new information on the microtubule-binding mode of the microtubule kinesin-13, MCAK, the authors use quantitative single-molecule studies to propose that MCAK preferentially binds to a GDP-Pi-tubulin portion of the microtubule end. However, the evidence provided to support this claim remains incomplete and would benefit from more rigorous methodology particularly the diffraction limited experiments do not provide sufficient spatial resolution to support the authors' conclusions. In addition, a more through discussion of the existing literature would further strengthen the manuscript.

Reviewer #1 (Public review):

The authors responded to multiple criticisms with additional data and more detailed statistics, in some instances improving the quality of the work. However, I had difficulty understanding some of the authors' responses. The logic was not always apparent, the writing was occasionally confusing or would benefit from more careful wording, and some of the provided responses were superficial or raised new concerns. In some cases, the underlying data needed to support their responses were not shown. Thus, the current version of the manuscript does not sufficiently resolve the following critical issues raised by myself and other reviewers.

(1) A clear new insight into a physiological process or cellular behavior remains lacking. The study largely confirms prior observations of MCAK binding to both the microtubule wall and end. However, it is still unclear whether direct binding to the tip-as opposed to accumulation via wall diffusion or interaction with other tip-binding proteins-is a significant mechanism.

(2) The newly revealed adenosine-nucleotide-dependent binding preferences do not help clarify MCAK's catalytic function or its mechanisms of tip recognition. Consequently, the final summary figure remains speculative and is not convincingly supported by the data. It is also unclear what exactly is meant by the "working model" (figure title), or by the claim of "a simple rule of how the end-binding regulators coordinate their activities" (abstract).

(3) As noted in my previous review, the effects of adding different adenosine nucleotides on MCAK binding to microtubules are much more pronounced than the differences in MCAK binding to tubulin with various guanosine-containing nucleotides, or to lattice versus tip (e.g., Fig. 5E). Therefore, the manuscript title-"MCAK recognizes the nucleotide-dependent feature at growing microtubule ends"-does not do justice to the scale of these effects.

(4) The title implies that MCAK selectively recognizes a feature determined by the tubulin-bound guanosine nucleotide. However, the authors frequently claim that MCAK binds to the "entire GTP cap." It appears that they exclude structural protrusions from their definition of the cap, which is debatable. Even using their definition, the conclusion that MCAK recognizes a specific "nucleotide-dependent feature" seems inconsistent with the claim that it binds uniformly across the cap. These distinctions were not made clear.

(5) Some important technical details are still absent. For example, when reading the authors' response to another reviewer's question, I could not find an explanation of how the kon values for end and wall binding were calculated. These calculations clearly require assumptions, e.g. about the number of binding sites, but these details are not described. In addition, the binding data are expressed in units per tubulin dimer, which are non-standard and make comparisons to other published results difficult. There are other instances where more technical detail would be desirable, but they are too numerous to list here.

(6) Several aspects of data presentation as graphs will make it difficult for other researchers to analyze or interpret the findings. Numerical Excel-style data sheets should be provided for all measurements, including raw data-not just the ratios or derived values shown in plots. Other, more significant issues include use of mean values for non-Gaussian distributions (e.g., dwell times); binding affinities inferred from single-concentration measurements, often under varying conditions (e.g., Figs. 3C, 4); and absence of side-by-side plotted controls (e.g., Fig. 6).

(7) While the authors have added some quantitative values and descriptive detail, the manuscript still lacks a critical comparison of their findings with existing literature. This weakens the impact of the study and limits the reader's ability to place the results in a broader context.

Reviewer #4 (Public review):

The revised manuscript from Chen et al. implements many of the changes requested by the 3 reviewers of the initial submission. These changes are well-described in the corresponding Response to Reviews document. Of course, not every request from the reviewers was addressed, and the following major concerns remain:

(1) The authors argue that MCAK binds to the same region as EB proteins, which they refer to as the "EB cap". Reviewers asked for experiments that would increase the size of the EB cap to create "comets" (e.g. by increasing the microtubule growth rate); the prediction is that the MCAK signal should increase in size as well. The authors declined to pursue these experiments. As a result, the EB signals and MCAK signals are diffraction-limited spots, as opposed to the predicted exponential decay signals characteristic of EB comets. The various diffraction-limited spots are then aligned with the diffraction-limited signal of the microtubule end. These alignments and sub-pixel comparisons are technically challenging. The revised manuscript does not go far enough to provide compelling evidence that all technical challenges were overcome. Thus, while the authors can safely conclude that MCAK, EBs, and the microtubule end do occupy the same diffraction-limited spot, more precise conclusions are not supported.

(2) The reviewers criticized the initial manuscript for neglecting key references, particularly Kinoshita et al., Science 2001. Indeed, I cannot fathom writing a manuscript about MCAK and XMAP215 without putting a citation to such a landmark paper front and center. The authors have responded by including more discussion of the relevant literature (and citing Kinoshita et al.). However, the revised manuscript is often still cursory in giving credit where credit is due, contextualizing the new data, and generally engaging with the scholarship on MCAK.

(3) The data presented does not include a simple measurement of the impact of MCAK on the catastrophe frequency of microtubules. The authors explain this absence by pointing out that their movies are short (5 min) and high frame rate (10 fps). While I understand that such imaging parameters are necessary to capture single molecule end-binding events, I do not understand why a separate set of experiments could not be performed. This type of "positive control" is often missing, as pointed out by the 3 reviewers.

(4) Salt conditions, protein concentrations, and other key experimental parameters are not varied, even when varying them would provide excellent tests of the authors' hypotheses.

In summary, the revised manuscript is improved in many ways, but the interested reader should look carefully at the previous reviews and compare the measurements presented here with those of other labs.

Author response:

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

Reviewer #1 (Public Review):

Major concerns:

(1) Is the direct binding of MCAK to the microtubule cap important for its in vivo function?

a.The authors claim that their "study provides mechanistic insights into understanding the end-binding mechanism of MCAK". I respectfully disagree. My concern is that the paper offers limited insights into the physiological significance of direct end-binding for MCAK activity, even in vitro. The authors estimate that in the absence of other proteins in vitro, ~95% of MCAK molecules arrive at the tip by direct binding in the presence of ~ physiological ATP concentration (1 mM). In cells, however, the major end-binding pathway may be mediated by EB, with the direct binding pathway contributing little to none. This is a reasonable concern because the apparent dissociation constant measured by the authors shows that MCAK binding to microtubules in the presence of ATP is very weak (69 uM). This concern should be addressed by 1) calculating relative contributions of direct and EB-dependent pathways based on the affinities measured in this and other published papers and estimated intracellular concentrations. Although there are many unknowns about these interactions in cells, a modeling-based analysis may be revealing. 2) the recapitulation of these pathways using purifying proteins in vitro is also feasible. Ideally, some direct evidence should be provided, e.g. based on MCAK function-separating mutants (GDP-Pi tubulin binding vs. catalytic activity at the curled protofilaments) that contribution from the direct binding of MCAK to microtubule cap in EB presence is significant.

We thank the reviewer for the thoughtful comments.

(1) We think that the end-binding affinity of MCAK makes a significant contribution for its cellular functions. To elucidate this concept, we now use a simple model shown in Supplementary Appendix-2 (see pages 49-51, lines 1246-1316). In this model, we simplified MCAK and EB1 binding to microtubule ends by considering only these two proteins while neglecting other factors (e.g. XMAP215). Specifically, we considered two scenarios: one in which both proteins freely diffuse in the cytoplasm and another where MCAK is localized to specific cellular structures, such as the centrosome or centromere. Based on the modeling results, we argue that MCAK's functional impact at microtubule ends derives both from its intrinsic end-binding capacity and its ability to strengthen the EB1-mediated end association pathway.

(2) We agree with the reviewer that MCAK exhibiting a lower end-binding affinity (69 µM) is indeed intriguing, as one might intuitively expect a stronger affinity, e.g. in the nanomolar range. Several factors may contribute to this observation. First, this could be partly due to the in vitro system employed, which may not perfectly replicate in vivo conditions, especially when considering cellular processes quantitatively. Variations in medium composition can significantly influence the binding state. For example, reducing salt concentration leads to a marked increase in MCAK’s binding affinity (Helenius et al., 2006; Maurer et al., 2011; McHugh et al., 2019). Additionally, while numerous binding events with short durations were detected, we excluded transient interactions from our analysis to facilitate quantification. This likely leads to an underestimation of the on-rate and, consequently, the binding affinity. Moreover, to minimize the interference of purification tags (His-tag), we ensured their complete removal during protein sample preparation. Previous studies reported that retaining the His-tag of MAPs affects the binding affinity to microtubules (Maurer et al., 2011; Zhu et al., 2009). Finally, a low affinity is not necessarily unexpected. Considering the microtubule end as a receptor with multiple binding sites for MCAK, the overall binding affinity is in the nanomolar range (260 nM). This does not necessarily contradict MCAK being a microtubule dynamics regulator as only a few MCAK molecules may suffice to induce microtubule catastrophe (as discussed on page 13, lines 408-441).

(3) Ideally, we would search for mutants that specifically interfere with the binding of GDP-Pi-tubulin or the curled protofilaments. However, the mutant we tested significantly impacts the overall affinity of MCAK to microtubules (both end and lattice), making it challenging to isolate and discuss the function of MCAK with respect to the binding to GDP-Pi-tubulin alone. Additionally, we also think that the GDP-Pi-tubulin in the EB cap and the tubulin in the curved protofilaments may share structural similarities. For instance, the tubulin dimers in both states may be less compact compared to those in the lattice, which could explain why MCAK recognizes both simultaneously (Manka and Moores, 2018). However, this remains a conjecture, as there is currently no direct evidence to support it.

b. As mentioned in the Discussion, preferential MCAK binding to tubulins near the MT tip may enhance MCAK targeting of terminal tubulins AFTER the MCAK has been "delivered" to the distal cap via the EB-dependent mechanism. This is a different targeting mechanism than the direct MCAK-binding. However, the measured binding affinity between MCAK and GMPCPP tubulins is so weak (69 uM), that this effect is also unlikely to have any impact because the binding events between MCAK and microtubule should be extremely rare. Without hard evidence, the arguments for this enhancement are very speculative.

Please see our response to the comment No. 1. Additionally, we have revised our discussion to discuss the end-binding affinity of MCAK as well as its physiological relevance (please see page 13, lines 408-441; and see Supplementary Appendix-2 in pages 49-51, lines 1246-1316).

(2) The authors do not provide sufficient justification and explanation for their investigation of the effects of different nucleotides in MCAK binding affinity. A clear summary of the nucleotide-dependent function of MCAK (introduction with references to prior affinity measurements and corresponding MCAK affinities), the justifications for this investigation, and what has been learned from using different nucleotides (discussion) should be provided. My take on these results is that by far the strongest effect on microtubule wall and tip binding is achieved by adding any adenosine, whereas differences between different nucleotides are relatively minor. Was this expected? What can be learned from the apparent similarity between ATP and AMPPNP effects in some assays (Fig 1E, 4C, etc) but not others (Fig 1D,F, etc)?

We thank the reviewer for this suggestion. We have revised the manuscript accordingly, and below are the main points of our response

(1) The experiment investigating the effects of different nucleotides on MCAK binding affinity was inspired by the previous studies demonstrating that kinesin-13 interactions with microtubules are highly dependent on their adenosine-bound states. For example, kinesin-13s tightly bind microtubules and prefer to form protofilament curls or rings with tubulin in the AMPPNP state, whereas kinesin-13s are considered to move along the microtubule lattice via one-dimensional diffusion in the ADP·Pi state (Asenjo et al., 2013; Benoit et al., 2018; Friel and Howard, 2011; Helenius et al., 2006). Based on these observations, we wondered whether MCAK's adenosine-bound states might similarly affect its binding preference for growing microtubule ends. We have made the motivation clear in the revised manuscript (please see page 7, lines 199-209).

(2) Our main finding regarding the effects of nucleotides is that MCAK shows differential end-binding affinity and preference based on its nucleotide state. First, MCAK shows the greatest preference for growing microtubule ends in the ATP state, supporting the idea that diffusive MCAK (MCAK·ATP) can directly bind to growing microtubule ends. Second, MCAK·ATP also demonstrates a binding preference for GTPγS microtubules and the ends of GMPCPP microtubules. The similar trends in binding preference suggest that the affinity for GDP·Pi-tubulin and GTP-tubulin likely underpins MCAK’s preference for growing microtubule ends. To clarify these points, we have added further discussions in the manuscript (please see page 8, lines 230-233; page9, lines 258-270 and pages 13-14, lines 443-458).

(3) It is not clear why the authors decided to use these specific mutant MCAK proteins to advance their arguments about the importance of direct tip binding. Both mutants are enzymatically inactive. Both show roughly similar tip interactions, with some (minor) differences. Without a clear understanding of what these mutants represent, the provided interpretations of the corresponding results are not convincing.

We thank the reviewer for this comment. In the revised manuscript, we no longer draw conclusions about the importance of end-binding based on the mutant data. Instead, we think that the mutant data provide insights into the structural basis of the end-binding preference. Therefore, we have rewritten the results in this section to more accurately reflect these findings (please see page 10, lines 295-327).

(4) GMPCPP microtubules are used in the current study to represent normal dynamic microtubule ends, based on some published studies. However, there is no consensus in the field regarding the structure of growing vs. GMPCPP-stabilized microtubule ends, which additionally may be sensitive to specific experimental conditions (buffers, temperature, age of microtubules, etc). To strengthen the authors' argument, Taxol-stabilized microtubules should be used as a control to test if the effects are specific. Additionally, the authors should consider the possibility that stronger MCAK binding to the ends of different types of microtubules may reflect MCAK-dependent depolymerization events on a very small scale (several tubulin rows). These nano-scale changes to tubulins and the microtubule end may lead to the accumulation of small tubulin-MCAK aggregates, as is seen with other MAPs and slowly depolymerizing microtubules. These effects for MCAK may also depend on specific nucleotides, further complicating the interpretation. This possibility should be addressed because it provides a different interpretation than presented in the manuscript.

Regarding the two points raised here, our thoughts are as following

(1) The end of GMPCPP-stabilized microtubules differs from that of growing microtubules, with the most obvious known difference being the absence of the region enriched in GDP-Pi-tubulin. We consider the end of GMPCPP microtubules as an analogue of the distal tip of growing microtubules, based on two key features: (1) curled protofilaments and (2) GMPCPP-tubulin, a close analogue of GTP-tubulin. Notably, both features are present at the ends of both GMPCPP-stabilized and growing microtubules. Moreover, we agree with the suggestion to use taxol-stabilized microtubules as a control. This would eliminate the second feature (absence of GTP-tubulin), allowing us to isolate the effect of the first feature. Therefore, we conducted this experiment, and our data showed that MCAK exhibits only a mild binding preference for the ends of taxol-stabilized microtubules, which is much less pronounced than for the ends of GMPCPP microtubules. This observation supports the idea that GMPCPP-stabilized ends closely resemble the growing ends of microtubules.

(2) The reviewer suggested that stronger MCAK binding to the ends of different types of microtubules might reflect MCAK-dependent depolymerization events on a very small scale. This is an insightful possibility, which we had overlooked in the original manuscript. Fortunately, we performed the experiments at the single-molecule concentrations. Upon reviewing the raw data, we found that under ATP conditions, the binding events of MCAK were not cumulative (see Fig. X1 below) and showed no evidence of local accumulation of MCAK-tubulin aggregates.

Author response image 1.

The representative kymograph showing GFP-MCAK binding at the ends and lattice of GMPCPP microtubules in the presence of 1 mM ATP (10 nM GFP-MCAK), which corresponded to Fig. 5A. The arrow: the end-binding of MCAK. Vertical bar: 1 s; horizontal bar: 2 mm.

(5) It would be helpful if the authors provided microtubule polymerization rates and catastrophe frequencies for assays with dynamic microtubules and MCAK in the presence of different nucleotides. The video recordings of microtubules under these conditions are already available to the authors, so it should not be difficult to provide these quantifications. They may reveal that microtubule ends are different (or not) under the examined conditions. It would also help to increase the overall credibility of this study by providing data that are easy to compare between different labs.

We thank the reviewer for this suggestion. In the revised manuscript, we have provided data on the growth rates, which are similar across the different nucleotide states (Fig. s1). However, due to the short duration of our recordings (usually 5 minutes, but with a high frame rate, 10 fps), we did not observe many catastrophe events, which prevented us from quantifying catastrophe frequency using the current dataset. Since we measured the binding kinetics of MCAK during the growing phase of microtubules, the similar growth rates and microtubule end morphologies suggest that the microtubule ends are comparable across the different conditions.

Reviewer #1 (Recommendations For The Authors):

a. Please provide more details about how the microtubule-bound molecules were selected for analysis (include a description of scripts, selection criteria, and filters, if any). Fig 1A arrows do not provide sufficient information.

We first measured the fluorescence intensity of each binding event. A probability distribution of these intensities was then constructed and fitted with a Gaussian function. A binding event was considered to correspond to a single molecule if its intensity fell within μ±2σ of the distribution. The details of the single-molecule screening process are now provided in the revised manuscript (see page17, lines 574-583).

b. Evidence that MCAK is dimeric in solution should be provided (gel filtration results, controls for Figs1A - bleaching, or comparison with single GFP fluorophore).

In the revised manuscript, we provide the gel filtration results of purified MCAK and other proteins used in this study. The elution volume of the peak for GFP-MCAK corresponded to a molecular weight range between 120 kDa (EB1-GFP dimer) and 260 kDa (XMAP215-GFP-his6), suggesting that GFP-MCAK exists as a dimer (~220 kDa) under experimental condition (please see Fig.s1 and page 5, lines 104-105). In addition, we also measured the fluorescence intensity of both MCAK^sN+M and MCAK. MCAK^sN+M is a monomeric mutant that contains the neck domain and motor domain (Wang et al., 2012). The average intensity of MCAK^sN+M is 196 A.U., about 65% of that of MCAK (300 A.U.). These two measurements suggest that the purified MCAK used in this study exists dimers (see Fig. s1).

c. Evidence that MCAK on microtubules represents single molecules should be provided (distribution of GFP brightness with controls - GFP imaged under identical conditions). Since assay buffers include detergent, which is not desirable, all controls should be done using the same assay conditions. The authors should rule out that their main results are detergent-sensitive.

(1) Regarding if MCAK on microtubules represent single molecules: please refer to our responses to the two points above.

(2) To rule out the effect of tween-20 (0.0001%, v/v), we performed additional control experiments. The results showed that it has no significant effect on microtubule-binding affinity of MCAK (see Figure below).

Author response image 2.

Tween-20 (0.0001%, v/v) has no significant effect on microtubule-binding affinity of MCAK. (A) The representative projection images of GFP-MCAK (5 nM) binding to taxol-stabled GDP microtubules in the presence of 1 mM AMPPNP with or without tween-20. The upper panel showed the results of the control experiments performed without MCAK. Scale bar: 5 mm. (B) Statistical quantification of the binding intensity of GFP-MCAK binding to GDP microtubules with or without tween-20 (53 microtubules from 3 assays and 70 microtubules from 3 assays, respectively). Data were presented as mean ± SEM. Statistical comparisons were performed using the two-tailed Mann-Whitney U test with Bonferroni correction, n.s., no significance.

d. How did the authors plot single-molecule intensity distributions? I am confused as to why the intensity distribution for single molecules in Fig 1D and 2A looks so perfectly smooth, non-pixelated, and broader than expected for GFP wavelength. Please provide unprocessed original distributions, pixel size, and more details about how the distributions were processed.

In the revised manuscript, we provided unprocessed original data in Fig. 1B and Fig. 2A. We thank the reviewer for pointing out this problem.

e. Many quantifications are based on a limited number of microtubules and the number of molecules is not provided, starting from Fig 1D and down. Please provide detailed statistics and explain what is plotted (mean with SEM?) on each graph.

We performed a thorough inspection of the manuscript and corrected the identified issues.

f. Plots with averaged data should be supplemented with error bars and N should be provided in the legend. E.g. Fig 1C - average position of MT and peak positions.

We agree with the reviewer. In the revised manuscript, we have made the changes accordingly (e.g. Fig. 2C).

g. Detailed information should be provided about protein constructs used in this work including all tags. The use of truncated proteins or charged/bulky tags can modify protein-microtubule interactions.

We agree with the reviewer. In the revised manuscript, we provide the information of all constructs (see Fig. s1 and the related descriptions in Methods, pages 15-16, lines 476-534).

h. Line 515: We estimated that the accuracy of microtubule end tracking was ~6 nm by measuring the standard error of the distribution of the estimated error in the microtubule end position. - evidence should be provided using the conditions of this study, not the reference to the prior work by others.

i. Line 520: We estimated that the accuracy of the measured position was ~2 nm by measuring the standard error of the fitting peak location". Please provide evidence.

Point h-i: we now provide detailed descriptions of how to estimate tracking and measurement accuracy and error in our work. Please see pages 18-19, lines 626-645.

j. Kymographs in Fig 5G are barely visible. Please provide single-channel greyscale images. What are the dim molecules diffusing on this microtubule?

We have incorporated the changes suggested by the reviewer. We think that some of the dim signals may result from stochastic background noise, while others likely represent transient bindings of MCAK. The exposure time in our experiments was approximately 0.05 seconds; if the binding duration were shorter than this, the signal would be lower (i.e. the “dim” signals). It is important to note that in this study, we selected binding events lasting at least 2 consecutive frames, meaning transient binding events were not included. This point has been clarified in the Methods section (see page17, lines 573-583).

k. Please provide a methods description for Fig 6. Did the buffer include 1 mM ATP? The presence of ATP would make these conditions more physiological. ATP concentration should be stated clearly in the main text or figure legend.

The buffer contains ATP. In the revised manuscript, we have provided the methods for the experiments of microtubule dynamics assay, as well as the analysis of microtubule lifetimes and catastrophe frequency (see page 17, lines 561-572 and page 20, lines 685-690).

l. Line 104: experiment was performed in BRB80 supplemented with 50 mM KCl and 1 mM ATP, providing a nearly physiological ion strength. Please provide a reference or add your calculations in Methods.

We have provided references on page 5, lines 101-104 of our manuscript.

m. What was the MCAK concentration in Figure 4? Did the microtubule shorten under any of these conditions?

In these experiments, we used a very low concentration of MCAK and taxol-stabilized microtubules, so there’s no microtubule shortening observed here. ATP: 10 nM GFP-MCAK; AMPPNP: 1 nM GFP-MCAK; ADP: 10 nM GFP-MCAK; APO state: 0.1 nM GFP-MCAK.

Other criticism:

Text improvements are recommended in the Discussion. For example, line 348: Fourth, the loss of the binding preference.. suggests that the binding preference .. is required for the optimal .. preference.

We thank the reviewer for pointing out this. In the revised manuscript, we conducted a thorough revision and review of the text.

Reviewer #2 (Public Review):

Summary:

In this manuscript, Chen et al. investigate the localization of microtubule kinesin-13 MCAK to the microtubule ends. MCAK is a prominent microtubule depolymerase whose molecular mechanisms of action have been extensively studied by a number of labs over the last ~twenty years. Here, the authors use single-molecule approaches to investigate the precise localization of MCAK on growing microtubules and conclude that MCAK preferentially binds to a GDP-Pi-tubulin portion of the microtubule end. The conclusions are speculative and not well substantiated by the data, making the impact of the study in its current form rather limited. Specifically, greater effort should be made to define the region of MCAK binding on microtubule ends, as well as its structural characteristics. Given that MCAK has been previously shown to effectively tip-track growing microtubule ends through an established interaction with EB proteins, the physiological relevance of the present study is unclear. Finally, the manuscript does not cite or properly discuss a number of relevant literature references, the results of which should be directly compared and contrasted to those presented here.

We thank the reviewer for the comments. As these suggestions are more thoroughly expressed in the following comments for authors, we will provide the responses in the corresponding sections, as shown below.

Reviewer #2 (Recommendations For The Authors):

Significant concerns:

(1) Establishing the precise localization of MCAK wrt microtubule end is highly non-trivial. More details should be provided, including substantial supplementary data. In particular, the authors claim ~6 nm accuracy in microtubule end positioning - this should be substantiated by data showing individual overlaid microtubule end intensity profiles as well as fits with standard deviations etc. Furthermore, to conclude that MCAK binds behind XMAP215, the authors should look at the localization of the two proteins simultaneously, on the same microtubule end. Notably, EB binding profiles are well known to exponentially decay along the microtubule lattice - this is not very apparent from the presented data. If MCAK's autonomous binding pattern matches that of EB, we should be seeing an exponentially-decaying localization for MCAK as well? However, averaged MCAK signals seem to only be fitted to Gaussian. Note that the EB binding region (i.e. position and size of the EB comet) can be substantially modulated by increasing the microtubule growth rate - this can be easily accomplished by increasing tubulin concentrations or the addition of XMAP215 (e.g. see Maurer et al. Cur Bio 2014). Thus to establish that MCAK on its own binds the same region as EB, experiments that directly modulate the size and the position of this region should be added.

(1) We thank the reviewer for this comment. Regarding the accuracy in microtubule end positioning, we now provide more details, and please see pages 18-19, lines 625-645 in the revised manuscript.

(2) Regarding the relative localization of XMAP215 and MCAK, we performed additional experiments to record their colocalizations simultaneously, on the same microtubule end. Our results showed that MCAK predominantly binds behind XMAP215, with 14.5% appearing within the XMAP215’s binding region. Please see Fig. 2.D-E and lines 184-197 in the revised manuscript.

(3) Regarding the exponential decay of the EB1 signal along microtubules, we observed that the position probability distribution measured in the present study follows a Gaussian distribution, and the expected exponential decay was not apparent. Since the exponential decay is thought to result from the time delay between tubulin polymerization and GTP hydrolysis, slower polymerization is expected to reduce this latency (Maurer et al., 2014). In our experiments, the growth rate was relatively low (~0.7 mm/min), much slower than the rate observed in cells, where the comet-shaped EB1 signal is most pronounced. The previous study has shown that the exponential decay of EB1 is more pronounced at growth rates exceeding 3 mm/min in vitro (Maurer et al., 2014). Therefore, we think that the relatively slow growth may account for the observed non-exponential decay distribution of the EB1 signals. The same reason may also explain the distribution of MCAK.

(4) We agree with the reviewer’s suggestion that altering microtubule growth rate is a valid and effective approach to regulate the EB cap length. However, the conclusion that MCAK binds to the EB region is supported by three lines of evidence: (1) the localization of MCAK at the ends of microtubules, (2) new experimental data showing that MCAK binds to the proximal end of the XMAP215 site, and (3) the tendency of MCAK to bind GTPγS microtubules, similar to EB1. Based on these findings, we did not pursue additional experiments to modify the length of the EB cap.

(2) Even if MCAK indeed binds behind XMAP215, there is no evidence that this region is defined by the GDP-Pi nucleotide state; it could still be curved protofilaments. GTPyS is an analogue of GTP - to what extent GTPyS microtubules exactly mimic the GDP-Pi-tubulin state remains controversial. Furthermore, nucleotide sensing for EB is thought to be achieved through its binding at the interface of four tubulin dimers. However MCAK's binding site is distinct, and it has been shown to recognize intradimer tubulin curvature. Thus it is not clear how MCAK would sense the nucleotide state. On the other hand, there is mounting evidence that the morphology of the growing microtubule end can be highly variable, and that curved protofilaments may be protruding off the growing ends for tens of nanometers or more, previously observed both by EM as well as by fluorescence (e.g. Mcintosh, Moores, Chretien, Odde, Gardner, Akhmanova, Hancock, Zanic labs). Thus, to establish that MCAK indeed localizes along the closed lattice, EM approaches should be used.

First, we conducted additional experiments that demonstrate MCAK indeed binds behind XMAP215, supporting the conclusion that MCAK interacts with the EB cap (please see Fig. 2 in the revised manuscript). Second, our argument that MCAK preferentially binds to GDP-Pi tubulin is based on two observations: (1) the binding regions of MCAK overlap with those of EB1, and (2) MCAK preferentially binds to GTPγS microtubules, which are considered a close analogue of GDP-Pi tubulin. Third, understanding the structural basis of how MCAK senses the nucleotide state of tubulin is beyond the scope of the present study. However, inspired by the reviewer’s suggestion, we looked into the structure of the MCAK-tubulin complex. The L2 loop of MCAK makes direct contact with the interdimer interface (Trofimova et al., 2018; Wang et al., 2017), which could provide a structural basis for recognizing the changes induced by GTP hydrolysis. While this remains a hypothesis, it is certainly a promising direction for future research. Forth, we agree with the reviewer that an EM approach would be ideal for establishing that MCAK localizes along the closed lattice. However, this is not the focus of the current study. Instead, we argue that MCAK binds to the EB cap, where at least some lateral interactions are likely to have formed.

(3) The physiological relevance of the study is rather questionable: MCAK has been previously established to be able to both diffuse along the microtubule lattice (e.g. Helenius et al.) as well as hitchhike on EBs (Gouveia et al.). Given the established localization of EBs to growing microtubule ends in cells, and apparently higher affinity of MCAK for EB vs. the microtubule end itself (although direct comparisons with the literature have not been reported here), the relevance of MCAK's autonomous binding to dynamic microtubule ends is dubious.

We thank the reviewer for raising the importance of physiological relevance. Please refer to our response to the comment No.1 of reviewer 1. Briefly, we think that the end-binding affinity of MCAK makes a significant contribution for its cellular functions. To elucidate this concept, we now use a simple model shown in Supplementary Appendix-2 (see pages 49-51, lines 1246-1316). In this model, we simplified MCAK and EB1 binding to microtubule ends by considering only these two proteins while neglecting other factors (e.g. XMAP215). Specifically, we considered two scenarios: one in which both proteins freely diffuse in the cytoplasm and another where MCAK is localized to specific cellular structures, such as the centrosome or centromere. Based on the modeling results, we argue that MCAK's functional impact at microtubule ends derives both from its intrinsic end-binding capacity and its ability to strengthen the EB1-mediated end association pathway.

(4) Finally, the study seriously lacks discussion of and comparison with the existing literature on this topic. There are major omissions in citing relevant literature, such as e.g. landmark study by Kinoshita et al. Science 2001. Several findings reported here directly contradict previous findings in the literature. Direct comparison with e.g. Gouveia et al findings, Helenius et al. findings, and others need to be included. For example, Gouveia et al reported that EB is necessary for MCAK plus-end-tracking in vitro (please see Figure 1 of their manuscript). The authors should discuss how they reconcile the differences in their findings when compared to this earlier study.

We thank the reviewer for this helpful suggestion. In the revised manuscript, we have updated the text description and included comparative discussions with other relevant studies in the Discussion section. Specifically, we added comparisons with the research on XMAP215 in page 14, lines 459-472 (Barr and Gergely, 2008; Kinoshita et al., 2001; Tournebize et al., 2000). Additionally, we have compared our findings with those of Gouveia et al. and Helenius et al. regarding MCAK's preference for binding microtubule ends in page 6, lines 145-157 and page 13, 408-441, respectively (Gouveia et al., 2010; Helenius et al., 2006).

Additional specific comments:

Figure 1

Gouveia et al. (Figure 1) reported that MCAK does not autonomously preferentially localize to growing tips. Specifically, Gouveia et al. found equal association rates of MCAK to both the lattice and the tip in the presence of EB3delT, an EB3 construct that does not directly interact with MCAK. How can these findings be reconciled with the results presented here?

We are uncertain why there was no observed difference in the on-rates to the lattice and the end in the study by Gouveia et al. Even when considering only the known affinity of MCAK for curved protofilaments at the distal tip of growing microtubules, we would still expect to observe an end-binding preference. After carefully comparing the experimental conditions, we nevertheless identified some differences. First, we used a 160 nm tip size to calculate the on-rate (k_on), whereas Gouveia et al. used a 450 nm tip. Using a longer tip size would naturally lead to a smaller(k_on) value. Note that we chose 160 nm for several reasons: (i) a previous cryo-electron tomography study has elucidated that the sheet structures of dynamic microtubule ends have an average length of around 180 nm (Guesdon et al., 2016); (ii) Analysis of fluorescence signals at dynamic microtubule ends has demonstrated that the taper length at the microtubule end is less than 180 nm (Maurer et al., 2014); (iii) in the present study, we estimated that the length of MCAK's end-binding region is approximately 160 nm. Second, in Gouveia et al., single-molecule binding events were recorded in the presence of 75 nM EB3ΔT, which could potentially create a crowded environment at the tip, reducing MCAK binding. Third, as mentioned in our response to Reviewer 1, we took great care to minimize the interference from purification tags (e.g., His-tag) by ensuring their complete removal during protein preparation. Previous studies reported that retaining the His-tag of MAPs led to a significant increase in binding for microtubules (Maurer et al., 2011; Zhu et al., 2009). We believe that some of the factors mentioned above, or their combined effects, may account for the differences in these two observations.

1C shows the decay of tubulin signal over several hundred nm - should show individual traces? How aligned? Doesn't this long decay suggest protruding protofilaments? (E.g. Odde/Gardner work).

(1) In the revised manuscript, we now show individual traces (e.g. in Fig. 1B and Fig. 2A). The average trace for tubulin signal with standard deviation was shown in Fig. 2C.

(2) The microtubule lattice was considered as a Gaussian wall and its end as a half-Gaussian in every frame. Use the peak position of the half-Gaussian of every frame to align and average microtubule end signals, during the dwell time. The average microtubule ends' half-Gaussion peak used as a reference to measure the intensity profile of individual single-molecule binding event in every frame (see page18, lines 607-624).

(3) We think that the decay of tubulin signal results from the convolution of the tapered end structure and the point spread function. In the revised manuscript, we have updated the Figures to provide unprocessed original data in Fig. 1B and Fig. 2A.

Please show absolute numbers of measurements in 1C (rather than normalized distribution only).

In the revised manuscript, we have included the raw data for both tubulin and MCAK signals as part of the methods description. In Fig. 1, using normalized values allows for the simultaneous representation of microtubule and protein signals on a unified graph.

How do the results in 1D-G compare with the previous literature? Particularly comparison of on-rates between this study and the Gouveia et al? Assuming 1 um = 1625 dimers, it appears that in the presence of EB3, the on-rate of MCAK to the tips reported in Gouveia et al. is an order of magnitude higher than reported here in the absence of EB3 (4.3 x 10E-4 vs. 2 x 10E-5). If so, and given the robust presence of EB proteins at growing microtubule ends in cells, this would invalidate the potential physiological relevance of the current study. Note that the dwell times measured in Gouveia et al. are also longer than those measured here.

Note that in Gouveia et al, the concentration of mCherry-EB3 was 75 nM, about 187.5 times higher than that of MCAK (0.4 nM). The relative concentrations of these two proteins are not always the case in cells. Regarding the physiological relevance of the end-binding affinity of MCAK itself, please refer to our response to the point No.1 of Reviewer 1.

Notably, Helenius et al reported a diffusion constant for MCAK of 0.38 um^2/s, which is more than an order of magnitude higher than reported here. The authors should comment on this!

In the revised manuscript, we have provided an explanation for the difference in diffusion coefficient. Please see page 6, line 142-157. In short, low salt condition facilitates rapid diffusion of MCAK.

Figure 2:

This figure is critical and really depends on the analysis of the tubulin signal. Note significant variability in tubulin signal between presented examples in 2A. Also, while 2C looks qualitatively similar, there appears to be significant variability over the several hundred nm from the tip along the lattice. This is the crucial region; statistical significance testing should be presented. More detailed info, including SDs etc. is necessary.

In the revised manuscript, we have provided raw data in Fig. 1B and Fig. 2A. Additionally, we have provided statistical analysis on the tubulin signals (Fig. 2C) and performed significance test. Please see page 5, lines 111-116 and page 7, lines 179-183 for detailed descriptions.

Insights into the morphology of microtubule ends based on TIRF imaging have been previously gained in the literature, with reports of extended tip structures/protruding protofilaments (see e.g. Coombes et al. Cur Bio 2013, based on the methods of Demchouk et al. 2011). Such analysis should be performed here as well, if we are to conclude that nucleotide state alone, as opposed to the end morphology, specifies MCAK's tip localization.

We appreciate the reviewer’s suggestion and agree that it provides a valid optical microscopy-based approach for estimating microtubule end morphology. However, this method did not establish a direct correlation between microtubule end morphology and tubulin nucleotide status. Therefore, we think that refining the measurement of microtubule end morphology will not necessarily provide more information to the understanding of tubulin nucleotide status at MCAK binding sites. Based on the available data in the present study, there are two main pieces of evidence supporting the idea that MCAK can sense tubulin nucleotide status: (1) the binding regions of MCAK and EB overlap significantly, and (2) MCAK shows a clear preference for binding to GTPγS microtubules, similar to EB1 (we provide a new control to support this, Fig. s4). Of course, we do not consider this to be a perfect set of evidence. As the reviewer has pointed out here and in other suggestions, future work should aim to further distinguish the nucleotide status of tubulin in the dynamic versus non-dynamic regions at the ends of microtubules, and to investigate the structural basis by which MCAK recognizes tubulin nucleotide status.

EB comet profile should be clearly reproduced. MCAK should follow the comet profile.

Please see our 3^rd response to the point 1 of this reviewer.

The conclusion that the MCAK binding region is larger than XMAP215 is not firm, based on the data presented. The authors state that 'the binding region of MCAK was longer than that of XMAP215'. What is the exact width of the region of the XMAP215 localization and how much longer is the MCAK end-binding region? Is this statistically significant?

We have revised this part in the revised manuscript (page 6, lines 167-172). The position probability distributions of MCAK and XMAP215 were significantly different (K-S test, p< 10^-5), and the binding region of MCAK (FWHM=185 nm) was significantly longer than that of XMAP215 (FWHM=123 nm).

MCAK localization with AMPPNP should also be performed here. Even low concentrations of MCAK have been shown to induce microtubule catastrophe/end depolymerization. This will dramatically affect microtubule end morphology, and thus apparent positioning of MCAK at the end.

In the end positioning experiment, we used a low concentration of MCAK (1 nM). Under this condition, microtubule dynamics remained unchanged, and the morphology of the microtubule ends was comparable across different conditions (with EB1, MCAK or XMAP215). Additionally, in the revised manuscript, we present a new experiment in which we recorded the localization of both MCAK and XMAP215 on the same microtubule. The results support the conclusion regarding their relative localization: most MCAK is found at the proximal end of the XMAP215 binding region, while approximately 15% of MCAK is located within the XMAP215 binding region. Please see Fig. 2D-E and page 7, lines 184-197 for the corresponding descriptions.

Figure 3:

For clearer presentation, projections showing two microtubule lattice types on the same image (in e.g. two different colors) should be shown first without MCAK, and then with MCAK.

We thank the reviewer for this suggestion. We have adjusted the figure accordingly. Please see Fig. 4 in the revised manuscript.

Please comment on absolute intensity values - scales seem to be incredibly variable.

The fluorescence value presented here is the result of multiple images being summed. Therefore, the difference in absolute values is influenced not only by the binding affinity of MCAK in different states to microtubules, but also by the number of images used. In this analysis, we are not comparing MCAK in different states, but rather evaluating the binding ability of MCAK in the same state on different types of microtubules.

Given that the authors conclude that MCAK binding mimics that of EB, EB intensity measurements and ratios on different lattice substrates should be performed as a positive control.

We performed additional experiments with EB1, in the revised manuscript, we provide the data as a positive control (please see Fig. s4).

Figure 4:

MCAK-nucleotide dependence of GMPCPP microtubule-end binding has been previously established (see e.g. Helenius et al, others?) - what is new here? Need to discuss the literature. This would be more appropriate as a supplemental figure?

In the present study, we reproduced the GMPCPP microtubule-end binding of MCAK in the AMPPNP state, as shown in several previous reports (Desai et al., 1999; Hertzer et al., 2006). Here, we also quantified the end to lattice binding preference, and our results showed that the nucleotide state-dependence shows the same trend as the binding preference of MCAK to the growing microtubule ends. Therefore, we prefer to keep this figure in the main text (Fig. 5).

Figure 5:

Please note that both MCAK mutants show an additional two orders of magnitude lower microtubule binding on-rates when compared to wt MCAK. This makes the analysis of preferential binding substrate for these mutants dubious.

We agreed with this point. We have rewritten this part. Please see page 10, lines 295-327, in the revised manuscript.

Figure 6:

Combined effects of XMAP215 and XKCM1 (MCAK) have been previously explored in the landmark study by Kinoshita et al. Science 2001, which should be cited and discussed. Also note that Moriwaki et al. JCB 2016 explored the combined effects of XMA215 and MCAK - which should be discussed here and compared to the current results.

We agree with the reviewer. We have revised the discussion on this part. Please see page 11, lines 329-342 and page 14, lines 459-472 in the revised manuscript.

Please report quantification for growth rate and lifetime.

In the revised manuscript, we provide all these data. Please see pages 11-12, lines 343-374.

To obtain any new quantitative information on the combined effects of the two proteins, at the very minimum, the authors should perform a titration in protein concentration.

We agree with the reviewer on this point. In our pilot experiments, we performed titration experiments to determine the appropriate concentrations of MCAK and XMAP215, respectively. We selected 50 nM for XMAP215, as it clearly enhances the growth rate and exhibits a mild promoting effect on catastrophe—two key effects of XMAP215 reported in previous studies (Brouhard et al., 2008; Farmer et al., 2021). Reducing the XMAP215 concentration eliminates the catastrophe-promoting effect, while increasing it would not much enhance the growth rate. For MCAK, we chose 20 nM, as it effectively promotes catastrophe; increasing the concentration beyond this point leads to no microtubule growth, at least in the MCAK-only condition. If there’s no microtubule growth, it would be difficult to quantify the parameters of microtubule dynamics, hindering a clear comparison of the combined versus individual effects. Therefore, we think that the concentrations used in this study are appropriate and representative. In the revised manuscript, we make this point clearer (see pages 11 and lines 329-342).

Finally, the writing could be improved for overall clarity.

We thank the reviewer for pointing out this. In the revised manuscript, we conducted a thorough revision and review of the text.

Reviewer #3 (Public Review):

The authors revisit an old question of how MCAK goes to microtubule ends, partially answered by many groups over the years. The authors seem to have omitted the literature on MCAK in the past 10-15 years. The novelty is limited due to what has previously been done on the question. Previous work showed MCAK targets to microtubule plus-ends in cells through association with EB proteins and Kif18b (work from Wordeman, Medema, Walczak, Welburn, Akhmanova) but none of their work is cited.

We thank the reviewer for the suggestion. Some of the referenced work has already been cited in our manuscript, such as studies on the interaction between MCAK and EB1. However, other relevant literature had not been properly cited. In the revised manuscript, we have added further discussion on this topic in the context of existing findings. Please refer to pages 3-4, lines 68-85, and pages 13, lines 425-441.

It is not obvious in the paper that these in vitro studies only reveal microtubule end targeting, rather than plus end targeting. MCAK diffuses on the lattice to both ends and its conformation and association with the lattice and ends has also been addressed by other groups-not cited here. I want to particularly highlight the work from Friel's lab where they identified a CDK phosphomimetic mutant close to helix4 which reduces the end preference of MCAK. This residue is very close to the one mutated in this study and is highly relevant because it is a site that is phosphorylated in vivo. This study and the mutant produced here suggest a charge-based recognition of the end of microtubules.

Here the authors analyze this MCAK recognition of the lattice and microtubule ends, with different nucleotide states of MCAK and in the presence of different nucleotide states for the microtubule lattice. The main conclusion is that MCAK affinity for microtubules varies in the presence of different nucleotides (ATP and analogs) which was partially known already. How different nucleotide states of the microtubule lattice influence MCAK binding is novel. This information will be interesting to researchers working on the mechanism of motors and microtubules. However, there are some issues with some experiments. In the paper, the authors say they measure MCAK residency of growing end microtubules, but in the kymographs, the microtubules don't appear dynamic - in addition, in Figure 1A, MCAK is at microtubule ends and does not cause depolymerization. I would have expected to see depolymerization of the microtubule after MCAK targeting. The MCAK mutants are not well characterized. Do they still have ATPase activity? Are they folded? Can the authors also highlight T537 and discuss this?

Finally, a few experiments are done with MCAK and XMAP215, after the authors say they have demonstrated the binding sites overlap. The data supporting this statement were not obvious and the conclusions that the effect of the two molecules are additive would argue against competing binding sites. Overall, while there are some interesting quantitative measurements of MCAK on microtubules - in particular in relation to the nucleotide state of the microtubule lattice - the insights into end-recognition are modest and do not address or discuss how it might happen in cells. Often the number of events is not recorded. Histograms with large SEM bars are presented, so it is hard to get a good idea of data distribution and robustness. Figures lack annotations. This compromises therefore their quantifications and conclusions. The discussion was hard to follow and needs streamlining, as well as putting their work in the context of what is known from other groups who produced work on this in the past few years.

We thank the reviewer for the comments. Regarding the physiological relevance of the end-binding of MCAK itself, please refer to our response to the point No.1 of reviewer 1. Moreover, as we feel that other suggestions are more thoroughly expressed in the following comments for authors, we will provide the responses in the corresponding sections, as shown below.

Reviewer #3 (Recommendations For The Authors):

Why, on dynamic microtubules, is MCAK at microtubule plus ends and does not cause a catastrophe?

At this concentration (10 nM MCAK with 16 mM tubulin in Fig. 1; 1 nM MCAK with 12 mM tubulin in Fig. 2), MCAK has little effect on microtubule dynamics in our experiments. Using TIRFM, we were able to observe individual MCAK binding events. Based on these observations, we think that in the current experimental condition, a single binding event of MCAK is insufficient to induce microtubule catastrophe; rather, it likely requires cumulative changes resulting from multiple binding events.

Do the MCAK mutants still have ATPase activity?

The ATPase activities of MCAK^K525A and MCAK^V298S are both reduced to about 1/3 of the wild-type (Fig. s6).

The intensities of GFP are not all the same on the microtubule lattice (eg 1A). See blue and white arrowheads. The authors could be looking at multiple molecules of GFP-MCAK instead of single dimers. How do they account for this possibility?

In the revised manuscript, we provide the gel filtration result of the purified MCAK, and the position of the peak corresponds to ~220 kDa, demonstrating that the purified MCAK in solution is dimeric (please see Fig.s1 and page 5, lines 101-103). We measured the fluorescence intensity of each binding event. A probability distribution of these intensities was then constructed and fitted with a Gaussian function. A binding event was considered to correspond to a single molecule if its intensity fell within μ±2σ of the distribution. The details of the single-molecule screening process are provided in the revised manuscript (see page 17, lines 574-583).

In addition, we also measured the fluorescence intensity of both MCAK^sN+M and MCAK. MCAK^sN+M is a monomeric mutant that contains the neck domain and motor domain (Wang et al., 2012). The average intensity of MCAK^sN+M is 196 A.U., about 65 % of that of MCAK (300 A.U.), suggesting that MCAK is a dimer (see Fig. s1). Moreover, we think that some of the dim signals may result from stochastic background noise, while others likely represent transient bindings of MCAK. The exposure time in our experiments was approximately 0.05 seconds; if the binding duration were shorter than this, the signal would be lower. It is important to note that in this study, we specifically selected binding events lasting at least 2 consecutive frames, meaning transient binding events were not included. This point has been clarified in the Methods section (see page 17, lines 568-569 and lines 574-583).

Could the authors provide a kymograph of an MT growing, in the presence of MCAK+AMPPNP? Can MCAK track the cap?

Under single-molecule conditions, we observed a single MCAK molecule briefly binding to the end of the microtubule. However, we did not record if MCAK at high concentrations could track microtubule ends under AMPPNP conditions.

In the experiments in Figure 6, the authors should also show the localization of MCAK and XMAP215 at microtubule plus ends in their kymographs to show the two molecules overlap.

Regarding the relative localization of XMAP215 and MCAK, we conducted additional experiments to record their colocalization simultaneously at the same microtubule end. Our results show that MCAK predominantly binds behind XMAP215, with 14.5% of MCAK binding within the XMAP215 binding region. Please see Fig. 2.D-E and page 7, lines 184-197 in the revised manuscript. However, we argue that the effects of XMAP215 and MCAK are additive, and their binding sites do not necessarily need to overlap for these effects to occur.

The authors do not report what statistical tests are done in their graphs, and one concern is over error propagation of their data. Instead of bar graphs, showing the data points would be helpful.

We have now shown all data points in the revised manuscript.

MCAK+AMPPNP accumulates at microtubule ends. Appropriate quotes from previous work should be provided.

We have made the revisions accordingly. Please see page 9, lines 273-276.

Controls are missing. An SEC profile for all purified proteins should be presented. Also, the authors need to explain if they report the dimeric or monomeric concentration of MCAK, XMAP215, etc...

We have provided the gel filtration result for all purified proteins in the revised manuscript (Fig.s1). Moreover, we now make it clear that the concentrations of MCAK and EB1 are monomeric concentration. Please see the legend for Fig. 1, line 893 in the revised manuscript.

Figure 1: the microtubules don't look dynamic at all. This is also why the authors can record MCAK at microtubule ends, because their structure is not changing.

The microtubules are dynamic, but they may appear non-dynamic due to the relatively slow growth rate and the high frame rate at which we are recording. We propose that individual binding events of MCAK induce structural changes at the nanoscopic or molecular scale, which are not detectable using TIRFM.

I recommend the authors measure the Kon and Koff for single GFP-MCAK mutant molecules and provide the information alongside their normalized and averaged binding intensities of GFP-MCAK in Fig 5. Showing data points instead of bar graphs would be better.

(1) We measured k_on and dwell time for mutants at growing microtubule end. However, we did not perform single-molecule tracking for MCAK’s binding on stabilized microtubules. This is mainly because the superimposed signal on the stable microtubule already indicates the changes in the mutant's binding affinity to different microtubule structures, and moreover, the binding of the mutants is highly transient, making accurate single-molecule tracking and calculations difficult.

(2) In the revised figure, we have included the data points in all plots.

When discussing how Kinesin-13 interacts with the lattice, the authors should quote the papers that report the organization of full-length Kinesin-13 on tubulin heterodimers: Trofimova et al, 2018; McHugh et al 2019; Benoit et al, 2018. It would reinforce their model and account for the full-length protein, rather than just the motor domain.

We thank the suggestion for the reviewer. In our manuscript, we have cited papers on full-length Kinesin-13 to discuss the interaction between MCAK and microtubule end-curved structure. Additionally, we have utilized the MCAK-tubulin crystal structure (PDB ID: 5MIO) in Fig. 6, as it depicts a human MCAK, which is consistent with the protein used in our study. This structure illustrates the interaction sites between MCAK and tubulin dimer, guiding our mutation studies on specific residues. Thus, we prefer to use the structure (PDB ID: 5MIO) in Fig.6.

Figure 5A. What type of model is this? A PDB code is mentioned. Is this from an X-ray structure? If so, mention it.

We have now included the structural information in the Figure legend (see page 37, lines 1045).

Figure 5B. It is not possible to distinguish the different microtubule lattices (GTPyS, GDP, and GMPCPP). The experiment needs to be better labelled.

We thank the reviewer for this comment. We have now rearranged the figure for better clarity (see Fig. 6).

"Figure 5D: what are the statistical tests? I don't understand " The statistical comparisons were made versus the corresponding value of 848 GFP-MCAK".

We have made this point clearer in the revised manuscript (see pages 38, line 1078-1080).

What is the "EB cap"? This needs explaining.

We provide this explanation for this, please see page 4, lines 87-89 in the revised manuscript.

Work from Friel and co-workers showed MCAK T537E did not have depolymerizing activity and a reduced affinity for microtubule ends. The work of the authors should be discussed with respect to this previously published work.

We thank the reviewer for this suggestion. In the revised manuscript, we have added discussions on this (see page 10, lines 303-307).

The concentration of protein used in the assays is not always described.

We have checked throughout the manuscript and made revisions accordingly.

"Having revealed the novel binding sites of MCAK in dynamic microtubule ends " should be on "we wondered how MCAK may work ..with EB1". This is not addressed so should be removed. Instead, they can quote the work from Akhmanova's lab. Realistically this section should be rephrased as there are other plus-end targeting molecules that compete with MCAK, not just XMAP215 and EB1.

We have rephrased this section as suggested by this reviewer to be more specific. Please see page 11, lines 329-342.

What is AMPCPP?

It should be “AMPPNP”

Typos in Figure 5.

Corrected

eLife Assessment

By taking advantage of noise in gene expression, this important study introduces a new approach for detecting directed causal interactions between two genes without perturbing either. The main theoretical result is supported by a proof. Preliminary simulations and experiments on small circuits are solid, but further investigations are needed to demonstrate the broad applicability and scalability of the method.

Reviewer #2 (Public Review):

Summary:

This paper describes a new approach to detecting directed causal interactions between two genes without directly perturbing either gene. To check whether gene X influences gene Z, a reporter gene (Y) is engineered into the cell in such a way that (1) Y is under the same transcriptional control as X, and (2) Y does not influence Z. Then, under the null hypothesis that X does not affect Z, the authors derive an equation that describes the relationship between the covariance of X and Z and the covariance of Y and Z. Violation of this relationship can then be used to detect causality.

The authors benchmark their approach experimentally in several synthetic circuits. In 4 positive control circuits, X is a TetR-YFP fusion protein that represses Z, which is an RFP reporter. The proposed approach detected the repression interaction in 2 of the 4 positive control circuits. The authors constructed 16 negative control circuit designs in which X was again TetR-YFP, but where Z was either a constitutively expressed reporter, or simply the cellular growth rate. The proposed method detected a causal effect in two of the 16 negative controls, which the authors argue is perhaps not a false positive, but due to an unexpected causal effect. Overall, the data support the potential value of the proposed approach.

Strengths:

The idea of a "no-causality control" in the context of detected directed gene interactions is a valuable conceptual advance that could potentially see play in a variety of settings where perturbation-based causality detection experiments are made difficult by practical considerations.

By proving their mathematical result in the context of a continuous-time Markov chain, the authors use a more realistic model of the cell than, for instance, a set of deterministic ordinary differential equations.

The authors have improved the clarity and completeness of their proof compared to a previous version of the manuscript.

Limitations:

The authors themselves clearly outline the primary limitations of the study: The experimental benchmark is a proof of principle, and limited to synthetic circuits involving a handful of genes expressed on plasmids in E. coli. As acknowledged in the Discussion, negative controls were chosen based on the absence of known interactions, rather than perturbation experiments. Further work is needed to establish that this technique applies to other organisms and to biological networks involving a wider variety of genes and cellular functions. It seems to me that this paper's objective is not to delineate the technique's practical domain of validity, but rather to motivate this future work, and I think it succeeds in that.

Might your new "Proposed additional tests" subsection be better housed under Discussion rather than Results?

I may have missed this, but it doesn't look like you ran simulation benchmarks of your bootstrap-based test for checking whether the normalized covariances are equal. It would be useful to see in simulations how the true and false positive rates of that test vary with the usual suspects like sample size and noise strengths.

It looks like you estimated the uncertainty for eta_xz and eta_yz separately. Can you get the joint distribution? If you can do that, my intuition is you might be able to improve the power of the test (and maybe detect positive control #3?). For instance, if you can get your bootstraps for eta_xz and eta_yz together, could you just use a paired t-test to check for equality of means?

The proof is a lot better, and it's great that you nailed down the requirement on the decay of beta, but the proof is still confusing in some places:

On pg 29, it says "That is, dividing the right equation in Eq. 5.8 with alpha, we write the ..." but the next equation doesn't obviously have anything to do with Eq. 5.8, and instead (I think) it comes from Eq 5.5. This could be clarified.

Later on page 29, you write "We now evoke the requirement that the averages xt and yt are stationary", but then you just repeat Eq. 5.11 and set it to zero. Clearly you needed the limit condition to set Eq. 5.11 to zero, but it's not clear what you're using stationarity for. I mean, if you needed stationarity for 5.11 presumably you would have referenced it at that step.

It could be helpful for readers if you could spell out the practical implications of the theorem's assumptions (other than the no-causality requirement) by discussing examples of setups where it would or wouldn't hold.

Author response:

The following is the authors’ response to the previous reviews

We have made the following small adjustments and resubmit the manuscript to be published as a Version of Record with eLife.

Changes in main text of the manuscript:

We have moved the “Proposed additional tests” subsection to the Discussion section as suggested by the referee. 

We have added a link to a Github repository and a link to a Zenodo data repository at the beginning of the Materials and Methods section in the “Data and materials availability” subsection. The Github repository contains simulation code and data, and single-cell data analysis code. The Zenodo link contains our experimental data (we await your confirmation before we publish it officially on Zenodo).   

Changes in the supplemental information files

We have fixed the typo on page 29 of the SI in which Eq. (8) was referred to in a derivation. It should be Eq. (5) instead. We thank the referee for catching this mistake which has now been corrected.

We have fixed a typo on page 29 of SI, in which the word “evoke” is now “invoke”.  

We have clarified the derivation on page 29 of the SI. The referee is correct that the limit condition was used to set the right-hand side of Eq. (5.11) to zero.

eLife Assessment

This important study reports an advancement in the diagnosis of Animal African Trypanosomosis (AAT), which adapts a CRISPR-based diagnostic tool (SHERLOCK4AAT) to detect different trypanosome species responsible for AAT. The evidence supporting the conclusions is convincing and in line with the current state-of-the-art diagnostics. This study will be of interest to the fields of Epidemiology, Public Health, and Veterinary Medicine.

Reviewer #1 (Public review):

Summary:

The authors developed SHERLOCK4AAT, a CRISPR-Cas13a-based diagnostic toolbox for detecting multiple trypanosome species responsible for animal African trypanosomiasis. They created species-specific assays targeting six prevalent parasite species and validated the system using dried blood spots from domestic pigs in Guinea and Côte d'Ivoire. Field testing revealed high infection rates (62.7% of pigs infected) and, notably, the presence of human-infective parasites in domestic animals.

Major Strengths:

This study represents a valuable application of CRISPR-based detection technology to veterinary diagnostics, with strong potential for practical implementation. The authors conducted comprehensive validation, including statistical analyses to determine sensitivity and specificity, and demonstrated field utility through large-scale testing of 424 samples from two geographically distinct regions. The detection of human-infective parasites in pigs at both sites provides important One Health insights supporting integrated disease surveillance and has direct implications for public health policy and disease elimination programs. The methodology is robust, incorporating Bayesian statistical modeling and offering clear practical advantages such as dried blood spot compatibility and detection of active infections. The revised manuscript also addresses implementation considerations, including cost, training needs, and field logistics.

Major Weaknesses:

Some technical limitations constrain broader applicability. The assay for one key parasite species (T. vivax) shows suboptimal sensitivity, which may limit its utility in detecting this important pathogen. The current assay design does not distinguish between closely related species within the same subgenus-an important factor for certain epidemiological studies. Additionally, some assays relied on synthetic controls due to unavailable biological material, and the discussion on potential cross-reactivity with related kinetoplastid parasites is limited.<br /> Achievement of Aims: The authors clearly achieved their primary objectives of developing a sensitive, species-specific diagnostic system and demonstrating its applicability in real-world settings. The detection of human-infective trypanosomes in domestic pigs provides valuable epidemiological evidence in support of One Health strategies and targeted disease elimination efforts.

Impact and Utility:

This work responds to a well-documented need in veterinary diagnostics, where current methods often lack sensitivity or species discrimination. The system offers practical benefits for resource-limited settings through a short assay duration and compatibility with dried blood spot samples. While certain performance limitations may restrict broader adoption, the species identification capability represents a substantial advancement over existing approaches. The findings enhance our understanding of parasite diversity in livestock and their potential role as zoonotic reservoirs, with implications extending beyond veterinary medicine to public health surveillance and policy development.

Context:

This study makes a timely and relevant contribution to diagnostic epidemiology and One Health surveillance frameworks. The field-adapted use of advanced molecular detection technologies represents a significant step toward improved disease monitoring in regions where trypanosomiasis poses ongoing threats to animal health, agriculture, and human livelihoods. The cross-disciplinary implications for veterinary medicine, public health, and disease elimination programs underscore the broader significance of this work.

Reviewer #2 (Public review):

Summary:

The manuscript is fundamental due to the significance of its findings. The strength of the evidence is compelling, and the manuscript is publishable since the corrections have been made.

Strengths:

Using a Novel SHERLOCK4AAT toolkit for diagnosis.

Identification of various sub-species of Trypanosomes.

Differentiating the animal sub-species from the human one.

Corrections Made:

Definite articles have been removed from the title.

The words of the title have been reduced to 15.

Typographical errors have been corrected.

Weaknesses:

None

Reviewer #3 (Public review):

Summary:

The study adapts CRISPR-based detection toolkit (SHERLOCK assay) using conserved and species-specific targets for the detection of some members of the Trypanosomatidae family of veterinary importance and species-specific assays to differentiate between the six most common animal trypanosomes species responsible for AAT (SHERLOCK4AAT). The assays were able to discriminate between Trypanozoon (T. b. brucei, T. evansi and T. equiperdum), T. congolense (Savanah, Forest Kilifi and Dzanga sangha), T. vivax, T. theileri, T. simiae and T. suis. The design of both broad and species-specific assays was based primarily on sequences of the 18S rRNA, GAPDH (Glyceraldehyde-3-phosphate dehydrogenase) and invariant flagellum antigen (IFX) genes for species identification. Most importantly the authors showed varying limit of detection for the different SHERLOCK assays which is somewhat comparable to PCR-derived molecular techniques currently used for detecting animal trypanosomes even though some of these methodologies have used other primers that target genes such as ITS1 and 7SL sRNA.

The data presented in the study are particularly useful and of significant interest for diagnosis of AAT in affected areas.

Strengths:

The assays convincingly allow for the analysis and detection of most trypanosomes in AAT

Weaknesses:

Inability for the assay to distinguish T. b. brucei, T. evansi and T. equiperdum using the 18S rRNA gene as well as the IFX gene not achieving the sensitivity requirements for detection of T. vivax. Both T. brucei brucei and T. vivax are the most predominant infective species in animals (in addition to T. congolense), therefore a reliable assay should be able to convincingly detect these to allow for proper use of diagnostic assay.

Author response:

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

Reviewer #1 (Public review): 

Summary: 

This study addresses a critical gap in veterinary diagnostics by developing a CRISPR-based diagnostic toolbox (SHERLOCK4AAT) for detecting animal African trypanosomosis. It describes the development and field deployment of SHERLOCK4AAT, a CRISPR-Cas13-based diagnostic toolbox for the eco-epidemiological surveillance of animal African trypanosomosis (AAT) in West Africa.The authors successfully created and validated species-specific assays for multiple trypanosomes, including T. congolense, T. vivax, T. theileri, T. simiae, and T. suis, alongside pan-trypanosomatid and pan-Trypanozoon assays. The field validation in pigs from Guinea and Côte d'Ivoire revealed high trypanosome prevalence (62.7%), frequent co-infections, and importantly identified T. b. gambiense in one animal at each site, suggesting pigs may serve as potential reservoirs for this human-infective parasite.

A major strength of the study lies in its methodological innovation. By adapting SHERLOCK to target both conserved and species-discriminating sequences, the authors achieved high sensitivity and specificity in detecting Trypanosoma species. Their use of dried blood spots, validated thresholds through ROC analyses, and statistical robustness (e.g., Bayesian latent class modeling) provides a strong foundation for their conclusions.

The results are significant: over 60% of pigs tested positive for at least one trypanosome species, with co-infections observed frequently and T. b. gambiense detected in pigs at both sites. These findings have direct implications for the role of animal reservoirs in human disease transmission and underscore the value of pigs as sentinel hosts in gHAT elimination efforts.

The limitations are well acknowledged, particularly the suboptimal sensitivity of the T. vivax assay and the reliance on synthetic controls for T. suis and T. simiae. However, these limitations do not undermine the overall conclusions, and the paper provides a clear roadmap for further assay refinement and implementation.

This study offers a timely, impactful, and well-substantiated contribution to the field. The SHERLOCK4AAT toolbox holds promise for improving AAT diagnostics in resource-limited settings and advancing One Health surveillance frameworks.

Thank you

Strengths: 

(1) The adaptation of SHERLOCK technology for AAT represents a significant technical advancement, offering higher sensitivity than traditional parasitological methods and the ability to detect multiple species simultaneously.

(2) Rigorously performed with validation using appropriate controls, ROC curve analyses, and Bayesian latent class modelling, establishing clear analytical sensitivity and specificity for most assays.

(3) Testing 424 pig samples across two countries provides robust evidence of the tool's utility and reveals important epidemiological insights about trypanosome diversity and prevalence.

(4) The identification of T. b. gambiense in pigs at both sites has significant implications for HAT elimination strategies and highlights the need for integrated One Health approaches.

(5) The use of dried blood spots and RNA detection for active infections makes the approach practical for field surveillance in resource-limited settings.

Thank you

Weaknesses: 

(1) The manuscript would benefit from more detailed discussion of practical considerations such as cost, equipment requirements, and training needs for implementing SHERLOCK in endemic areas and rural settings which would improve applicability.

This is now adressed in the revised discussion (end of the first section).

(2) Limited discussion of pig selection criteria: More justification for choosing pigs as sentinel animals and discussion of potential limitations of this approach would strengthen the manuscript.

Yes, this is now more clearly explained in the revised discussion (beginning of the first section).

(3) More details on why certain genes were targeted would strengthen the methods.

The first result section ‘Selection of targets for broad and species-specific SHERLOCK assays targeting AAT species (SHERLOCK4AAT)’ is already dedicated to extensively explaining target selection, hence we’re afraid we don’t know what could be added.  

(4) Table formatting could be improved for readability. 

(5) Some figures are complex and would benefit from additional explanations in the legends.

We have tried to improve these two aspects as much as possible in the revised manuscript.

Reviewer #2 (Public review): 

Summary: 

The manuscript is important due to the significance of the findings. The strength of evidence is convincing.

Thank you

Strengths: 

(1) Using a Novel SHERLOCK4AAT toolkit for diagnosis. 

(2) Identification of various sub-species of Trypanosomes. 

(3) Differentiating the animal subspecies from the human one. 

Thank you

Weaknesses: 

(1) The title is too long, and the use of definite articles should be reduced in the title.

The title has been improved in the revised version.

(2) The route of blood sample collection in the animals should be well defined and explained.

This has been more clearly explained in the revised method section.

Reviewer #3 (Public review):

Summary: 

The study adapts CRISPR-based detection toolkit (SHERLOCK assay) using conserved and species-specific targets for the detection of some members of the Trypanosomatidae family of veterinary importance and species-specific assays to differentiate between the six most common animal trypanosome species responsible for AAT (SHERLOCK4AAT). The assays were able to discriminate between Trypanozoon (T. b. brucei, T. evansi, and T. equiperdum), T. congolense (Savanah, Forest Kilifi, and Dzanga sangha), T. vivax, T. theileri, T. simiae, and T. suis. The design of both broad and species-specific assays was based primarily on sequences of the 18S rRNA, GAPDH (Glyceraldehyde-3-phosphate dehydrogenase), and invariant flagellum antigen (IFX) genes for species identification. Most importantly, the authors showed varying limits of detection for the different SHERLOCK assays, which is somewhat comparable to PCR-derived molecular techniques currently used for detecting animal trypanosomes, even though some of these methodologies have used other primers that target genes such as ITS1 and 7SL sRNA. <br /> The data presented in the study are particularly useful and of significant interest for the diagnosis of AAT in affected areas.

Thank you

Strengths: 

The assays convincingly allow for the analysis and detection of most trypanosomes in AAT.

Thank you

Weaknesses: 

Inability for the assay to distinguish T. b. brucei, T. evansi, and T. equiperdum using the 18S rRNA gene, as well as the IFX gene, not achieving the sensitivity requirements for detection of T. vivax.  Both T. brucei brucei and T. vivax are the most predominant infective species in animals (in addition to T. congolense), therefore, a reliable assay should be able to convincingly detect these to allow for proper use of the diagnostic assay.

We agree with this point and aim to improve the toolbox for future studies.

Reviewer #1 (Recommendations for the authors):

(1) Provide additional details on the practicality of SHERLOCK deployment in the field, including training, costs, and infrastructure (potential challenges for field deployment, including suggestions for how to overcome these barriers).

This is now adressed in the revised discussion (end of the first section).

(2) Provide more detailed justification for choosing pigs as the main study species and discuss potential benefits and limitations of extending the approach to other livestock species.

Yes, this is now more clearly explained in the revised discussion (beginning of the first section).

(3) Add a comparison table comparing SHERLOCK4AAT performance metrics (sensitivity, specificity, LoD) with existing molecular diagnostic methods for AAT for ease of reference.

There are dozens of different serological, immunological and molecular approaches with highlty variable levels of sensitivity and specificities already reviewed and compared in detail in two references from 2022 (Desquesnes et al. a and b), which we have cited, as well as in a newly added reference (EBHODAGHE F acta trop 2018). Hence, we decided to only refer to the most comparable studies in the present article.

(4) Review complex figures and improve legends for better readability and interpretation.

We have tried to improve this as much as possible in the revised manuscript.

Reviewer #2 (Recommendations for the authors): 

(1) Reduce the number of words in the title from 28 to not more than 20.

The title has been improved in the revised version.

(2) Specify the particular route of collection of blood samples in the various animals.

Yes, this is now more clearly explained in the revised method section.

(3) Correct all typographical errors. 

We have tried to improve this as much as possible in the revised manuscript.

Thanks. I wish you the best in your publication process. 

Thank you

Reviewer #3 (Recommendations for the authors): 

Minor comments 

(1) The authors can expand the discussion to include other recent diagnostic assays for Animal trypanosomiasis, such as those that target other genes like tubulin.

Please see response to Review 1 point #3 above.

(2) The cost-effectiveness of the use of the assay can be discussed since the assay is expected to be used for work in some resource-deprived areas. For example, will it cost a researcher less to do a diagnosis with this assay relative to what is already available?

This is now adressed in the revised discussion (end of the first section).

(3) Is Cote d'Ivoire more endemic for AAT than Guinea? Will this account for the apparently consistent differences in the percentage of positive samples, or just because of the type of samples used from the two locations?

As the sampling method, sample preservation and sample analysis were the same for both groups - yes, it appears that pigs, at least for domesticated ones, in the study region of Cote d'Ivoire were more frequently infected than those in the study region of Guinea. It is however risky to extrapolate these observations to the AAT prevalence in the entire countries and/or to other mammals.

(4) Can the authors comment on how long one can store the samples for an effective and reliable assay?

The samples can be stored for several months at ambient temperature in a sealed bag with silica gel packages to reduce humidity. We have added this detail in the revised methods section.

(5) It is not clear whether the authors used conventional molecular diagnostics to compare the data obtained from this particular cohort of animals as reference is made to published data. It is not surprising that the SHERLOCK performed better than using parasitology-based methodology.

This is now adressed in the revised discussion.

(6) (Figure 4D-5D) should be 4D and 5D.

Thank you, this has been corrected.

eLife Assessment

This useful study integrates experimental methods from materials science with psychophysical methods to investigate how frictional stabilities influence tactile surface discrimination. The authors argue that force fluctuations arising from transitions between frictional sliding conditions facilitate the discrimination of surfaces with similar friction coefficients. However, the reliance on friction data from an artificial finger, combined with correlational analyses that fall short of establishing a mechanistic link to perception, renders the findings incomplete.

Reviewer #2 (Public review):

This is a revised version of a paper I reviewed previously.

Again, the purpose of the paper is to suggest that common metrics, such as friction or any given physical property of the surface, are probably inadequate to predict the perception of the surface or its discriminability. Instead, the authors propose a very interesting and original idea that, instead, frictional instabilities are related to fine touch perception (title).

Overall, the authors have put much effort into improving the manuscript, enhancing clarity, and avoiding overstatements. And I feel the narrative is indeed much improved and less ambiguous.

However, the authors have systematically avoided addressing the main comment of all reviewers: the link made between the mock finger passive experiment and the active human psychophysics is incorrect and should not be done, because its interpretation could be flawed.<br /> - First, this link is very weak (the correlation of 6 datapoints is barely significant).<br /> - Second, the real and mock fingers have very different properties (think about moisture, compliance, roughness,...).<br /> - Third, the comparison is made between a passive and well-controlled experiment and an active exploration. Yet, the comparison metrics (number of events) are clearly dependent on exploration procedures.

In your response to my comments:<br /> "We have made changes throughout the manuscript to acknowledge that our findings are correlative, clarifying this throughout, and incorporating into the discussion how our work may enable biomechanical measurements and tactile decision making models"

The authors admit that the analysis is flawed, yet they did not remove it. If they cannot demonstrate that the mock finger and the human finger behave the same way during the perceptual experiment, then they should remove Fig2 that combines apples and oranges. OR, they should look at the active exploration data and compute the same metrics on that data.

"This "weird choice" is the central innovation of this paper. This choice was necessary because we demonstrated that the common usage of friction coefficient is fundamentally flawed: we see that friction coefficient suggests that surface which are more different would feel more similar - indeed the most distinctive surfaces would be two surfaces that are identical, which is clearly spurious. "

They did not "demonstrate" such a flaw. Again, the difference in friction is between the mock finger trials. At the very least, the authors should verify that it is true of the active human experiment.

"To fully implement this, a decision-making model is necessary because, as a counter example, a participant could have generated 10 swipes of SFW and 1 swipe of a Sp, but the Sp may have been the most important event for making a tactile decision. This type of scenario is not compatible with the analysis suggested - and similar counterpoints can be made for other types of seemingly straightforward analysis."

The suggested analyses are straightforward and would be much more valuable than the data from the mock finger, even with the potential variability stated above.

"We recognize that, with all factors being equal, this sample size is on the smaller end"

Yet, the authors did not collect additional data to confirm their findings.

Reviewer #3 (Public review):

Strengths:

The paper describes a new perspective on friction perception, with the hypothesis that humans are sensitive to the instabilities of the surface rather than the coefficient of friction. The paper is very well written and with a comprehensive literature survey.

One of the central tools used by the author to characterize the frictional behavior is the frictional instabilities maps. With these maps, it becomes clear that two different surfaces can have both similar and different behavior depending on the normal force and the speed of exploration. It puts forward that friction is a complicated phenomenon, especially for soft

The psychophysics study is centered around an odd-one-out protocol, which has the advantage of avoiding any external reference to what would mean friction or texture for example. The comparisons are made only based on the texture being similar or not.

The results show a significant relationship between the distance between frictional maps and the success rate in discriminating two kinds of surface.

Weaknesses:

The main weakness of the paper comes from the fact that the frictional maps and the extensive psychophysics study are not made at the same time, nor with the same finger. The frictional maps are produced with an artificial finger made out of PDMS which is a poor substitute for the complex tribological properties of skin.

The evidence would have been much stronger if the measurement of the interaction was done during the psychophysical experiment. In addition, because of the protocol, the correlation is based on aggregates rather than on individual interactions. However the current data already bring new light on the nature of frictional oscillation and their link to perception.

The authors compensate with a third experiment where they used a 2AFC protocol and an online force measurement. But the results of this third study fail to solidify the relation.

No map of the real finger interaction is shown, bringing doubt to the validity of the frictional map for something as variable as human fingers.

Reviewer #4 (Public review):

Summary:

In this paper, Derkaloustian et al. look at the important topic of what affects fine touch perception. The observations that there may be some level of correlation with instabilities are intriguing. They attempted to characterize different materials by counting the frequency (occurrence #, not of vibration) of instabilities at various speeds and forces of a PDMS slab pulled lengthwise over the material. They then had humans perform the same vertical motion to discriminate between these samples. They correlated the % correct in discrimination with differences in frequency of steady sliding over the design space as well as other traditional parameters such as friction coefficient and roughness.

The authors pose an interesting hypothesis and make an interesting observation about the occurrences of instability regimes in different materials while in contact with PDMS, which is interesting for the community to see in publication. It should be noted however that the finger is complex, and there are many factors that may be over simplified, and perhaps even incorrect, with the use of the PDMS finger. There are trends, such as the trend of surfaces that are more similar in PDMS friction coefficient being easier to discriminate than those with more different PDMS friction coefficient, that contradict multiple other papers in the literature (Fehlberg et al., 2024; Smith and Scott, 1996). This may be due to the PDMS finger not being representative of the real finger conditions. A measurement of friction and the instabilities with a human finger, or demonstration that the PDMS finger is producing the same results (friction coefficient, instabilities, etc.) as a human finger, is needed.

Strengths:

The strength of this paper is in its intriguing hypothesis and important observation that instabilities may contribute to what humans are detecting as differences in these apparently similar samples.

Weaknesses:

There is are significant weaknesses in the representativeness of the PDMS finger, the vertical motion, and the speed of sliding to real human exploration. The real finger has multiple layers with different moduli. In fact, the stratum corneum cells, which are the outer layer at the interface and determine the friction, have much higher modulus than PDMS. In addition, the flat contact area can cause shifting of contact points. Both can contribute to making the PDMS finger have much more stick slip than a real finger. In fact, if you look at the regime maps, there is very little space that has steady sliding. This does not represent well human exploration of surfaces. We do not tend to use force and velocity that will cause extensive stick slip (frequent regions of 100% stick slip) and, in fact, the speeds used in the study are on the slow side, which also contributes to more stick slip. At higher speeds and lower forces, all of the materials had steady sliding regions. Further, on these very smooth surfaces, the friction and stiction are more complex and cannot dismiss considerations such as finger material property change with sweat pore occlusion and sweat capillary forces. Also, the vertical motion of both the PDMS finger and the instructed human subjects is not the motion that humans typically use to discriminate between surfaces.

This all leads to the critical question, why is the friction, normal force, and velocity not measured during the measured human exploration using the real human finger? An alternative would be showing that the PDMS finger reproduces the results of the human finger. I have checked the author's previous papers with this setup and did not find one that showed that the PDMS finger produced the same results as a human finger (Carpenter et al., 2018; Dhong et al., 2018; Nolin et al., 2022, 2021). The reviewer is not asking to do a more detailed psychophysical study with a decision-making model. All that is being asked is to use a human finger for the friction coefficient and instability measurements at typical human forces and speeds, or at least doing these measurements with both for one or two samples to show that the PDMS finger produces the same results as a human finger. The authors posed an extremely interesting hypothesis that humans may alter their speed to feel the instability transition regions. This is something that could be measured with a real finger but is not likely to be correlated accurately enough to match regime boundaries determined with such a simplified artificial finger.

References

Carpenter CW, Dhong C, Root NB, Rodriquez D, Abdo EE, Skelil K, Alkhadra MA, Ramírez J, Ramachandran VS, Lipomi DJ. 2018. Human ability to discriminate surface chemistry by touch. Mater Horiz 5:70-77. doi:10.1039/C7MH00800G<br /> Dhong C, Kayser LV, Arroyo R, Shin A, Finn M, Kleinschmidt AT, Lipomi DJ. 2018. Role of fingerprint-inspired relief structures in elastomeric slabs for detecting frictional differences arising from surface monolayers. Soft Matter 14:7483-7491. doi:10.1039/C8SM01233D<br /> Fehlberg M, Monfort E, Saikumar S, Drewing K, Bennewitz R. 2024. Perceptual Constancy in the Speed Dependence of Friction During Active Tactile Exploration. IEEE Transactions on Haptics 17:957-963. doi:10.1109/TOH.2024.3493421<br /> Nolin A, Licht A, Pierson K, Lo C-Y, Kayser LV, Dhong C. 2021. Predicting human touch sensitivity to single atom substitutions in surface monolayers for molecular control in tactile interfaces. Soft Matter 17:5050-5060. doi:10.1039/D1SM00451D<br /> Nolin A, Pierson K, Hlibok R, Lo C-Y, Kayser LV, Dhong C. 2022. Controlling fine touch sensations with polymer tacticity and crystallinity. Soft Matter 18:3928-3940. doi:10.1039/D2SM00264G<br /> Smith AM, Scott SH. 1996. Subjective scaling of smooth surface friction. Journal of Neurophysiology 75:1957-1962. doi:10.1152/jn.1996.75.5.1957

Author response:

The following is the authors’ response to the original reviews

eLife Assessment

This useful study integrates experimental methods from materials science with psychophysical methods to investigate how frictional stabilities influence tactile surface discrimination. The authors argue that force fluctuations arising from transitions between frictional sliding conditions facilitate the discrimination of surfaces with similar friction coefficients. However, the reliance on friction data obtained from an artificial finger, together with the ambiguous correlative analyses relating these measurements to human psychophysics, renders the findings incomplete.

Our main goal with this paper was to show that the most common metric, i.e. average friction coefficient—widely used in tactile perception and device design – is fundamentally unsound, and to offer a secondary parameter that is compatible with the fact that human motion is unconstrained, leading to dynamic interfacial mechanics.

We understand the Reviewers wanted, through biomechanical measurements, to demonstrate that humans using instabilities. This is seemingly reasonable, but in individual responses, we explain the significant challenges and fundamental unknowns to those experiments. We believe this paper sets forth an important step to approach this problem. At the same time, we have made several changes in the discussion, conclusion, and title to clarify that our study is correlative between mechanical characterization and human testing.

In short, there are still several fundamental unknowns that prevented us from basing the study around biomechanical measurements: (1) a decision-making model would need to be created, but it is unknown if tactile decision making follows other models, (2) it is further unknown what constitutes “tactile evidence”, though at our manuscript’s conclusion, we propose that friction instabilities are better suited for to be tactile evidence than the averaging of friction coefficients from a narrow range of human exploration (3) in the design of samples, from a friction mechanics and materials perspective, it is not at this point, possible to pre-program surfaces a priori to deliver friction instabilities and instead must be experimentally determined – especially when attempting to achieve this in controlled surfaces that do not create other overriding tactile cues, like macroscopic bumps or large differences in surface roughness. (4) Given that the basis for tactile percepts, like which object feels “rougher” or “smoother” is not sufficiently established, it is necessary to use a 3-alternative forced choice task which avoids asking objects along a preset perceptual dimension – a challenge recognized by Reviewer 3. However, this would bring in issues of memory in the decision-making model. (5) The prior points are compounded by the fact that, we believe, tactile exploration must be performed in an unconstrained manner, i.e., without an apparatus generating motion onto a stationary finger. Work by Liu et al. (IEEE ToH, 2024) showed that recreating friction obtained during free exploration onto a stationary finger was uninterpretable by the participants, hinting at the importance of efference copies.[1] We believe that many of the above-mentioned issues constitutes a significant advance in knowledge and would require discussion and dissemination with the community.

Our changes to the manuscript

Page 1 & SI Page 1, Title

“Alternatives to Friction Coefficient: Fine Touch Perception Correlates with Frictional Instabilities”

Reviewer 1 (Public review):

Summary:

In this paper, Derkaloustian et. al look at the important topic of what affects fine touch perception. The observations that there may be some level of correlation with instabilities are intriguing. They attempted to characterize different materials by counting the frequency (occurrence #, not of vibration) of instabilities at various speeds and forces of a PDMS slab pulled lengthwise over the material. They then had humans make the same vertical motion to discriminate between these samples. They correlated the % correct in discrimination with differences in frequency of steady sliding over the design space as well as other traditional parameters such as friction coefficient and roughness. The authors pose an interesting hypothesis and make an interesting observation about the occurrences of instability regimes in different materials while in contact with PDMS, which is interesting for the community to see in the publication. It should be noted that the finger is complex, however, and there are many factors that may be quite oversimplified with the use of the PDMS finger, and the consideration and discounting of other parameters are not fully discussed in the main text or SI. Most importantly, however, the conclusions as stated do not align with the primary summary of the data in Figure 2.

Strengths:

The strength of this paper is in its intriguing hypothesis and important observation that instabilities may contribute to what humans are detecting as differences in these apparently similar samples.

We thank Reviewer 1 for their time on the manuscript, recognizing the approach we took, and offering constructive feedback. We believe that our conclusions, in fact, are supported by the primary summary of the data in Fig. 2 but we believe that our use of R² could have led to misinterpretation. The trend with friction coefficient and percent correct was indeed statistically significant but was spurious because the slope was negative. In the revision, we add clarifying comments throughout, change from R² to r as to highlight the negative trend, and adjust the figures to better focus on friction coefficient.

Finally, we added a new section to discuss the tradeoffs between using a real human finger versus a mock finger, and which situations may warrant the use of one or the other. In short, for our goal of characterizing surfaces to be used in tactile experiments, we believe a mock finger is more sustainable and practical than using real humans because human fingers are unique per participant, humans move their fingers at constantly changing pressures and velocities, and friction generated during free exploring human cannot be satisfactorily replicated by moving a sample onto a stationary finger. But, we do not disagree that for other types of experiments, characterizing a human participant directly may be more advantageous.

Weaknesses:

Comment 1

The most important weakness is that the findings do not support the statements of findings made in the abstract. Of specific note in this regard is the primary correlation in Figure 2B between SS (steady sliding) and percent correct discrimination. Of specific note in this regard is the primary correlation in Figure 2B between SS (steady sliding) and percent correct discrimination. While the statistical test shows significance (and is interesting!), the R-squared value is 0.38, while the R-squared value for the "Friction Coefficient vs. Percent Correct" plot has an R-squared of 0.6 and a p-value of < 0.01 (including Figure 2B). This suggests that the results do not support the claim in the abstract: "We found that participant accuracy in tactile discrimination was most strongly correlated with formations of steady sliding, and response times were negatively correlated with stiction spikes. Conversely, traditional metrics like surface roughness or average friction coefficient did not predict tactile discriminability."

We disagree that the trend with friction coefficient suggests the results do not support the claim because the correlation was found to be negative. However, we could have made the comparison more apparent and expanded on this point, given its novelty.

While the R² value corresponding to the “Friction Coefficient vs. Percent Correct” plot is notably higher, our results show that the slope is negative, which would be statistically spurious. This is because a negative correlation between percent correct (accuracy in discriminating surfaces) and difference in friction coefficient means that the more similar two surfaces are (by friction coefficient), the easier it would be for people to tell them apart. That is, it incorrectly concludes that two identical surfaces would be much easier to tell apart than two surfaces with greatly different friction coefficients.

This is counterintuitive to nearly all existing results, but we believe our samples were well-positioned to uncover this trend by minimizing variability, by controlling multiple physical parameters in the samples, and that the friction coefficient — typically calculated in the field as an average friction coefficient — ignores all the dynamic changes in forces present in elastic systems undergoing mesoscale friction, i.e., human touch, as seen in Fig. 1 in a mock finger and Fig. 3 in a real finger. By demonstrating this statistically spurious trend, we believe this strongly supports our premise that an alternative to friction coefficient is needed in the design of tactile psychophysics and haptic interfaces.

We believe that this could have been misinterpreted, so we took several steps to improve clarity, given the importance of this finding: we separated the panel on friction coefficient to its own panel, we changed from R² to r throughout, and we added clarifying text. We also added a small section focusing on this spurious trend.

Our changes to the manuscript

Page 1, Abstract

“In fact, the typical method of averaging friction coefficients led to a spurious correlation which erroneously suggests that distinct objects should feel identical and identical objects should feel distinct.”

Page 7

“As Fig. 1 was constructed from friction measurements, we can also calculate an average friction coefficient, µ, by averaging the friction coefficient obtained at each of the 16 combinations of masses and velocities (Table 1). This calculation is a standard approach in tactile studies for summarizing friction measurements, or in some cases, surfaces are never characterized at multiple masses and velocities. However, summarizing friction data in this manner has been considered as conceptually questionable by others from a mechanics perspective.[3] Fig. 1 shows that the type of instabilities and friction forces encountered on a single surface can vary widely depending on the conditions. As a result, large variations in the friction coefficient are expected, depending on the mass and velocity — even though measurements originate from the same surface. This variability in friction coefficient can be seen with the large interquartile range of friction coefficients, which shows that the variation in friction coefficient across a single surface is similar, or even larger, than the differences in average friction coefficient across two different surfaces. The observation that friction coefficients vary so widely on a single surface calls into question the approach of analyzing how humans may perceive two different objects based on their average friction coefficients.”

Page 9, Fig. 2 Caption

“D) GLMM of accuracy vs. difference in average friction coefficient , showing a negative correlation. E) GLMMs of accuracy vs. other commonly used material properties or parameters: ΔAverage roughness R_a, ΔHurst exponent H, and ΔWater contact angle hysteresis (º) (N = 10 participants_, _n = 600 total trials).”

Page 9

“Considering all instabilities individually, we found that only steady sliding was a positive, statistically significant predictor. (r \= 0.62, p < 0.05, shown in Fig. 2B).”

Page 10

“To compare the value of looking at frictional instabilities, we also performed GLMM fits on common approaches in the field, like a friction coefficient or material property typically used in tactile discrimination, shown in Fig. 2D-E. Interestingly, in Fig. 2D, we observed a spurious, negative correlation between friction coefficient (typically and often problematically simplified as across all tested conditions) and accuracy (r = -0.64, p < 0.01); that is, the more different the surfaces are by friction coefficient, the less people can tell them apart. This spurious correlation would be the opposite of intuition, and further calls into question the common practice of using friction coefficients in touch-related studies. Interestingly, this spurious correlation was also found by Gueorguiev et al.[21] The alternative, two-term model which includes adhesive contact area for friction coefficient[32] was even less predictive (see Fig. S6A of SI). We believe such a correlation could not have been uncovered previously as our samples are minimal in their physical variations. Yet, the dynamic changes in force even within a single sample are not considered, despite being a key feature of mesoscale friction during human touch.

We investigate different material properties in Fig. 2E. Differences in average roughness R_a (or other parameters, like root mean square roughness R_rms (Fig. S6A of SI) did not show a statistically significant correlation to accuracy. Though roughness is a popular parameter, correlating any roughness parameter to human performance here could be moot: the limit of detecting roughness differences has previously been defined as 13 nm on structured surfaces[36] and much higher for randomly rough surfaces,[49] all of which are magnitudes larger than the roughness differences between our surfaces. The differences in contact angle hysteresis – as an approximation of the adhesion contributions[50] – do not present any statistically significant effects on performance.”

Page 11-12

“Despite the correlative nature of this study, we still obtained high correlations compared to existing biomechanical studies[4,19,21], which we speculate is because instabilities are an important predictive phenomenon for models of human touch. We believe that biomechanical studies, including more sophisticated techniques, like spatially resolved force maps from digital image correlation[5,42] may yield stronger correlations and results if they analyze data based on instabilities.

Added References

(2) Khamis, H. et al. Friction sensing mechanisms for perception and motor control: passive touch without sliding may not provide perceivable frictional information. J. Neurophysiol. 125, 809– 823 (2021).

(6) Olczak, D., Sukumar, V. & Pruszynski, J. A. Edge orientation perception during active touch. J. Neurophysiol. 120, 2423–2429 (2018).

Comment 2, Part 1

Along the same lines, other parameters that were considered such as the "Percent Correct vs. Difference in Sp" and "Percent Correct vs. Difference in SFW" were not plotted for consideration in the SI. It would be helpful to compare these results with the other three metrics in order to fully understand the relationships.

We have added these plots to the SI. We note that we had checked these relationships and discussed them briefly, but did not include the plot. The plots show that the type of instability was not as helpful as its presence or absence.

Our changes to the manuscript

Page 9

“Furthermore, a model accounting for slow frictional waves alone specifically shows a significant, negative effect on performance (p < 0.01, Fig. S5 of SI), suggesting that in these samples and task, the type of instability was not as important.”

“Fig. S5. GLMM fits of participant accuracy vs. the differences in instability incidence for individual instability types. Left: accuracy vs. differences in formation of slow frictional waves (SFW) between pairs. P1 and P5 have the same x-axis value and are shifted for clarity. Right: accuracy vs. differences in formation of stiction spikes (Sp).”

SI Page 4

“and no correlation between accuracy and stiction spikes (Fig. S5).”

Comment 2, Part 2

Other parameters such as stiction magnitude and differences in friction coefficient over the test space could also be important and interesting.

We agree these are interesting and have thought about them. We are aware that others, like Gueorguiev et al., have studied stiction magnitudes, and though there was a correlation, the physical differences in surface roughness (glass versus PMMA) investigated made it unclear if these could be generalized further.[3] We are unsure how to proceed here with a satisfactory analysis of stiction magnitude, given that stiction spikes are not always generated. In fact, Fig. 1 shows that for many velocities and pressures, stiction spikes are not formed. In ongoing work, however, we are always cognizant that if stiction spikes are a dominant factor, then a secondary analysis on their magnitude would be important. We offer some speculation on why stiction spikes may be overrepresented in the literature:

(1) They are prone to being created if the finger was loaded for a long time onto a surface prior to movement, thus creating adhesion by contact aging which is unlike active human exploration. We avoid this by discarding the first pull in our measurements, which is a standard practice in mechanical characterization if contact aging needs to be avoided.

(2) The ranges of velocities and pressures explored by others were small.

(3) In an effort to generate strong tactile stimuli, highly adhesive or rough surfaces are used.

(4) Stiction spikes are visually distinctive on a plot, but we are unaware of any mechanistic reason that mechanoreceptors would be particularly sensitive to this low frequency event over other signals.

We interpret “difference in friction coefficient over the test space” to be, for a single surface, like C4, to find the highest average friction for a condition of single velocity and mass and subtract that from the lowest average friction for a condition of single velocity and mass. We calculated the difference in friction coefficient in the typical manner of the field, by averaging all data collected at all velocities and masses and assigning a single value for all of a surface, like C4. We had performed this, and have the data, but we are wary of overinterpreting secondary and tertiary metrics because they do not have any fundamental basis in traditional tribology, and this value, if used by humans, would suggest that they rapidly explore a large parameter space to find a “maximum” and “minimum” friction. Furthermore, the range in friction across the test space, after averaging, can be smaller than the range of friction experienced at different masses and velocities on a single surface. We have tabulated and newly included these values (the interquartile range of friction coefficients of different masses and velocities per surface) in Table 1.

Fig. 2D shows a GLMM fit between percent correct responses across our pairs and the differences in friction coefficient for each pair, where we see a spurious negative correlation. As we had the data of all average friction coefficients for each condition for a given material, we also looked at the difference in maximum and minimum friction coefficients. For our tested pairs, these differences also lined up on a statistically significant, negative GLMM fit (r = -0.86, p < 0.005). However, the values for a given surface can vary drastically, with an interquartile range of 1.20 to 2.09 on a single surface. We fit participant accuracy to the differences in these IQRs across pairs. This also led to a negative GLMM fit (r = -0.65, p < 0.05). However, we are hesitant to add this plot to the manuscript for the reasons stated previously.

Comment 3, Part 1

Beyond this fundamental concern, there is a weakness in the representativeness of the PDMS finger, the vertical motion, and the speed of sliding to real human exploration.

Overall, this is a continuous debate that we think offers two solutions, and we are not advocating for an “either-or” case. There is always a tradeoff between using a synthetic model of a finger versus a real human finger, and there is a place for both models. That is, while our mock finger will be “better” the more similar it is to a human finger, it is not our goal to fully replace a human finger. Rather our goal is to provide a consistent method of characterizing surfaces that is sufficiently similar to human touch as to be a useful and predictive tool.

The usefulness of the mock finger is in isolating the features of each surface that is independent of human variability, i.e., instabilities that form without changing loading conditions between sliding motions or even within one sliding motion. Of course, with this method, we still require confirmation of these features still forming during human exploration, which we show in Fig. 3. We believe that this method of characterizing surfaces at the mesoscale will ultimately lead to more successful human studies on tactile perception. Currently, and as shown in the paper, characterizing surfaces through traditional techniques, such as a commercial tribometer (friction coefficient, using a steel or hard metal ball), roughness (via atomic force microscopy or some other metrology), surface energy are less or not at all predictive. Thus, we believe this mock finger is better than the current state-of-the-art characterizing surfaces (we are also aware of a commercial mock finger company, but we were unable to purchase or obtain an evaluation model).

One of the main – and severe – limitations of using a human finger is that all fingers are different, meaning any study focusing on a particular user may not apply to others or be recreated easily by other researchers. We do not think it is feasible to set a standard for replication around a real human finger as that participant may no longer be available, or willing to travel the world as a “standard”. Furthermore, the method in which a person changes their pressures and velocities is different. We note that this is a challenge unique to touch perception – how an object is touched changes the friction generated, and thus the tactile stimulus generated, whereas a standardized stimulus is more straightforward for light or sound.

However, we do emphasize that we have strongly considered the balance between feasibility and ecological validity in the design of a mock finger. We have a mock finger, with the three components of stiffness of a human finger (more below). Furthermore, we have also successfully used this mock finger in correlations with human psychophysics in previous work, where findings from our mechanical experiments were more predictive of human performance[4–7] than other available methods.

Our changes to the manuscript Added (Page 2-3)

“Mock finger as a characterization tool

We use a mechanical setup with a PDMS (poly(dimethylsiloxane)) mock finger to derive tactile predictors as opposed to direct biomechanical measurements on human participants. While there is a tradeoff in selecting a synthetic finger over a real human finger to modeling human touch, human fingers themselves are also highly variable[23] both in their physical shape and their use during human motion. Our goal is to design a consistent method of characterization of samples that can be easily accessed by other researchers and does not rely on a standard established around single human participant. We believe that sufficient replication of surface, bulk properties, and contact geometry results in characterization that isolates consistent features of surfaces that are not derived from human-to-human variability. We have used this approach to successfully correlate human results with mock finger characterization previously.[8,9,24]

The major component of a human finger, by volume, is soft tissue (~56%),[25] resulting in an effective modulus close to 100 kPa.[26,27] In order to achieve this same softness, we crosslink PDMS in a 1×1×5 cm mold at a 30:1 elastomer:crosslinker ratio. In addition, two more features in the human finger impart significant mechanical differences. Human fingers have a bone at the fingertip, the distal phalanx,[26–28, 8–10]which we mimic with an acrylic “bone” within our PDMS network. The stratum corneum, the stiffer, glassier outer layer of skin,[29] is replicated with the surface of the mock finger glassified, or further crosslinked, after 8 hours of UV-Ozone treatment.30 This treatment also modifies the surface properties of the native PDMS to align with those of a human finger more closely: it minimizes the viscoelastic tack at the surface, resulting in a comparable non-sticky surface. Stabilizing after one day after treatment, the mock finger surface obtains a moderate hydrophilicity (~60º), as is typically observed for a real finger.[11,31]

The initial contact area formed before a friction trace is collected is a rectangle of 1×1 cm. While this shape is not entirely representative of a human finger with curves and ridges, human fingers flatten out enough to reduce the effects of curvature with even very light pressures.[31–33] This implies that for most realistic finger pressures, the contact area is largely load-independent, which is more accurately replicated with a rectangular mock finger.

Lastly, we consider the role of fingerprint ridges. A key finding of our previous work is that while fingerprints enhanced frictional dynamics at certain conditions, key features were still maintained with a flat finger.[11] Furthermore, for some loading conditions, the more amplified signals could also result in more similar friction traces for different surfaces. We have observed good agreement between these friction traces and human experiments.[8,9,22,34]”

Page 3-4, Materials and Methods

“Mock Finger Preparation

Friction forces across all six surfaces were measured using a custom apparatus with a polydimethylsiloxane (PDMS, Dow Sylgard 184) mock finger that mimics a human finger’s mechanical properties and contact mechanics while exploring a surface relatively closely.[8,9] PDMS and crosslinker were combined in a 30:1 ratio to achieve a stiffness of 100 kPa comparable to a real finger, then degassed in a vacuum desiccator for 30 minutes. We are aware that the manufacturer recommended crosslinking ratio for Sylgard 184 is 10:1 due to potential uncrosslinked liquid residues,[35] but further crosslinking concentrated at the surface prevents this. The prepared PDMS was then poured into a 1×1×5 cm mold also containing an acrylic 3D-printed “bone” to attach applied masses on top of the “fingertip” area contacting a surface during friction testing. After crosslinking in the mold at 60ºC for 1 hour, the finger was treated with UV-Ozone for 8 hours out of the mold to minimize viscoelastic tack.

Mechanical Testing

A custom device using our PDMS mock finger was used to collect macroscopic friction force traces replicating human exploration.[8,9] After placing a sample surface on a stage, the finger was lowered at a slight angle such that an initial 1×1 cm rectangle of “fingertip” contact area could be established. We considered a broad range of applied masses (M \= 0, 25, 75, and 100 g) added onto the deadweight of the finger (6 g) observed during a tactile discrimination task. The other side of the sensor was connected to a motorized stage (V-508 PIMag Precision Linear Stage, Physikinstrumente) to control both displacement (4 mm across all conditions) and sliding velocity (v \= 5, 10, 25, and 45 mm s^-1). Forces were measured at all 16 combinations of mass and velocity via a 250 g Futek force sensor (k \= 13.9 kN m^-1) threaded to the bone, and recorded at an average sampling rate of 550 Hz with a Keithley 7510 DMM digitized multimeter. Force traces were collected in sets of 4 slides, discarding the first due to contact aging. Because some mass-velocity combinations were near the boundaries of instability phase transitions, not all force traces at these given conditions exhibited similar profiles. Thus, three sets were collected on fresh spots for each condition to observe enough occurrences of multiple instabilities, at a total of nine traces per combination for each surface.”

Added References

(23) Infante, V. H. P. et al. The role of skin hydration, skin deformability, and age in tactile friction and perception of materials. Sci. Rep. 15, 9935 (2025).

(24) Nolin, A., Lo, C.-Y., Kayser, L. V. & Dhong, C. B. Transparent and Electrically Switchable Thin Film Tactile Actuators Based on Molecular Orientation. Preprint at https://doi.org/10.48550/arXiv.2411.07968 (2024).

(25) Murai, M., Lau, H.-K., Pereira, B. P. & Pho, R. W. H. A cadaver study on volume and surface area of the fingertip. J. Hand Surg. 22, 935–941 (1997).

(26) Abdouni, A. et al. Biophysical properties of the human finger for touch comprehension: influences of ageing and gender. R. Soc. Open Sci. (2017) doi:10.1098/rsos.170321.

(27) Cornuault, P.-H., Carpentier, L., Bueno, M.-A., Cote, J.-M. & Monteil, G. Influence of physico-chemical, mechanical and morphological fingerpad properties on the frictional distinction of sticky/slippery surfaces. J. R. Soc. Interface (2015) doi:10.1098/rsif.2015.0495.

(28) Qian, K. et al. Mechanical properties vary for different regions of the finger extensor apparatus. J. Biomech. 47, 3094–3099 (2014).

(29) Yuan, Y. & Verma, R. Measuring microelastic properties of stratum corneum. Colloids Surf. B Biointerfaces 48, 6–12 (2006).

(30) Fu, Y.-J. et al. Effect of UV-Ozone Treatment on Poly(dimethylsiloxane) Membranes: Surface Characterization and Gas Separation Performance. Langmuir 26, 4392–4399 (2010).

Comment 3, Part 2

The real finger has multiple layers with different moduli. In fact, the stratum corneum cells, which are the outer layer at the interface and determine the friction, have a much higher modulus than PDMS. The real finger has multiple layers with different moduli. In fact, the stratum corneum cells, which are the outer layer at the interface and determine the friction, have a much higher modulus than PDMS.

We have approximated the softness of the finger with 100 kPa crosslinked PDMS, which is close to what has been reported for the bulk of a human fingertip.[9,10] However, as mentioned in the Materials and Methods, there are two additional features of the mock finger that impart greater strength. The PDMS surrounds a rigid, acrylic bone comparable to the distal phalanx, which provides an additional layer of higher modulus.[8] Additionally, the 8-hour UV-Ozone treatment decreases the viscoelastic tack of the pristine PDMS by glassifying, or further crosslinking the surface of the finger,[12] therefore imparting greater stiffness at the surface similar to the contributions of the stratum corneum, along with a similar surface energy.[13] This technique is widely used in wearables,[14] soft robotics,[15] and microfluidics[16] to induce both these material changes. Additionally, the finger is used at least a day after UV-Ozone treatment is completed to generate a stable surface that is moderately hydrophilic, similar to the outermost layer of human skin.[17]

Comment 3, Part 3

In addition, the slanted position of the finger can cause non-uniform pressures across the finger. Both can contribute to making the PDMS finger have much more stick-slip than a real finger.

To ensure that there is minimal contribution from the slanted position of the finger, an initial contact area of 1×1 cm is established before sliding and recording friction measurements. As the PDMS finger is a soft object, the portion in contact with a surface flattens and the contact area remains largely unchanged during sliding. Any additional stick-slip after this alignment step is caused by contact aging at the interface, but the first trace we collect is always discarded to only consider stick-slip events caused by surface chemistry. We recognize that it is difficult to completely control the pressure distribution due to the planar interface, but this is also expected when humans freely explore a surface.

Comment 3, Part 4

In fact, if you look at the regime maps, there is very little space that has steady sliding. This does not represent well human exploration of surfaces. We do not tend to use a force and velocity that will cause extensive stick-slip (frequent regions of 100% stick-slip) and, in fact, the speeds used in the study are on the slow side, which also contributes to more stick-slip. At higher speeds and lower forces, all of the materials had steady sliding regions.”

We are not aware of published studies that extensively show that humans avoid stickslip regimes. In fact, we are aware familiar with literature where stiction spike formation is suppressed – a recent paper by AliAbbasi, Basdogan et. al. investigates electroadhesion and friction with NaCl solution-infused interfaces, resulting in significantly steadier forces.[18] We also directly showed evidence of instability formation that we observed during human exploration in Fig. 3B-C. These dynamic events are common, despite the lack of control of normal forces and sliding velocities. We also note that Reviewer 1, Comment 2, Part 2 was suggesting that we further explore possible trends from parameterizing the stiction spike.

We note that many studies have often not gone at the velocities and masses required for stiction spikes – even though these masses and velocities would be routinely seen in free exploration – this is usually due to constraints of their equipment.[19] Sliding events during human free exploration of surfaces can exceed 100 mm/s for rapid touches. However, for the surfaces investigated here, we observe that large regions of stick-slip can emerge at velocities as low as 5 mm/s depending on the applied load. The incidence of steady sliding appears more dependent on the applied mass, with almost no steady sliding observed at or above 75 g. Indeed, the force categorization along our transition zones is the main point of the paper.

Comment 3, Part 5

Further, on these very smooth surfaces, the friction and stiction are more complex and cannot dismiss considerations such as finger material property change with sweat pore occlusion and sweat capillary forces. Also, the vertical motion of both the PDMS finger and the instructed human subjects is not the motion that humans typically use to discriminate between surfaces.

We did not describe the task sufficiently. Humans were only given the instruction to slide their finger along a single axis from top to bottom of a sample, not vertical as in azimuthal to gravity. We have updated our wording in the manuscript to reflect this.

Page 4

“Participants could touch for as long as they wanted, but were asked to only use their dominant index fingers along a single axis to better mimic the conditions for instability formation during mechanical testing with the mock finger.”

Page 11

“The participant was then asked to explore each sample simultaneously, and ran over each surface in strokes along a single axis until the participant could decide which of the two had “more friction”.”

Comment 3, Part 6

Finally, fingerprints may not affect the shape and size of the contact area, but they certainly do affect the dynamic response and detection of vibrations.”

We are aware of the nuance. Our previous work on the role of fingerprints on friction experienced by a PDMS mock finger showed enhanced signals with the incorporation of ridges on the finger and used a rate-and-state model of a heterogenous, elastic body to find corresponding trends (though there is no existing model of friction that can accurately model experiments on mesoscale friction).[11] The key conclusion was that a flat finger still preserved key dynamic features, and the presence of stronger or more vibrations could result in more similar forces for different surfaces depending on the sliding conditions.

This is also in the context that we are seeking to provide a reasonable and experimentally accessible method to characterize surfaces, which will always be better as we get closer in replicating a true human finger. But our goal here was to replicate the finger sufficiently for use in human studies. We believe the more appropriate metric of success is if the mock finger is more successful than replacing traditional characterization experiments, like friction coefficient, roughness, surface energy, etc.

Comment 4

This all leads to the critical question, why are friction, normal force, and velocity not measured during the measured human exploration and in a systematic study using the real human finger? The authors posed an extremely interesting hypothesis that humans may alter their speed to feel the instability transition regions. This is something that could be measured with a real finger but is not likely to be correlated accurately enough to match regime boundaries with such a simplified artificial finger.

We are excited that our manuscript offers a tractable manner to test the hypothesis that tactile decision-making models use friction instabilities as evidence. However, we lay out the challenges and barriers, and how the scope of this paper will lead us in that direction. We also clarify that our goals are to provide a method to characterize samples to better design tactile interfaces in haptics or in psychophysical experiments and raise awareness that the common methods of sample characterization in touch by an average friction coefficient or roughness is fundamentally unsound. Throughout the paper, we have made changes to reflect that our study, at this point, is only correlative.

As discussed in the summary, and with additional detail here, to further support our findings through observation on humans would require answering:

(1) Which one, or combination of, of the multiple swipes that people make responsible for a tactile decision? (There is a need for a decision-making model)

(2) Establish what is, or may be, tactile evidence.

(3) Establish tactile decision-making models are similar or different than existing decision-making models.

(4) Design a task that does not require the use of subjective tactile descriptors, like “which one feels rougher”, which we have seen causes confusion in participants, which will likely require accounting for memory effects.

We elaborate these points below:

To successfully perform this experiment, we note that freely exploring humans make multiple strokes on a surface. Therefore, we would need to construct a decision-making model. It has not yet been demonstrated whether tactile decision making follows visual decision making, but perhaps to start, we can assume it does. Then, in the design of our decision-making paradigm, we immediately run into the problem: What is tactile evidence?

From Fig. 3C, we already can see that identifying evidence is challenging. Prior to this manuscript, people may have chosen the average force, or the highest force. Or we may choose the average friction force. Then, after deciding on the evidence, we need to find a method to manipulate the evidence, i.e., create samples or a machine that causes high friction, etc. We show that during the course of human touch, due to the dynamic nature of friction, the average can change a large amount and sample design becomes a central barrier to experiments. Others may suggest immobilizing the finger and applying a known force, but given how much friction changes with human exploration, there is no known method to make a machine recreate temporally and spatially varying friction forces during sliding onto a stationary finger. Finally, perhaps most importantly, in addition to mechanical challenges, a study by Liu, Colgate et al. showed that even if they recorded the friction (2D) of a finger exploring a surface and then replicated the same friction forces onto a finger, the participant could not determine which surface the replayed friction force was supposed to represent.[1] This supports that the efference copy is important, that the forces in response to expected motion are important to determine friction. Finally, there is no known method to design instabilities a priori. They must be found through experiments. Especially since if we were to introduce, say a bump or a trough, then we bring in confounding variables to how participants tell surfaces apart.

Furthermore, even if we had some consistent method to create tactile “evidence”, the paradigm also deserves some consideration. In our experience, the 3-AFC task we perform is important because the vocabulary for touch has not been established. That is, in 3-AFC, by asking to determine which one sample is unlike the others, we do not have to ask the participant questions like “which one is rougher” or “which one has less friction”. In contrast, 2-AFC, which is better for decision-making models because it does not include memory, requires the asking of a perceptual question like: “which one is rougher?”. In our ongoing work, taking two silane coatings, we found that participants could easily identify which surface is unlike the others above chance in a 3-AFC, but participants, even within their own trials, could not consistently identify one silane as perceptually “rougher” by 2-AFC. To us, this calls into question the validity of tactile descriptors, but is beyond the scope of this manuscript.

This is not our only goal, but in the context of human exploration, in this manuscript here, we believed it was important to identify a mechanical parameter that was consistent with how humans explore surfaces, but was also a parameter that could characterize to some consistent property of a surface – irrespective of whether a human was touching it. We thought that designing human decision-making models and paradigms around the friction coefficient would not be successful.

Given the scope of these challenges, we do not think it would be possible to establish these conceptual sequences in a single manuscript. However, we think that our manuscript brings an important step forward to approach this problem.

Reviewer 2 (Public review):

Summary:

In this paper, the authors want to test the hypothesis that frictional instabilities rather than friction are the main drivers for discriminating flat surfaces of different sub-nanometric roughness profiles.

They first produced flat surfaces with 6 different coatings giving them unique and various properties in terms of roughness (picometer scale), contact angles (from hydrophilic to hydrophobic), friction coefficient (as measured against a mock finger), and Hurst exponent.

Then, they used those surfaces in two different experiments. In the first experiment, they used a mock finger (PDMS of 100kPA molded into a fingertip shape) and slid it over the surfaces at different normal forces and speeds. They categorized the sliding behavior as steady sliding, sticking spikes, and slow frictional waves by visual inspection, and show that the surfaces have different behaviors depending on normal force and speed. In a second experiment, participants (10) were asked to discriminate pairs of those surfaces. It is found that each of those pairs could be reliably discriminated by most participants.

Finally, the participant's discrimination performance is correlated with differences in the physical attributes observed against the mock finger. The authors found a positive correlation between participants' performances and differences in the count of steady sliding against the mock finger and a negative correlation between participants' reaction time and differences in the count of stiction spikes against the mock finger. They interpret those correlations as evidence that participants use those differences to discriminate the surfaces.

Strengths:

The created surfaces are very interesting as they are flat at the nanometer scale, yet have different physical attributes and can be reliably discriminated.

We thank Reviewer 2 for their notes on our manuscript. The responses below address the reviewer’s comments and recommendations for revised work.

Weaknesses:

Comment 1

In my opinion, the data presented in the paper do not support the conclusions. The conclusions are based on a correlation between results obtained on the mock finger and results obtained with human participants but there is no evidence that the human participants' fingertips will behave similarly to the mock finger during the experiment. Figure 3 gives a hint that the 3 sliding behaviors can be observed in a real finger, but does not prove that the human finger will behave as the mock finger, i.e., there is no evidence that the phase maps in Figure 1C are similar for human fingers and across different people that can have very different stiffness and moisture levels.

We have made changes throughout the manuscript to acknowledge that our findings are correlative, clarifying this throughout, and incorporating into the discussion how our work may enable biomechanical measurements and tactile decision making models.

The mechanical characterization conducted with the mock finger seeks to extract significant features of friction traces of a set of surfaces to use as predictors of tactile discriminability. The goal is to find a consistent method to characterize surfaces for use in tactile experiments that can be replicated by others and used prior to any human experiments. However, in the overall response and in a response to a similar comment by Reviewer 1 (recreated below), we also explain why we believe experiments on humans to establish this fact is not yet reasonable.

First, we discuss the mock finger. The PDMS finger is treated to have comparable surface and bulk properties to a human finger. We have approximated the softness of the finger with 100 kPa crosslinked PDMS, which is close to what has been reported for the bulk of a human fingertip.[9,10] However, as mentioned in the Materials and Methods, there are two additional features of the mock finger that impart greater strength. The PDMS surrounds a rigid, acrylic bone comparable to the distal phalanx, which provides an additional layer of higher modulus.[8] Additionally, the 8-hour UV-Ozone treatment decreases the viscoelastic tack of the pristine PDMS by glassifying, or further crosslinking the surface of the finger,[12] therefore imparting greater stiffness at the surface similar to the contributions of the stratum corneum, along with a similar surface energy.[13] Additionally, the finger is used at least a day after UV-Ozone treatment is completed in order for the surface to return to moderate hydrophilicity, similar to the outermost layer of human skin.[17] We also discuss the shape of the contact formed. To ensure that there is minimal contribution from the slanted position of the finger, an initial contact area of 1×1 cm is established before sliding and recording friction measurements. As the PDMS finger is a soft object, the portion in contact with a surface flattens and the contact area remains largely unchanged during sliding. Any additional stick-slip after this alignment step is caused by contact aging at the interface, but the first trace we collect is always discarded to only consider stick-slip events caused by surface chemistry. We recognize that it is difficult to completely control the pressure distribution due to the planar interface, but this is also expected when humans freely explore a surface. Finally, we consider flat vs. fingerprinted fingers. Our previous work on the role of fingerprints on friction experienced by a PDMS mock finger showed enhanced signals with the incorporation of ridges on the finger and used a rate-and-state model of a heterogenous, elastic body to find corresponding trends.[11] The key conclusion was that a flat finger still preserved key dynamic features, and the presence of stronger or more vibrations could result in more similar forces for different surfaces depending on the sliding conditions. We note that we have subsequently used this flat mock finger in correlations with human psychophysics in previous work, where findings from our mechanical experiments were predictive of human performance.[4–7] We have added these details to the manuscript.

With this adequately similar mock finger, we collected friction traces at controlled conditions of normal force and velocity in order to extract the signals unique to each material that are not caused by the influence of human variability. For example, we observe the smallest regions of steady sliding on our phase maps (Fig. 1C) for short-chain alkylsilanes C4 and C5, while the increased intermolecular forces of other silanes increase the incidence of steady sliding. We have also previously shown that comparisons of similarly collected mechanical data is predictive of human performance, using the crosscorrelations between signals of two different materials.[4–7] While different participants produce different raw signals, we see that broad categories of stick-slip, i.e. instabilities, can be extracted (Fig. 3B-C) and used as a cue in a tactile discrimination task. As mentioned above, we have provided an additional section about the usefulness of our mock finger, as well as its structure, in the main manuscript.

Second, we lay out the challenges and barriers to demonstrating this in humans in the manner requested by the reviewer, and how the scope of this paper will lead us in that direction. We also clarify that our goals are to provide a method to characterize samples to better design tactile interfaces in haptics or in psychophysical experiments and raise awareness that the common methods of sample characterization in touch by an average friction coefficient or roughness is fundamentally unsound.

As discussed in the summary, and with additional detail here, to further support our findings through observation on humans would require answering:

(1) Which one, or combination of, of the multiple swipes that people make responsible for a tactile decision?

(2) Establish what is, or may be, tactile evidence.

(3) Establish tactile decision-making models are similar or different than existing decision-making models.

(4) Test the hypothesis, in these models, that friction instabilities are evidence, and not some other unknown metric.

(5) Design a task that does not require the use of subjective tactile descriptors, like “which one feels rougher”, which we see cause confusion in participants, which will likely require accounting for memory effects.

We elaborate these points below:

To successfully perform this experiment, we note that freely exploring humans make multiple strokes on a surface. Therefore, we would need to construct a decision-making model. It has not yet been demonstrated whether tactile decision making follows visual decision making, but perhaps to start, we can assume it does. Then, in the design of our decision-making paradigm, we immediately run into the problem: What is tactile evidence?

From Fig. 3C, we already can see that identifying evidence is challenging. Prior to this manuscript, people may have chosen the average force, or the highest force. Or we may choose the average friction force. Then, after deciding on the evidence, we need to find a method to manipulate the evidence, i.e., create samples or a machine that causes high friction, etc. We show that during the course of human touch, due to the dynamic nature of friction, the average can change a large amount and sample design becomes a central barrier to experiments. Others may suggest immobilizing the finger and applying a known force, but given how much friction changes with human exploration, there is no known method to make a machine recreate temporally and spatially varying friction forces during sliding onto a stationary finger. Finally, perhaps most importantly, in addition to mechanical challenges, a study by Liu, Colgate, et al. showed that even if they recorded the friction (2D) of a finger exploring a surface and then replicated the same friction forces onto a finger, the participant could not determine which surface the replayed friction force was supposed to represent.[1] This supports that the efference copy is important, that the forces in response to expected motion are important to determine friction. Finally, there is no known method to design instabilities a priori. They must be found through experiments, especially since if we were to introduce, say a bump or a trough, then we bring in confounding variables to how participants tell surfaces apart.

Furthermore, even if we had some consistent method to create tactile “evidence”, the paradigm also deserves some consideration. In our experience, the 3-AFC task we perform is important because the vocabulary for touch has not been established. That is, in 3-AFC, by asking to determine which one sample is unlike the others, we do not have to ask the participant questions like “which one is rougher” or “which one has less friction”. In contrast, 2-AFC, which is better for decision-making models because it does not include memory, requires the asking of a perceptual question like: “which one is rougher?”. In our ongoing work, taking two silane coatings, we found that participants could easily identify which surface is unlike the others above chance in a 3-AFC, but participants, even within their own trials, could not consistently identify one silane as perceptually “rougher” by 2-AFC. To us, this calls into question the validity of tactile descriptors, but is beyond the scope of the current manuscript.

This is not our only goal, but in the context of human exploration, in this manuscript here, we believed it was important to identify a mechanical parameter that was consistent with how humans explore surfaces, but was also a parameter that could characterize to some consistent property of a surface – irrespective of whether a human was touching it. We thought that designing human decision-making models and paradigms around the friction coefficient would not be successful.

Given the scope of these challenges, we do not think it would be possible to establish this conceptual sequence in a single manuscript.

See Reviewer 1, comment 3part 3 for changes to the manuscript

Comment 2

I believe that the authors collected the contact forces during the psychophysics experiments, so this shortcoming could be solved if the authors use the actual data, and show that the participant responses can be better predicted by the occurrence of frictional instabilities than by the usual metrics on a trial by trial basis, or at least on a subject by subject basis. I.e. Poor performers should show fewer signs of differences in the sliding behaviors than good performers.

To fully implement this, a decision-making model is necessary because, as a counter example, a participant could have generated 10 swipes of SFW and 1 swipe of a Sp, but the Sp may have been the most important event for making a tactile decision. This type of scenario is not compatible with the analysis suggested — and similar counterpoints can be made for other types of seemingly straightforward analysis.

While we are interested and actively working on this, the study here is critical to establish types of evidence for a future decision-making model. We know humans change their friction constantly during real exploration, so it is unclear which of these constantly changing values we should input into the decision making model, and the future challenges we anticipate are explained in Weaknesses, Comment 1.

Comment 3

The sample size (10) is very small.

We recognize that, with all factors being equal, this sample size is on the smaller end. However, we emphasize the degree of control of samples is far above typical, with minimal variations in sample properties such as surface roughness, and every sample for every trial was pristine. Furthermore, the sample preparation (> 300 individual wafers were used) became a factor. Although not typically appropriate, and thus not included in the manuscript, a post-hoc power analysis for our 100 trials of our pair that was closest to chance, P4, (53%, closest to chance at 33%) showed a power of 98.2%, suggesting that the study was appropriately powered.

Reviewer 2 (Recommendations for the authors):

Comment 1

Differences in SS and Sp (Table 2) are NOT physical or mechanical differences but are obtained by counting differences in the number of occurrences of each sliding behavior. It is rather a weird choice.

We disagree that differences in SS and Sp are not physical or mechanical, as these are well-established phenomena in the soft matter and tribology literature.[20–22] These are known as “mechanical instabilities” and generated due to the effects of two physical phenomena: the elasticity of the finger (which is constant in our mechanical testing) and the friction forces present (which change per sample type). The motivation behind using these different shapes is that the instabilities, in some conditions, can be invariant to external factors like velocity. This would be quite advantageous for human exploration because, unlike friction coefficient, which changes with nearly any factor, including velocity and mass, the instabilities being invariant to velocity would mean that we are accurately characterizing a unique identifier of the surface even though velocity may be variable.

This “weird choice” is the central innovation of this paper. This choice was necessary because we demonstrated that the common usage of friction coefficient is fundamentally flawed: we see that friction coefficient suggests that surface which are more different would feel more similar – indeed the most distinctive surfaces would be two surfaces that are identical, which is clearly spurious. Furthermore, Table 1 now includes the range of friction generated on a surface, the range of friction coefficients of a single surface is large – of order the differences in friction between two surfaces. This is expected in soft sliding systems and emphasizes our issue with the use of average friction coefficient in psychophysical design. One potential explanation for why we were able to see this is effect is because our surfaces have similar (< 0.6 nm variability) roughness, removing potential confounding factors from large scale roughness, and this type of low roughness control has not been widely used in tactile studies to the best of our knowledge.

Comment 2

Figures 2B-C: why are the x-data different than Table 2?

The x-data in Fig. 2B-C are the absolute differences in the number of occurrences measured for a given instability type or material property out of 144 pulls. Modeling the human participant results in our GLMMs required the independent variables to be in this form rather than percentages. We initially chose to list percent differences in Table 2 to highlight the ranges of differences instead of an absolute value, but have added both for clarity.

Our changes to the manuscript

Page 7

“To determine if humans can detect these three different instabilities, we selected six pairs of surfaces to create a broad range of potential instabilities present across all three types. These are summarized in Table 2, where the first column for each instability is the difference in occurrence of that instability formed between each pair, and the second is the percent difference.”

“Thus, when comparing C4 versus C4-APTMS, they have a difference in steady sliding of 20 out of a maximum 144 pulls, for a |ΔSS| of 13.9%. The absolute value is taken to compare total differences present, as the psychophysical task does not distinguish between sample order.”

Comment 3

We constructed a set of coated surfaces with physical differences which were imperceptible by touch but created different types of instabilities based on how quickly a finger is slid and how hard a human finger is pressed during sliding." Yet, in your experiment, participants could discriminate them, so this is incoherent.

To clarify the point, macroscopic objects can differ in physical shape and in chemical composition. What we meant was that the physical differences, i.e., roughness, were below a limit (Skedung et al.) that participants, without a coating, would not be able to tell these apart.[23] Therefore, the reason people could tell our surfaces apart was due to the chemical composition of the surface, and not any differences in roughness or physical effects like film stiffness (due to the molecular-scale thinness of the surface coatings, they are mechanically negligible). However, we concede that at the molecular scale, the traditional macroscopic distinction between physical and chemical is blurred.

We have made minor revisions to the wording in the abstract. We clarify that the surface coatings had physical differences in roughness that were smaller than 0.6 nm, which based purely on roughness, would not be expected to be distinguishable to participants. Therefore, the reason participants can tell these surfaces apart is due to differences in friction generated by chemical composition, and we were able to minimize contributions from physical differences in the sample our study.

Our changes to the manuscript

Page 1, Abstract

“Here, we constructed a set of coated surfaces with minimal physical differences that by themselves, are not perceptible to people, but instead, due to modification in surface chemistry, the surfaces created different types of instabilities based on how quickly a finger is slid and how hard a human finger is pressed during sliding.”

“In one experiment, we used a mechanical mock finger to quantify and classify differences in instability formation from different coated surfaces. In a second experiment, participants perform a discrimination task using the same coated surfaces. Using the data from these two experiments, we found that human discrimination response times were faster with surfaces where the mock finger produced more stiction spikes and discrimination accuracy was higher where the mock finger produced more steady sliding. Conversely, traditional metrics like surface roughness or average friction coefficient did not relate to tactile discriminability. In fact, the typical method of averaging friction coefficients led to a spurious correlation which erroneously suggests that distinct objects should feel identical and identical objects should feel distinct—similar to findings by others. Friction instabilities may offer a more predictive and tractable framework of fine touch perception than friction coefficients, which would accelerate the design of tactile interfaces.”

Reviewer 3 (Public review):

Strengths

The paper describes a new perspective on friction perception, with the hypothesis that humans are sensitive to the instabilities of the surface rather than the coefficient of friction. The paper is very well written and with a comprehensive literature survey.

One of the central tools used by the author to characterize the frictional behavior is the frictional instabilities maps. With these maps, it becomes clear that two different surfaces can have both similar and different behavior depending on the normal force and the speed of exploration. It puts forward that friction is a complicated phenomenon, especially for soft materials.

The psychophysics study is centered around an odd-one-out protocol, which has the advantage of avoiding any external reference to what would mean friction or texture for example. The comparisons are made only based on the texture being similar or not.

The results show a significant relationship between the distance between frictional maps and the success rate in discriminating two kinds of surface.

We thank Reviewer 3 for their notes and interesting discussion points on our manuscript. Below, we address the reviewer’s feedback and comments on related works.

Weaknesses:

Comment 1

The main weakness of the paper comes from the fact that the frictional maps and the extensive psychophysics study are not made at the same time, nor with the same finger. The frictional maps are produced with an artificial finger made out of PDMS which is a poor substitute for the complex tribological properties of skin.

A similar comment was made by Reviewers 1 and 2. We agree in part and have made changes throughout that our study is correlative, but presents an important step forward to these biomechanical measurements and corresponding decision making models.

We are not claiming that our PDMS fingers are superior to real fingers, but rather, we cannot establish standards in the field by using real human fingers that vary between subjects and researchers. We believe the mock finger we designed is a reasonable mimic of the human finger by matching surface energy, heterogeneous mechanical structure, and the ability to test multiple physiologically relevant pressures and sliding velocities.

We achieve a heterogeneous mechanical structure with the 3 primary components of stiffness of a human finger. The effective modulus of ~100 kPa, from soft tissue,[9,10] is obtained with a 30:1 ratio of PDMS to crosslinker. The PDMS also surrounds a rigid, acrylic bone comparable to the distal phalanx, which provides an additional layer of higher modulus.[8] Additionally, the 8-hour UV-Ozone treatment decreases the viscoelastic tack of the pristine PDMS by glassifying, or further crosslinking the surface of the finger,[12] therefore imparting greater stiffness at the surface similar to the contributions of the stratum corneum, along with a similar surface energy.[13] The finger is used at least a day after UV-Ozone treatment is completed in order for the surface to return to moderate hydrophilicity, similar to the outermost layer of human skin.[17] We also discuss the shape of the contact formed. To ensure that there is minimal contribution from the slanted position of the finger, an initial contact area of 1×1 cm is established before sliding and recording friction measurements. As the PDMS finger is a soft object, the portion in contact with a surface flattens and the contact area remains largely unchanged during sliding. We recognize that it is difficult to completely control the pressure distribution due to the planar interface, but this variation is also expected when humans freely explore a surface. Finally, we consider flat vs. fingerprinted fingers. Our previous work on the role of fingerprints on friction experienced by a PDMS mock finger showed enhanced signals with the incorporation of ridges on the finger and used a rate-andstate model of a heterogenous, elastic body to find corresponding trends.[11] The key conclusion was that a flat finger still preserved key dynamic features, and the presence of stronger or more vibrations could result in more similar forces for different surfaces depending on the sliding conditions. We note that we have subsequently used the controlled mechanical data collected with this flat mock finger in correlations with human psychophysics in previous work, where findings from our mechanical experiments were predictive of human performance.[4–7] Ultimately, we see from our prior work and here that, despite the drawbacks of our mock finger, it outperforms other standard characterization technique in providing information about the mesoscale that correlates to tactile perception. We have added these details to the manuscript.

We also note that an intermediate option, replicating real fingers, even in a mold, may also inadvertently limit trends from characterization to a specific finger. One of the main – and severe – limitations of using a human finger is that all fingers are different, meaning any study focusing on a particular user may not apply to others or be recreated easily by other researchers. We cannot set a standard for replication around a real human finger as that participant may no longer be available, or willing to travel the world as a “standard”. Furthermore, the method in which a single person changes their pressures and velocities as they touch a surface is highly variable. We also note that in the Summary Response, we noted that a study by Colgate et al. (IEEE ToH 2024) demonstrated that efference copies may be important, and thus constraining a human finger and replaying the forces recorded during free exploration will not lead to the participant identifying a surface with any consistency. Thus, it is important to allow humans to freely explore surfaces, but creates nearly limitless variability in friction forces.

This is also against the backdrop that we are seeking to provide a method to characterize surfaces. Indeed, the more features we replicate in the mock finger to a human finger, the more likely it is that the mechanical data will correlate to human performance. However, we have used this technique several times to achieve stronger correlations to human data than other available techniques. We believe the metric of success should be in comparison to the available characterization technique, rather than a 1:1 reconstruction of forces of an arbitrary human finger. Indeed, a 1:1 reconstruction of forces of an arbitrary human finger would be limited to the finger of a single individual, perhaps even to that individual on a given day.

See Reviewer1 weaknesses, comment 2 part 2 for changes to the manuscript

Comment 2

The evidence would have been much stronger if the measurement of the interaction was done during the psychophysical experiment. In addition, because of the protocol, the correlation is based on aggregates rather than on individual interactions.

We agree that this would have helped further establish our argument, but in the overall statement and in other reviewer responses, we describe the significant challenges to establishing this.

To fully implement this, a decision-making model is necessary because, as a counter example, a participant could have generated 10 swipes of SFW and 1 swipe of a Sp, but the Sp may have been the most important event for making a tactile decision. We also clarify that our goals are to provide a method to characterize samples to better design tactile interfaces in haptics or in psychophysical experiments.

As discussed in the summary, and expanded on here, in our view, to develop a decision-making model, the challenges are as follows:

(1) Which one, or combination of, of the multiple swipes that people make responsible for a tactile decision?

(2) Establish what is, or may be, tactile evidence.

(3) Establish tactile decision-making models are similar or different than existing decision-making models.

(4) Test the hypothesis, in these models, that friction instabilities are evidence, and not some other unknown metric.

(5) Design a task that does not require the use of subjective tactile descriptors, like “which one feels rougher”, which we see cause confusion in participants, which will likely require accounting for memory effects.

(6) Design samples that vary in the amount of evidence generated, but this evidence cannot be controlled directly. Rather, the samples indirectly vary evidence by how likely it is for a human to generate different types of friction instabilities during standard exploration.

We elaborate these points below:

To successfully perform this experiment, we note that freely exploring humans make multiple strokes on a surface. Therefore, we would need to construct a decision-making model. It has not yet been demonstrated whether tactile decision making follows visual decision making, but perhaps to start, we can assume it does. Then, in the design of our decision-making paradigm, we immediately run into the problem: What is tactile evidence?

From Fig. 3C, we already can see that identifying evidence is challenging. Prior to this manuscript, people may have chosen the average force, or the highest force. Or we may choose the average friction force. Then, after deciding on the evidence, we need to find a method to manipulate the evidence, i.e., create samples or a machine that causes high friction, etc. We show that during the course of human touch, due to the dynamic nature of friction, the average can change a large amount and sample design becomes a central barrier to experiments. Others may suggest to immobilize the finger and applying a known force, but given how much friction changes with human exploration, there is no known method to make a machine recreate temporally and spatially varying friction forces during sliding onto a stationary finger. Finally, perhaps most importantly, in addition to mechanical challenges, a study by Liu, Colgate et al. showed that even if they recorded the friction (2D) of a finger exploring a surface and then replicated the same friction forces onto a finger, the participant could not determine which surface the replayed friction force was supposed to represent.[1] This supports that the efference copy is important, that the forces in response to expected motion are important to determine friction. Finally, there is no known method to design instabilities a priori. They must be found through experiments, especially since if we were to introduce, say a bump or a trough, then we bring in confounding variables to how participants tell surfaces apart.

Furthermore, even if we had some consistent method to create tactile “evidence”, the paradigm also deserves some consideration. In our experience, the 3-AFC task we perform is important because the vocabulary for touch has not been established. That is, in 3-AFC, by asking to determine which one sample is unlike the others, we do not have to ask the participant questions like “which one is rougher” or “which one has less friction”. In contrast, 2-AFC, which is better for decision-making models because it does not include memory, requires the asking of a perceptual question like: “which one is rougher?”. In our ongoing work, taking two silane coatings, we found that participants could easily identify which surface is unlike the others above chance in a 3-AFC, but participants, even within their own trials, could not consistently identify one silane as perceptually “rougher” by 2-AFC. To us, this calls into question the validity of tactile descriptors, but is beyond the scope of the current manuscript.

This is not our only goal, but in the context of human exploration, in this manuscript here, we believed it was important to identify a mechanical parameter that was consistent with how humans explore surfaces, but was also a parameter that could characterize to some consistent property of a surface – irrespective of whether a human was touching it. We thought that designing human decision-making models and paradigms around the friction coefficient would not be successful.

Given the scope of these challenges, we do not think it would be possible to establish this conceptual sequence in a single manuscript.

Comment 3

The authors compensate with a third experiment where they used a 2AFC protocol and an online force measurement. But the results of this third study, fail to convince the relation.

With this experiment, our central goal was to demonstrate that the instabilities we have identified with the PDMS finger also occur with a human finger. Several instances of SS, Sp, and SFW were recorded with this setup as a participant touched surfaces in real time.

Comment 4

No map of the real finger interaction is shown, bringing doubt to the validity of the frictional map for something as variable as human fingers.

Real fingers change constantly during exploration, and friction is state-dependent, meaning that the friction will depend on how the person was moving the moment prior. Therefore, a map is only valid for a single human movement – even if participants all were instructed to take a single swipe and start from zero motion, humans are unable to maintain constant velocities and pressures. Clearly, this is not sustainable for any analysis, and these drawbacks apply to any measured parameter, whether instabilities suggested here, or friction coefficients used throughout. We believe the difficulty of this approach emphasizes why a standard map of characterization of a surface by a mock finger, even with its drawbacks, is a viable path forward.

Reviewer 3 (Recommendations for the authors):

Comment 1

It would be interesting to comment on a potential connection between the frictional instability maps and Schalamack waves.

Schallamach waves are a subset of slow frictional waves (SFW). Schallamach waves are very specifically defined in the field. They occur when pockets of air that form between a soft sliding object and rigid surface which then propagate rear-to-front (retrograde waves) relative to motion of the sliding motion and form buckles due to adhesive pinning. Wrinkles then form at the detached portion of the soft material, until the interface reattaches and the process repeats.[24] There is typically a high burden of proof to establish a Schallamach wave over a more general slow frictional wave. We note that it would be exceedingly difficult to design samples that can reliably create subsets of SFW, but we are aware that this may be an interesting question at a future point in our work.

Comment 2

The force sensors look very compliant, and given the dynamic nature of the signal, it is important to characterize the frequency response of the system to make sure that the fluctuations are not amplified.

Thank you for noticing. We mistyped the sensor spring constant as 13.9 N m^-1 instead of kN m^-1. However, below we show how the instabilities are derived from the mechanics at the interface due to the compliance of the finger. The “springs” of the force sensor and PDMS finger are connected in parallel. Since k_sensor = 13.9 kN m^-1, the spring constant of the system overall reflects the compliance of the finger, and highlights the oscillations arising solely from stick-slip. A sample calculation is shown below.

Author response image 1.

Fitting a line to the initial slope of the force trace for C6 gives the equation y = 25.679x – 0.2149. The slope here represents force data over time data, and is divided by the velocity (25 mm/s) to determine the spring constant of the system k_total == 1027.16 N/m. This value is lower than k_sensor = 13.9 kN/m, indicating that the “springs” representing the force sensor and PDMS finger are connected in parallel:

. The finger is the compliant component of the system, with k_finger = 1.11 kN/m, and of course, real human fingers are also compliant so this matches our goals with the design of the mock finger.

Our changes to the manuscript

(Page 4) (k = 13.9 kN m¹)

Comment 3

The authors should discuss about the stochastic nature of friction: - Wiertlewski, Hudin, Hayward, IEEE WHC 2011 Greenspon, McLellan, Lieber, Bensmaia, JRSI 2020.

We believe that, given the references, this comment on “stochastic” refers to the macroscopically-observable fluctuations (i.e., the mechanical “noise” which is not due to instrument noise) in friction arising from the discordant network of stick-slip phenomena occurring throughout the contact zone, and not the stochastic nature of nanoscale friction that occurs thermal fluctuations nor due to statistical distributions in bond breaking associated with soft contact.

We first note that our small-scale fluctuations do not arise from a periodic surface texture that dominates in the frequency regime. However, even on our comparatively smooth surfaces, we do expect fluctuations due to nanoscale variation in contact, generation of stick-slip across at microscale length scales that occur either concurrently or discordantly across the contact zone, and the nonlinear dependence of friction to nearly any variation in state and composition.[11]

Perhaps the most relevant to the manuscript is that a major advantage of analysis by friction is that it sidesteps these ever-present microscale fluctuations, leading to more clearly defined classifiers or categories during analysis. Wiertlewski et. al. showed repeated measurements in their systems ultimately gave rise to consistent frequencies[25] (we think their system was in a steady sliding regime and the patterning gave rise to underlying macroscopic waves). These consistent frequencies, at least in soft systems and absent obvious macroscopic patterned features, would be expected to arise from the instability categories and we see them throughout.

Comment 4

It is stated that "we observed a spurious, negative correlation between friction coefficient and accuracy".

What makes you qualify that correlation as spurious?

We mean this as in the statistical definition of “spurious”.

This correlation would indicate that by the metric of friction coefficient, more different surfaces are perceived more similarly. Thus, two very different surfaces, like Teflon and sandpaper, by friction coefficient would be expected to feel very similar. Two nearly identical surfaces would be expected to feel very different – but of course, humans cannot consistently distinguish two identical surfaces. This finding is counterintuitive and refutes that friction coefficient is a reliable classifier of surfaces by touch. We do not think it is productive to determine a mechanism for a spurious correlation, but perhaps one reason we were able to observe this is because our study, to the best of our knowledge, is unique for having samples that are controlled in their physical differences in roughness and surface features.

See response to Reviewer 1 weaknesses, comment 1 for changes to the manuscript

Comment 5

The authors should comment on the influence of friction on perceptual invariance. Despite inducing radially different frictional behavior for various conditions, these surfaces are stably perceived. Maybe this is a sign that humans extract a different metric?

We agree – we are excited that frictional instabilities may offer a more stable perceptual cue because they are not prone to fluctuations (as discussed in Comment 3) and instability formation, in many conditions, is invariant to applied pressures and velocities – thus forming large zones where a human may reasonable encounter a given instability.

Raw friction is highly prone to variation during human exploration (in alignment with Recommendations for the authors, Comment 3), but ongoing work seeks to explain tactile constancy, or the ability to identify objects despite these large changes in force. Very recently published work by Fehlberg et. al. identified the role of modulating finger speed and normal force in amplifying the differences in friction coefficient between materials in order to identify them,[26] and we postulate that their work may be streamlined and consistent with the idea of friction instabilities, though we have not had a chance to discuss this in-depth with the authors yet.

We think that the instability maps show a viable path forward to how surfaces are stably perceived, and instabilities themselves show a potential mechanism: mathematically, instabilities for given conditions can be invariant to velocity or mass, creating zones where a certain instability is encountered. This reduces the immense variability of friction to a smaller, more stable classification of surfaces (e.g., a 30% SS surface or a 60% SS surface). A given surface will typically produce the same instability at a specific condition (we found some boundaries of experimental parameters are very condition sensitive, but many conditions are not), whereas a single friction trace which is highly prone to variation is not a stable metric.

Added Reference

(53) M. Fehlberg, E. Monfort, S. Saikumar, K. Drewing and R. Bennewitz, IEEE Trans. Haptics, 2024, 17, 957–963.

References

(1) Liu, Z., Kim, J.-T., Rogers, J. A., Klatzky, R. L. & Colgate, J. E. Realism of Tactile Texture Playback: A Combination of Stretch and Vibration. IEEE Trans. Haptics 17, 441–450 (2024).

(2) Waters, I., Alazmani, A. & Culmer, P. Engineering Incipient Slip Into Surgical Graspers to Enhance Grasp Performance. IEEE Transactions on Medical Robotics and Bionics 2, 541–544 (2020).

(3) Gueorguiev, D., Bochereau, S., Mouraux, A., Hayward, V. & Thonnard, J.-L. Touch uses frictional cues to discriminate flat materials. Sci Rep 6, 25553 (2016).

(4) Carpenter, C. W. et al. Human ability to discriminate surface chemistry by touch. Mater. Horiz. 5, 70– 77 (2018).

(5) Nolin, A. et al. Predicting human touch sensitivity to single atom substitutions in surface monolayers for molecular control in tactile interfaces. Soft Matter 17, 5050–5060 (2021).

(6) Nolin, A. et al. Controlling fine touch sensations with polymer tacticity and crystallinity. Soft Matter 18, 3928–3940 (2022).

(7) Swain, Z. et al. Self-Assembled Thin Films as Alternative Surface Textures in Assistive Aids with Users Who are Blind. J. Mater. Chem. B (2024) doi:10.1039/D4TB01646G.

(8) Qian, K. et al. Mechanical properties vary for different regions of the finger extensor apparatus. J Biomech 47, 3094–3099 (2014).

(9) Abdouni, A. et al. Biophysical properties of the human finger for touch comprehension: influences of ageing and gender. Royal Society Open Science (2017) doi:10.1098/rsos.170321.

(10) Cornuault, P.-H., Carpentier, L., Bueno, M.-A., Cote, J.-M. & Monteil, G. Influence of physicochemical, mechanical and morphological fingerpad properties on the frictional distinction of sticky/slippery surfaces. Journal of The Royal Society Interface (2015) doi:10.1098/rsif.2015.0495.

(11) Dhong, C. et al. Role of fingerprint-inspired relief structures in elastomeric slabs for detecting frictional differences arising from surface monolayers. Soft Matter 14, 7483–7491 (2018).

(12) Fu, Y.-J. et al. Effect of UV-Ozone Treatment on Poly(dimethylsiloxane) Membranes: Surface Characterization and Gas Separation Performance. Langmuir 26, 4392–4399 (2010).

(13) Yuan, Y. & Verma, R. Measuring microelastic properties of stratum corneum. Colloids Surf B Biointerfaces 48, 6–12 (2006).

(14) Yu, G. et al. A wearable pressure sensor based on ultra-violet/ozone microstructured carbon nanotube/polydimethylsiloxane arrays for electronic skins. Nanotechnology 29, 115502 (2018).

(15) Zheng, L. et al. Dual-Stimulus Smart Actuator and Robot Hand Based on a Vapor-Responsive PDMS Film and Triboelectric Nanogenerator. ACS Appl. Mater. Interfaces 11, 42504–42511 (2019).

(16) Ma, K., Rivera, J., Hirasaki, G. J. & Biswal, S. L. Wettability control and patterning of PDMS using UV–ozone and water immersion. Journal of Colloid and Interface Science 363, 371–378 (2011).

(17) Mavon, A. et al. Sebum and stratum corneum lipids increase human skin surface free energy as determined from contact angle measurements: A study on two anatomical sites. Colloids and Surfaces B: Biointerfaces 8, 147–155 (1997).

(18) AliAbbasi, E. et al. Effect of Finger Moisture on Tactile Perception of Electroadhesion. IEEE Trans. Haptics 17, 841–849 (2024).

(19) Corniani, G. et al. Sub-surface deformation of individual fingerprint ridges during tactile interactions.

eLife 13, (2024).

(20) Israelachvili, J. N. Intermolecular and Surface Forces. (Academic Press, 2011).