Author response:

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

eLife Assessment

This valuable study reports the development of a novel organoid system for studying the emergence of autorhythmic gut peristaltic contractions through the interaction between interstitial cells of Cajal and smooth muscle cells. While the utility of the organoids for studying hindgut development is well illustrated by showing, for example, a previously unappreciated potential role for smooth muscle cells in regulating the firing rate of interstitial cells of Cajal, some of the functional analyses are incomplete. There are some concerns about the specificity and penetrance of perturbations and the reproducibility of the phenotypes. With these concerns properly addressed, this paper will be of interest to those studying the development and physiology of the gut.

We greatly appreciate constructive comments raised by the Editors and all the Reviewers. We have newly conducted pharmacological experiments using Nifedipine, a L-type Ca²⁺ blocker known to operate in smooth muscles (new Fig 7). The treatment abrogated not only the oscillation of SMCs but also that in ICCs, further corroborating our model that not only ICC-to-SMC interactions but also the reverse direction, namely SMC-to-ICC feedback signals, are operating to achieve coordinated/stable rhythm of gut contractile organoids.

Concerning the issues of the specificity and penetrance in pharmacological experiments with gap junction inhibitors, we have carefully re-examined effects by multiple blockers (CBX and 18b-GA) at different concentrations (new Fig 5D and Fig. S3B).We have newly found that: (1) the effects observed by CBX (100 µM) that the latency of Ca²⁺ peaks between ICCs (preceding) and SMCs (following) was abolished are not seen by 18b-GA at any concentrations including 100 µM, implying that the latency of Ca²⁺ peaks between these cells is governed by connexin(s) that are not inhibited by18bGA. Such difference in inhibiting effects by these two drugs were previously reported in multiple model systems including guts (Daniel et al., 2007; Parsons & Huizinga, 2015; Schultz et al., 2003).

Regarding the penetrance of the drugs, we have carried out earlier administration (Day 3) of the gap junction inhibitor, either CBX (100 µM) or 18b-GA (100 µM), in the course of organoidal formation in culture when cells are still at 2D to exclude a possible penetrance problem (new Fig. S3C). There treatments render no or little effects to the patterns of organoidal contractions in a way similar to the drug administration at Day 7. As already shown in the first version, CBX (100 µM) eliminates the latency of Ca²⁺ peaks, we believe that this drug successfully penetrates into the organoid and exerts its specific effects.

Unfortunately, due to very unstable condition in climate including extreme heat and sporadically occurring bird flu epidemic since the last summer in Japan, the poultry farm must have faced problems. In the course of revision experiments, we got in a serious trouble at multiple times with unhealthy eggs/embryos lasting from last summer until present. These unfortunate incidents did not allow us to engage in the revision experiments as fully as we originally planned. Nevertheless, we did our very best within a limited time fame, and we believe that the revised version is suitable as a final version of an eLife article.

Public Reviews:

Reviewer #1 (Public Review):

Summary:

In this study, the authors developed an organoid system that contains smooth muscle cells (SMCs) and interstitial cells of Cajal (ICCs; pacemaker) but few enteric neurons, and generates rhythmic contractions as seen in the developing gut. The stereotypical arrangements of SMCs and ICCs in the organoid allowed the authors to identify these cell types in the organoid without antibody staining. The authors took advantage of this and used calcium imaging and pharmacology to study how calcium transients develop in this system through the interaction between the two types of cells. The authors first show that calcium transients are synchronized between ICC-ICC, SMC-SMC, and SMC-ICC. They then used gap junction inhibitors to suggest that gap junctions are specifically involved in ICC-to-SMC signaling. Finally, the authors used an inhibitor of myosin II to suggest that feedback from SMC contraction is crucial for the generation of rhythmic activities in ICCs. The authors also show that two organoids become synchronized as they fuse and SMCs mediate this synchronization.

Strengths:

The organoid system offers a useful model in which one can study the specific roles of SMCs and ICCs in live samples.

Thank you very much for the constructive comments.

Weaknesses:

Since only one blocker each for gap junction and myosin II was used, the specificities of the effects were unclear.

We appreciate these comments. We have addressed those of “weaknesses” as described in “Responses to the eLife assessment” (please see above).

Reviewer #2 (Public Review):

Summary:

In this study, Yagasaki et al. describe an organoid system to study the interactions between smooth muscle cells (SMCs) and interstitial cells of Cajal (ICCs). While these interactions are essential for the control of rhythmic intestinal contractility (i.e., peristalsis), they are poorly understood, largely due to the complexity of and access to the in vivo environment and the inability to co-culture these cell types in vitro for long term under physiological conditions. The "gut contractile organoids" organoids described herein are reconstituted from stromal cells of the fetal chicken hindgut that rapidly reorganize into multilayered spheroids containing an outer layer of smooth muscle cells and an inner core of interstitial cells. The authors demonstrate that they contract cyclically and additionally use calcium imagining to show that these contractions occur concomitantly with calcium transients that initiate in the interstitial cell core and are synchronized within the organoid and between ICCs and SMCs. Furthermore, they use several pharmacological inhibitors to show that these contractions are dependent upon non-muscle myosin activity and, surprisingly, independent of gap junction activity. Finally, they develop a 3D hydrogel for the culturing of multiple organoids and found that they synchronize their contractile activities through interconnecting smooth muscle cells, suggesting that this model can be used to study the emergence of pacemaking activities. Overall, this study provides a relatively easy-to-establish organoid system that will be of use in studies examining the emergence of rhythmic peristaltic smooth muscle contractions and how these are regulated by interstitial cell interactions. However, further validation and quantification will be necessary to conclusively determine show the cellular composition of the organoids and how reproducible their behaviors are.

Strengths:

This work establishes a new self-organizing organoid system that can easily be generated from the muscle layers of the chick fetal hindgut to study the emergence of spontaneous smooth muscle cell contractility. A key strength of this approach is that the organoids seem to contain few cell types (though more validation is needed), namely smooth muscle cells (SMCs) and interstitial cells of Cajal (ICCs). These organoids are amenable to live imaging of calcium dynamics as well as pharmacological perturbations for functional assays, and since they are derived from developing tissues, the emergence of the interactions between cell types can be functionally studied. Thus, the gut contractile organoids represent a reductionist system to study the interactions between SMCs and ICCs in comparison to the more complex in vivo environment, which has made studying these interactions challenging.

Thank you very much for the constructive comments.

Weaknesses:

The study falls short in the sense that it does not provide a rigorous amount of evidence to validate that the gut organoids are made of bona fide smooth muscle cells and ICCs. For example, only two "marker" proteins are used to support the claims of cell identity of SMCs and ICCs. At the same time, certain aspects of the data are not quantified sufficiently to appreciate the variance of organoid rhythmic contractility. For example, most contractility plots show the trace for a single organoid. This leads to a concern for how reproducible certain aspects of the organoid system (e.g. wavelength between contractions/rhythm) might be, or how these evolve uniquely over time in culture. Furthermore, while this study might be able to capture the emergence of ICC-SMC interactions as they related to muscle contraction and pacemaking, it is unclear how these interactions relate to adult gastrointestinal physiology given that the organoids are derived from fetal cells that might not be fully differentiated or might have distinct functions from the adult. Finally, despite the strength of this system, discoveries made in it will need to be validated in vivo. Thank you very much for the comments, which are helpful to improve our MS. In the revised version, we have additionally used antibody against desmin, known to be a maker for mature SMCs (new Fig 3B). The signal is seen only in the peripheral cells overlapping with the αSMA staining (line 169-170).

Concerning the reproducibility, while contractility changes were shown for a representative organoid in the original version, experiments had been carried out multiple times, and consistent data were reproduced as already mentioned in the text of the first version of MS. However, we agree with this reviewer that it must be more convincing if we assess quantitatively. We have therefore conducted quantitative assessments of organoidal contractions and Ca²⁺ transients (new Fig. 2B, new Fig. 4D, new Fig 5D, E, new Fig. 6B, new Fig. 7B, new Fig. 8C, new Fig. S2, S3). Details such as repeats of experiments and size of specimens are carefully described in the revised version (Figure legends)

In particular, in place of contraction numbers/time, we have plotted “contraction intervals” between two successive peaks (Fig. 2B and others). Actually, with your suggestion, we have tried to perform a periodicity analysis of organoid contractions. Unfortunately, no clear value has been obtained, probably because the contractions/Ca²⁺ transitions are not as “regularly periodical” as seen in conventional physics. This led us to perform the peak-interval analysis. Methods to quantify the contraction intervals are carefully explained in the revised version.

As already mentioned in the “Our provisional responses” following the receipt of Reviewers’ comments, we agree that our organoids derived from embryonic hind gut (E15) might not necessarily recapitulate the full function of cells in adult. However, it has well been accepted in the field of developmental biology that studies with embryonic tissue/cells make a huge contribution to unveil complicated physiological cell functions. Nevertheless, we have carefully considered in the revised version so that the MS would not send misleading messages. We agree that in vivo validation of our gut contractile organoid must be wonderful, and this is a next step to go.

Reviewer #3 (Public Review):

Summary:

The paper presents a novel contractile gut organoid system that allows for in vitro studying of rudimentary peristaltic motions in embryonic tissues by facilitating GCaMPlive imaging of Ca²⁺ dynamics, while highlighting the importance and sufficiency of ICC and SMC interactions in generating consistent contractions reminiscent of peristalsis. It also argues that ENS at later embryonic stages might not be necessary for coordination of peristalsis.

Strengths:

The manuscript by Yagasaki, Takahashi, and colleagues represents an exciting new addition to the toolkit available for studying fundamental questions in the development and physiology of the hindgut. The authors carefully lay out the protocol for generating contractile gut organoids from chick embryonic hindgut, and perform a series of experiments that illustrate the broader utility of these organoids for studying the gut. This reviewer is highly supportive of the manuscript, with only minor requests to improve confidence in the findings and broader impact of the work. These are detailed below.

Thank you very much for the constructive comments.

Weaknesses:

(1) Given that the literature is conflicting on the role GAP junctions in potentiating communication between intestinal cells of Cajal (ICCs) and smooth muscle cells (SMCs), the experiments involving CBX and 18Beta-GA are well-justified. However, because neither treatment altered contractile frequency or synchronization of Ca++ transients, it would be important to demonstrate that the treatments did indeed inhibit GAP junction function as administered. This would strengthen the conclusion that GAP junctions are not required, and eliminate the alternative explanation that the treatments themselves failed to block GAP junction activity.

Thank you for these comments, and we agree. In the revised version, we have verified the drugs, CBX and 18b-GA, using dissociated embryonic heart cells in culture, a well-established model for the gap junction study (new Fig. S3D, line 237-239). Expectedly, both inhibitors abrogate the rhythmic beats of heart cells, and importantly, cells’ beats resume after wash-out of the drug.

(2) Given that 5uM blebbistatin increases the frequency of contractions but 10uM completely abolishes contractions, confirming that cell viability is not compromised at the higher concentration would build confidence that the phenotype results from inhibition of myosin activity. One could either assay for cell death, or perform washout experiments to test for recovery of cyclic contractions upon removal of blebbistatin. The latter may provide access to other interesting questions as well. For example, do organoids retain memory of their prior setpoint or arrive at a new firing frequency after washout?

We greatly appreciate these suggestions and also interesting ideas to explore! In the revised version, we have newly conducted washout experiments (new Fig. 6B) (10 µM drug is washed-out from culture medium), and found that contractions resume, showing that cell viability is not compromised at 10 µM concentration (line 257-259). Intriguingly, the resumed rhythm appears more regular than that before drug administration. Thus, the contraction rhythm of the organoid might be determined by cellcell interactions at any given time rather than by memory of their prior setpoint. This is an interesting issue we would like to further explore in the future. These issues, although potentially interesting, are not mentioned in the text of the revised version, since it is too early to interpret there observations.

(3) Regulation of contractile activity was attributed to ICCs, with authors reasoning that Tuj1+ enteric neurons were only present in organoids in very small numbers (~1%).

However, neuronal function is not strictly dependent on abundance, and some experimental support for the relative importance of ICCs over Tuj1+ cells would strengthen a central assumption of the work that ICCs the predominant cell type regulating organoid contraction. For example, one could envision forming organoids from embryos in which neural crest cells have been ablated via microdissection or targeted electroporation. Another approach would be ablation of Tuj1+ cells from the formed organoids via tetrodotoxin treatment. The ability of organoids to maintain rhythmic contractile activity in the total absence of Tuj1+ cells would add confidence that the ICCs are indeed the driver of contractility in these organoids.

We agree. In the revised version, we have conducted TTX administration (new Fig. S2C). Changes in contractility by this treatment is not detected, supporting the argument that neural cells/activities are not essential for rhythmic contractions of the organoid (line 178-181).

(4) Given the implications of a time lag between Ca++ peaks in ICCs and SMCs, it would be important to quantify this, including standard deviations, rather than showing representative plots from a single sample.

In the revised version, we have elaborated a series of quantitative assessments as mentioned above (please see our responses to the “eLife assessments” at the beginning of these correspondences). The latency between Ca²⁺ peaks in ICCs and SMCs is shown in new Fig. 4D, in which measured value is 700 msec-terraced since the time-lapse imaging was performed with 700 msec intervals (as already described in the first version).

117 peaks for 14 organoids have been assessed (line 218).

(5) To validate the organoid as a faithful recreation of in vivo conditions, it would be helpful for authors to test some of the more exciting findings on explanted hindgut tissue. One could explant hindguts and test whether blebbistatin treatment silences peristaltic contractions as it does in organoids, or following RCAS-GCAMP infection at earlier stages, one could test the effects of GAP junction inhibitors on Ca++ transients in explanted hindguts. These would potentially serve as useful validation for the gut contractile organoid, and further emphasize the utility of studying these simplified systems for understanding more complex phenomena in vivo.

Thank you very much for insightful comments. We would love to explore these issues in near future. Just a note is that it was previously reported that Nifedipine silences peristaltic contractions in ex-vivo cultured gut (Chevalier et al., 2024; Der et al., 2000).

(6) Organoid fusion experiments are very interesting. It appears that immediately after fusion, the contraction frequency is markedly reduced. Authors should comment on this, and how it changes over time following fusion. Further, is there a relationship between aggregate size and contractile frequency? There are many interesting points that could be discussed here, even if experimental investigation of these points is left to future work.

It would indeed be interesting to explore how cell communications affect/determine the contraction rhythm, and our novel organoids must serve as an excellent model to address these fundamental questions. We have observed multiple times that when two organoids fuse, they undergo “pause”, and resume coordinated contractions as a whole, and we have mentioned such notice briefly in the revised version (line 282). To know what is going on during this pause time should be tempting. In addition, we have an impression that the larger in size organoids grow, the slower rhythm they count. We would love to explore this in near future.

(7) Minor: As seen in Movie 6 and Figure 6A, 5uM blebbistatin causes a remarkable increase in the frequency of contractions. Given the regular periodicity of these contractions, it is a surprising and potentially interesting finding, but authors do not comment on it. It would be helpful to note this disparity between 5 and 10 uM treatments, if not to speculate on what it means, even if it is beyond the scope of the present study to understand this further.

We assume that the increase in the frequency of contractions at 5 µM might be due to a shorter refractory period caused by a decreasing magnitude (amplitude) of contraction. We have made a short description in the revised text (line 256-257).

(8) Minor: While ENS cells are limited in the organoid, it would be helpful to quantify the number of SMCs for comparison in Supplemental Figure S2. In several images, the number of SMCs appears quite limited as well, and the comparison would lend context and a point of reference for the data presented in Figure S2B.

In the revised version, the number of SMCs has been counted and added in Fig. S2B. Contrary to that SMCs are more abundant than ICCs in an intact gut, the proportion is reversed in our organoid (line 181-183). It might due to treatments during cell dissociation/plating.

(9) Minor: additional details in the Figure 8 legend would improve interpretation of these results. For example, what is indicated in orange signal present in panels C, G and H? Is this GCAMP?

We apologize for this confusion. In the revised version, we have added labeling directly in the photos of new Fig. 9 (old Fig. 8). For C, G and H, the left photo is mRuby3+GCaMP6s, and the right one is GCaMP6s only.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

I have a few comments for the authors to consider:

(1) Figure 4C: The authors propose that calcium signals propagate from ICC to SMC based on the results presented in this figure. While it is observed that the peak of the calcium signal in ICC precedes that in SMC, it's worth noting that the onset of the rise in calcium signals occurs simultaneously in ICC and SMC. Doesn't this suggest that they are activated simultaneously? The latency observed for the peaks of calcium signals could reflect different kinetics of the rise in calcium concentration in the two types of cells rather than the order of calcium signal propagation.

We greatly appreciate these comments. We have re-examined kinetics of GCaMP signals in ICC and SMC, but we did not succeed in validating rise points precisely. We agree that the possibility that the rise in calcium signals could be occurring simultaneously. To clarify these issues, analyses with higher resolution is required, such as using GCaMP6f or GCaMP7/8. Nevertheless, the disappearance of the latency of Ca²⁺ peak by CBX implies a role of gap junction in ICC to SMC signaling. In the revised version, we replaced the wording “rise” by “peak” when the latency is discussed.

(2) Figure 5C: The specific elimination of the latency in the calcium signal peaks between ICC and SMC is interesting. However, I am curious about how gap junction inhibitors specifically eliminate the latency between ICC and SMC without affecting other aspects of calcium transients in these cells, such as amplitude and synchronization among ICCs and/or SMCs. Readers of the manuscript would expect some discussion on possible mechanisms underlying this specificity. Additionally, I wonder if the elimination of the latency was observed consistently across all samples examined. The authors should provide information on the frequency and number of samples examined, and whether the elimination occurs when 18-beta-GA is used.

In the revised version, we have elaborated quantitative demonstration. For the effects by CBX on latency or Ca²⁺ peaks, a new graph has been added to new Fig 5, in which 100 µM eliminated the latency. Intriguingly, the latency appears to be attributed to a gap junction that is not inhibited by18-beta-GA (please see new Fig. S3E). As already mentioned above, inhibiting activity of both CBX and 18-beta-GA has been verified using dissociated cells of embryonic heart, a popular model for gap junction studies.

At present, we do not know how gap junction(s) contribute to the latency of Ca²⁺ peaks without affecting synchronization among ICCs and/or SMCs (we have not addressed amplitude of the oscillation in this study). Actually, it was surprising to us to find that GJ’s contribution is very limited. We do not exclude the importance of GJs, and currently speculate that GJs might be important for the initiation of contraction/oscillation signals, whereas the requirement of GJs diminishes once the ICC-SMC interacting rhythm is established. What we observed in this study might be the synchronization signals AFTER these interactions are established (Day 7 of organoidal culture). Upon the establishment, it is possible that mechanical signaling elicited by smooth muscles’ contraction might become prominent as a mediator for the (stable) synchronization, as implicated by experiments with blebbistatin and Nifedipin, the latter being newly added to the revised version (new Fig. 7). We have added such speculation, although briefly in Discussion (line 374-377)

(3) Figure 6: The significant effects of blebbistatin on calcium dynamics in both ICC and SMC are intriguing. However, since only one blocker is utilized, the specificity of the effects is unclear. If other blockers for muscle contraction are available, they should be employed. Considering that a rise in calcium concentration precedes contraction, calcium transients should persist even if muscle contraction is inhibited. One concern is whether blebbistatin inadvertently rendered the cells unhealthy. The authors should demonstrate at least that contraction and calcium transients recover after removal of the drug. The frequency and number of samples examined should be shown, as requested for Figure 5C above.

Thank you for these critical comments. A possible harmfulness of the drugs was also raised by other reviewers, and we have therefore conducted wash-out experiments in the revised version (new Fig. 6B). Contractions resume after wash-out showing that cell viability is not compromised at 10 µM concentration. The number of samples examined has been described more explicitly in the revised version. Regarding the blocker of SMC, we have newly carried out pharmacological assays using nifedipine, a blocker of a L-type Ca²⁺ channel known to operate in smooth muscle cells (new Fig 7) (Chevalier et al., 2024; Der et al., 2000). As already explained in the “Responses to eLife assessment”, the treatment abrogated ICCs’ rhythm and synchronous Ca²⁺ transients between ICCs and SMCs, further corroborating our model that not only ICC-to-SMC interactions but also SMC-to-ICC feedback signals are operating to achieve coordinated/stable rhythm of gut contractile organoids of Day 7 culture (please also see our responses shown above for Comment (2)).

Reviewer #2 (Recommendations For The Authors):

Major:

(1) The claim that organoids contain functional SMCs and ICCs is insufficient as it currently relies on only c-Kit and aSMA antibodies. This conclusion could be additionally supported by staining with other markers of contractile smooth muscle (e.g. TAGLN and MYH14) and an additional accepted marker of ICCs (e.g. ANO1/TMEM16). Moreover, it should be demonstrated whether these cells are PDGFRA+, as PDGFRA is a known marker of other mesenchymal fibroblast cell types. These experiments would additionally rule out whether these cells were simply less differentiated myofibroblasts. Given that there might not be available antibodies that react with chicken protein versions, the authors could support their conclusions using alternative approaches, such as fluorescent in situ hybridization. A more thorough approach, such as single-cell RNA sequencing to compare the cell composition of the in vitro organoids to the in vivo colon, would fully justify the use of these organoids as a system for studying in vivo cell physiology.

With these suggestions provided, we have newly stained contractile organoids with anti-desmin antibody, known to be a marker for differentiated SMCs. As shown in new Fig. 3B, desmin-positive cells perfectly overlapped with aSMA-staining, indicating that the peripherally enclosing cells are SMCs. Regarding the interior cells, as this Reviewer concerned, there are no antibodies against ANO1/TMEM16 which are available for avian specimens. The anti- c-Kit antibody used in this study is what we raised in our hands by spending years (Yagasaki et al., 2021)), in which the antibody was carefully validated in intact guts of chicken embryos by multiple methods including Western Blot analyses, immunostaining, and in situ hybridization. We have attempted several times to perform organoidal whole-mount in situ hybridization for expression of PDGFRα, but we have not succeeded so far. In addition, as explained to the Editor, the very unhealthy condition of purchased eggs these past 7 months did not allow us to continue any further. We are planning to interrogate cell types residing in the central area of the organoid, results of which will be reported in a separate paper in near future.

(2) The key ICC-SMC relationship and physiological interaction seems to arise developmentally, but the mechanisms of this transition are not well defined (Chevalier 2020). To further support the claim that ICC-SMC interactions can be interrogated in this system, this study would benefit from establishing organoids at distinct developmental stages to (a) show that they have unique contractile profiles, and (b) demonstrate that they evolve over time in vitro toward an ICC-driven mechanism.

We agree with these comments. We tried to prepare gut contractile organoids derived from different stages of development, and we had an impression that slightly younger hindguts are available for the organoid preparations. In addition, not only the hindgut, but also midgut and caecum also yield organoids. However, since formed organoids derived from these “non-E15 hindgut” vary substantially in shapes, contraction frequencies/amplitudes etc., we are currently not ready to report these preliminary observations. Instead, we decided to optimize and elaborate in vitro culture conditions by focusing on the E15 hindgut, which turned out to be most stable in our hands. Nevertheless, it is tempting to see how organoid evolves over time during gut development.

(3) This manuscript would be greatly enhanced by a functional examination of the prospective organoid ICCs. For example, the authors could test whether the c-Kit inhibitor Imatinib, which has previously been used to impair ICC differentiation and function in the developing chick gut (Chevalier 2020), has an effect on contractility at different stages.

Following the paper of (Chevalier 2020), we had already conducted similar experiments with Imatinib in the culture with our organoids, but we did not see detectable effects. In that paper, the midgut of younger embryos was used, whereas we used E15 hindgut to prepare organoids. It would be interesting to see if we add Imanitib earlier during organoidal formation, and this is a next step to go.

(4) It is claimed that there is a 690s msec delay in SMC spike relative to ICC spike, however, it is unclear where this average is derived from and whether the organoid calcium trace shown in Figure 4C is representative of the data. The latency quantification should be shown across multiple organoids, and again in the case of carbenoxolone treatment, to better understand the variations in treatment.

We apologize that the first version failed to clearly demonstrate quantitative assessments. In the revised version, we have elaborated quantitative assessments (117 peaks for 14 organoids) (line 216-218). In new Fig. 4D, measured value is 700 msecterraced since as already mentioned in the first version, the time-lapse imaging was performed with 700 msec intervals.

(5) As above, a larger issue is that only single traces are shown for each organoid. This makes it challenging to understand the variance in contractile properties across multiple organoids. While contraction frequencies are shown several times, the manuscript would benefit from additional quantifications, such as rhythm (average wavelength between events) in control and perturbed conditions.

We have substantially elaborated quantitative assessments (please also see our responses to the “Public Review”). In particular, in place of contraction numbers/time, we have plotted “contraction intervals” between two successive peaks (Fig. 2B and others). Actually, we have tried to perform a periodicity analysis of organoid contractions. Unfortunately, no clear value has been obtained, probably because the contractions/Ca²⁺ transitions are not as “regularly periodical” as seen in conventional physics. This led us to perform the peak-interval analysis. Methods to quantify the contraction intervals are carefully explained in the revised version.

(6) The synchronicity observed between ICCs and SMCs within the organoid is interesting, and should be emphasized by making analyses more quantitative so as to understand how consistent and reproducible this phenomenon is across organoids. Moreover, one of the most exciting parts of the study is the synchronicity established between organoids in the hydrogel system, but it is insufficiently quantified. For example, how rapidly is pacemaking synchronization achieved?

As we replied above to (5), and described in the responses to the “Public Review”, we have substantially elaborated quantitative assessments in the revised version. Concerning the synchronicity between ICCs and SMCs, our data explicitly show that as long as the organoid undergoes healthy contraction, they perfectly match their rhythm (Fig. 4) making it difficult to display quantitatively. Instead, to demonstrate such synchronicity more convincingly, we have carefully described the number of peaks and the number of independent organoids we analyzed in each of Figure legends. In the experiments with hydrogels, the time required for two organoids to start/resume synchronous contraction varies greatly. For example, for the experiment shown in new Fig 9F, it takes 1 day to 2 days for cells crawling out of organoids and cover the surface of the hydrogel. In the experiments shown in new Fig. 8, two organoids undergo “pause” before resuming contractions. In the revised version, we have briefly mentioned our notice and speculation that active cell communications take place during this pausing time, (line 282-283 in Result and line 437-439 in Discussion). We agree with this reviewer saying that the pausing time is potentially very interesting. However, it is currently difficult to quantify these phenomena. More elaborate experimental design might be needed.

(7) Smooth muscle layers in vivo are well organized into circular and longitudinal layers. To establish physiological relevance, the authors should demonstrate if these organoids have multiple layers (though it looks like just a single outer layer) and if they show supracellular organization across the organoid.

The immunostaining data suggest that peripherally lining cells are of a single layer, and we assume that they might be aligned in register with contracting direction. However, to clarify these issues, observation with higher resolution would be required.

(8) To further examine whether the organoids contain true functional ICCs, the authors should test whether their calcium transients are impacted by inhibitors of L-type calcium channels, such as nifedipine and nicardipine. These channels have been demonstrated to be important for SMCs but not ICCs, so one might expect to see continued transients in the core ICCs but a loss of them in SMCs (Lee et al., 1999; PMID: 10444456)

We appreciate these comments. We have accordingly conducted new experiments with Nifedipine. Contrary to the expectation, Nifedipine ceases not only organoidal contractions, but also ICC activities (and its resulting synchronization) (new Fig. 7). These findings actually corroborate our model already mentioned in the first version that ICCs receive mechanical feedback from SMC’s contraction to stably maintain their oscillatory rhythm. We believe that the additional findings with Nifedipine have improved the quality of our paper. Concerning the central cells in the organoid, we have additionally used anti-desmin antibody known to mark differentiated SMCs. Desmin signals perfectly overlap with those of aSMA in the peripheral single layer, supporting that the peripheral cells are SMCs and central cells are ICCs. The anti c-Kit antibody used in this study is what we raised in our hands by spending years (Yagasaki et al., 2021)), in which the antibody was carefully validated in intact guts of chicken embryos by multiple methods including Western Blot analyses, immunostaining, and in situ hybridization.

ANO1/TMEM16 are known to stain ICCs in mice. Antibodies against ANO1/TMEM16 available for avian specimens are awaited.

(9) Despite Tuj1+ enteric neurons only making up a small fraction of the organoids, the authors should still functionally test whether they regulate any aspect of contractility by treating organoids with an inhibitor such as tetrodotoxin to rule out a role for them.

Thank you for these advices, which are also raised by other reviewers. We have conducted TTX administration (new Fig. S2C). Changes in contractility by this treatment is not detected, supporting the argument that neural cells/activities are not essential for rhythmic contractions of the organoid (line 178-181).

(10) Finally, the manuscript is written to suggest that the focus of the study is to establish a system to interrogate ICC-SMC interactions in gut physiology and peristalsis. However, the organoids designed in this study are derived from the fetal precursors to the adult cell types. Thus, they might not accurately portray the adult cell physiology. I don't believe that this is a downfall, but rather a strength of the study that should be emphasized. That is, the focus could be shifted toward stressing the power of this new system as a reductionist, self-organizing model to examine the developmental emergence of contractile synchronization in the intestine - in particular that arising through ICC-SMC interactions.

We appreciate these advices. In the revised MS, we are careful so that our findings do not necessarily portray the physiological functions in adult gut.

Minor:

More technical information could be used in the methods:

(1) What concentration of Matrigel is used for coating, and what size were the wells that cells were deposited into?

We have added, “14-mm diameter glass-bottom dishes (Matsunami, D11130H)” and “undiluted Matrigel (Corning, 354248) at 38.5°C for 20 min” (line 471473).

(2) How were organoids transferred to the hydrogels? And were the hydrogels coated?

We have added “Organoids were transferred to the hydrogel using a glass capillary” (line 560-561).

(3) Tests for significance and p values should be added where appropriate (e.g. Figure S3B).

We have added these in Figure legend of new Fig. S3.

Reviewer #3 (Recommendations For The Authors):

This is an exciting study, and while the majority of our comments are minor suggestions to improve the clarity and impact of findings, it would be important to verify the effective disruption of GAP junction function with CBX or 18Beta-GA treatments before concluding they are not required for coordination of contractility and initiation by ICCs. It is possible that sufficient contextual support exists in the literature for the nature of treatments used, but this may need to be conveyed within the manuscript to allay concerns that the results could be explained by ineffective inhibition of GAP junctions.

Thank you very much for these advices. In the revised version, we have newly carried out experiments with dissociated embryonic heart cells cultured in vitro, a model widely used for gap junction studies (Fig. S3D). Both CBX or 18b-GA exert efficient inhibiting activity on contractions of heart cells. We have added the following sentence, “The inhibiting activity of the drugs used here was verified using embryonic heart culture (line 237-239)”.

eLife Assessment

The study presents a comprehensive multi-approach and functional investigation of RBMX2 as a host factor involved in Mycobacterium bovis pathogenesis and its potential role in promoting epithelial-mesenchymal transition and lung cancer progression. The findings are valuable since the possible connection between M. bovis and lung cancer and the underlying mechanisms provides a promising direction for future research. The evidence is solid with methods, data, and analyses broadly supporting the claims, albeit with minor weaknesses that, if addressed, will make the evidence stronger. The study remains of great interest to microbiology, oncology, and drug discovery scientists.

Reviewer #1 (Public review):

Summary:

This manuscript presents a compelling study identifying RBMX2 as a novel host factor upregulated during Mycobacterium bovis infection.

The study demonstrates that RBMX2 plays a role in:

(1) Facilitating M. bovis adhesion, invasion, and survival in epithelial cells.

(2) Disrupting tight junctions and promoting EMT.

(3) Contributing to inflammatory responses and possibly predisposing infected tissue to lung cancer development.

By using a combination of CRISPR-Cas9 library screening, multi-omics, coculture models, and bioinformatics, the authors establish a detailed mechanistic link between M. bovis infection and cancer-related EMT through the p65/MMP-9 signaling axis. Identification of RBMX2 as a bridge between TB infection and EMT is novel.

Strengths:

This topic and data are both novel and significant, expanding the understanding of transcriptomic diversity beyond RBM2 in M. bovis responsive functions.

Weaknesses:

(1) The abstract and introduction sometimes suggest RBMX2 has protective anti-TB functions, yet results show it facilitates pathogen adhesion and survival. The authors need to rephrase claims to avoid contradiction.

(2) While p65/MMP-9 is convincingly implicated, the role of MAPK/p38 and JNK is less clearly resolved.

(3) Metabolomics results are interesting but not integrated deeply into the main EMT narrative.

(4) A key finding and starting point of this study is the upregulation of RBMX2 upon M. bovis infection. However, the authors have only assessed RBMX2 expression at the mRNA level following infection with M. bovis and BCG. To strengthen this conclusion, it is essential to validate RBMX2 expression at the protein level through techniques such as Western blotting or immunofluorescence. This would significantly enhance the credibility and impact of the study's foundational observation.

(5) The manuscript would benefit from a more in-depth discussion of the relationship between tuberculosis (TB) and lung cancer. While the study provides experimental evidence suggesting a link via EMT induction, integrating current literature on the epidemiological and mechanistic connections between chronic TB infection and lung tumorigenesis would provide important context and reinforce the translational relevance of the findings.

Author response:

Public Reviews:

Reviewer #1 (Public review):

Summary:

This manuscript presents a compelling study identifying RBMX2 as a novel host factor upregulated during Mycobacterium bovis infection.

The study demonstrates that RBMX2 plays a role in:

(1) Facilitating M. bovis adhesion, invasion, and survival in epithelial cells.

(2) Disrupting tight junctions and promoting EMT.

(3) Contributing to inflammatory responses and possibly predisposing infected tissue to lung cancer development.

By using a combination of CRISPR-Cas9 library screening, multi-omics, coculture models, and bioinformatics, the authors establish a detailed mechanistic link between M. bovis infection and cancer-related EMT through the p65/MMP-9 signaling axis. Identification of RBMX2 as a bridge between TB infection and EMT is novel.

Strengths:

This topic and data are both novel and significant, expanding the understanding of transcriptomic diversity beyond RBM2 in M. bovis responsive functions.

Weaknesses:

(1) The abstract and introduction sometimes suggest RBMX2 has protective anti-TB functions, yet results show it facilitates pathogen adhesion and survival. The authors need to rephrase claims to avoid contradiction.

We sincerely appreciate the reviewer's valuable feedback regarding the need to clarify RBMX2's role throughout the manuscript. We have carefully revised the text to ensure consistent messaging about RBMX2's function in promoting M. bovis infection. Below we detail the specific modifications made:

(1) Introduction Revisions:

Changed "The objective of this study was to elucidate the correlation between host genes and the susceptibility of M.bovis infection" to "The objective of this study was to identify host factors that promote susceptibility to M.bovis infection"

Revised "RBMX2 polyclonal and monoclonal cell lines exhibited favorable phenotypes" to "RBMX2 knockout cell lines showed reduced bacterial survival"

Replaced "The immune regulatory mechanism of RBMX2" with "The role of RBMX2 in facilitating M.bovis immune evasion"

(2) Results Revisions:

Modified "RBMX2 fails to affect cell morphology and the ability to proliferate and promotes M.bovis infection" to "RBMX2 does not alter cell viability but significantly enhances M.bovis infection"

Strengthened conclusion in Figure 4: "RBMX2 actively disrupts tight junctions to facilitate bacterial invasion"

(3) Discussion Revisions:

Revised screening description: "We screened host factors affecting M.bovis susceptibility and identified RBMX2 as a key promoter of infection"

Strengthened concluding statement: "In summary, RBMX2 drives TB pathogenesis by compromising epithelial barriers and inducing EMT"

These targeted revisions ensure that:

All sections consistently present RBMX2 as promoting infection; the language aligns with our experimental finding; potential protective interpretations have been eliminated. We believe these modifications have successfully addressed the reviewer's concern while maintaining the manuscript's original structure and scientific content. We appreciate the opportunity to improve our manuscript and thank the reviewer for this constructive suggestion.

(2) >While p65/MMP-9 is convincingly implicated, the role of MAPK/p38 and JNK is less clearly resolved.

We sincerely appreciate the reviewer's insightful comment regarding the roles of MAPK/p38 and JNK in our study. Our experimental data clearly demonstrated that RBMX2 knockout significantly reduced phosphorylation levels of p65, p38, and JNK (Fig. 5A), indicating potential involvement of all three pathways in RBMX2-mediated regulation.

Through systematic functional validation, we obtained several important findings:

In pathway inhibition experiments, p65 activation (PMA treatment) showed the most dramatic effects on both tight junction disruption (ZO-1, OCLN reduction) and EMT marker regulation (E-cadherin downregulation, N-cadherin upregulation);

p38 activation (ML141 treatment) exhibited moderate effects on these processes;

JNK activation (Anisomycin treatment) displayed minimal impact.

Most conclusively, siRNA-mediated silencing of p65 alone was sufficient to:

Restore epithelial barrier function

Reverse EMT marker expression

Reduce bacterial adhesion and invasion

These results establish a clear hierarchy in pathway importance: p65 serves as the primary mediator of RBMX2's effects, while p38 plays a secondary role and JNK appears non-essential under our experimental conditions. We have now clarified this relationship in the revised Discussion section to strengthen this conclusion.

This refined understanding of pathway hierarchy provides important mechanistic insights while maintaining consistency with all our experimental data. We thank the reviewer for this valuable suggestion that helped improve our manuscript.

(3) Metabolomics results are interesting but not integrated deeply into the main EMT narrative.

Thank you for this constructive suggestion. In this article, we detected the metabolome of RBMX2 knockout and wild-type cells after Mycobacterium bovis infection, which mainly served as supporting evidence for our EMT model. However, we did not conduct an in-depth discussion of these findings. We have now added a detailed discussion of this section to further support our EMT model.

ADD:Meanwhile, metabolic pathways enriched after RBMX2 deletion, such as nucleotide metabolism, nucleotide sugar synthesis, and pentose interconversion, primarily support cell proliferation and migration during EMT by providing energy precursors, regulating glycosylation modifications, and maintaining redox balance; cofactor synthesis and amino sugar metabolism participate in EMT regulation through influencing metabolic remodeling and extracellular matrix interactions; chemokine and cGMP-PKG signaling pathways may further mediate inflammatory responses and cytoskeletal rearrangements, collectively promoting the EMT process.

(4) A key finding and starting point of this study is the upregulation of RBMX2 upon M. bovis infection. However, the authors have only assessed RBMX2 expression at the mRNA level following infection with M. bovis and BCG. To strengthen this conclusion, it is essential to validate RBMX2 expression at the protein level through techniques such as Western blotting or immunofluorescence. This would significantly enhance the credibility and impact of the study's foundational observation.

Thank you for your comment. We have supplemented the experiments in this part and found that Mycobacterium bovis infection can significantly enhance the expression level of RBMX2 protein.

(5) The manuscript would benefit from a more in-depth discussion of the relationship between tuberculosis (TB) and lung cancer. While the study provides experimental evidence suggesting a link via EMT induction, integrating current literature on the epidemiological and mechanistic connections between chronic TB infection and lung tumorigenesis would provide important context and reinforce the translational relevance of the findings.

We sincerely appreciate the valuable comments from the reviewer. We fully agree with your suggestion to further explore the relationship between tuberculosis (TB) and lung cancer. In the revised manuscript, we will add a new paragraph in the Discussion section to systematically integrate the current literature on the epidemiological and mechanistic links between chronic tuberculosis infection and lung cancer development, including the potential bridging roles of chronic inflammation, tissue damage repair, immune microenvironment remodeling, and the epithelial-mesenchymal transition (EMT) pathway. This addition will help more comprehensively interpret the clinical implications of the observed EMT activation in the context of our study, thereby enhancing the biological plausibility and clinical translational value of our findings.

ADD:There is growing epidemiological evidence suggesting that chronic TB infection represents a potential risk factor for the development of lung cancer. Studies have shown that individuals with a history of TB exhibit a significantly increased risk of lung cancer, particularly in areas of the lung with pre-existing fibrotic scars, indicating that chronic inflammation, tissue repair, and immune microenvironment remodeling may collectively contribute to malignant transformation 74. Moreover, EMT not only endows epithelial cells with mesenchymal features that enhance migratory and invasive capacity but is also associated with the acquisition of cancer stem cell-like properties and therapeutic resistance 75. Therefore, EMT may serve as a crucial molecular link connecting chronic TB infection with the malignant transformation of lung epithelial cells, warranting further investigation in the intersection of infection and tumorigenesis.

Reviewer #2 (Public review):

Summary:

I am not familiar with cancer biology, so my review mainly focuses on the infection part of the manuscript. Wang et al identified an RNA-binding protein RBMX2 that links the Mycobacterium bovis infection to the epithelial-Mesenchymal transition and lung cancer progression. Upon mycobacterium infection, the expression of RBMX2 was moderately increased in multiple bovine and human cell lines, as well as bovine lung and liver tissues. Using global approaches, including RNA-seq and proteomics, the authors identified differential gene expression caused by the RBMX2 knockout during M. bovis infection. Knockout of RBMX2 led to significant upregulations of tight-junction related genes such as CLDN-5, OCLN, ZO-1, whereas M. bovis infection affects the integrity of epithelial cell tight junctions and inflammatory responses. This study establishes that RBMX2 is an important host factor that modulates the infection process of M. bovis.

Strengths:

(1) This study tested multiple types of bovine and human cells, including macrophages, epithelial cells, and clinical tissues at multiple timepoints, and firmly confirmed the induced expression of RBMX2 upon M. bovis infection.

(2) The authors have generated the monoclonal RBMX2 knockout cell lines and comprehensively characterized the RBMX2-dependent gene expression changes using a combination of global omics approaches. The study has validated the impact of RBMX2 knockout on the tight-junction pathway and on the M. bovis infection, establishing RBMX2 as a crucial host factor.

Weaknesses:

(1) The RBMX2 was only moderately induced (less than 2-fold) upon M. bovis infection, arguing its contribution may be small. Its value as a therapeutic target is not justified. How RBMX2 was activated by M. bovis infection was unclear.

Thank you for your valuable and constructive comments. In this study, we primarily utilized the CRISPR whole-genome screening approach to identify key factors involved in bovine tuberculosis infection. Through four rounds of screening using a whole-genome knockout cell line of bovine lung epithelial cells infected with Mycobacterium bovis, we identified RBMX2 as a critical factor.

Although the transcriptional level change of RBMX2 was less than two-fold, following the suggestion of Reviewer 1, we examined its expression at the protein level, where the change was more pronounced, and we have added these results to the manuscript.

Regarding the mechanism by which RBMX2 is activated upon M. bovis infection, we previously screened for interacting proteins using a Mycobacterium tuberculosis secreted and membrane protein library, but unfortunately, we did not identify any direct interacting proteins from M. tuberculosis (https://doi.org/10.1093/nar/gkx1173).

(2) Although multiple time points have been included in the study, most analyses lack temporal resolution. It is difficult to appreciate the impact/consequence of M. bovis infection on the analyzed pathways and processes.

We appreciate the valuable comments from the reviewers. Although our study included multiple time points post-infection, in our experimental design we focused on different biological processes and phenotypes at distinct time points:

During the early phase (e.g., 2 hours post-infection), we focused on barrier phenotypes; during the intermediate phase (e.g., 24 hours post-infection), we concentrated more on pathway activation and EMT phenotypes;

And during the later phase (e.g., 48–72 hours post-infection), we focused more on cell death phenotypes, which were validated in another FII article (https://doi.org/10.3389/fimmu.2024.1431207).

We also examined the impact of varying infection durations on RBMX2 knockout EBL cellular lines via GO analysis. At 0 hpi, genes were primarily related to the pathways of cell junctions, extracellular regions, and cell junction organization. At 24 hpi, genes were mainly associated with pathways of the basement membrane, cell adhesion, integrin binding and cell migration By 48 hpi, genes were annotated into epithelial cell differentiation and were negatively regulated during epithelial cell proliferation. This indicated that RBMX2 can regulate cellular connectivity throughout the stages of M. bovis infection.

For KEGG analysis, genes linked to the MAPK signaling pathway, chemical carcinogen-DNA adducts, and chemical carcinogen-receptor activation were observed at 0 hpi. At 24 hpi, significant enrichment was found in the ECM-receptor interaction, PI3K-Akt signaling pathway, and focal adhesion. Upon enrichment analysis at 48 hpi, significant enrichment was noted in the TGF-beta signaling pathway, transcriptional misregulation in cancer, microRNAs in cancer, small cell lung cancer, and p53 signaling pathway.

Reviewer #3 (Public review):

Summary:

This study investigates the role of the host protein RBMX2 in regulating the response to Mycobacterium bovis infection and its connection to epithelial-mesenchymal transition (EMT), a key pathway in cancer progression. Using bovine and human cell models, the authors have wisely shown that RBMX2 expression is upregulated following M. bovis infection and promotes bacterial adhesion, invasion, and survival by disrupting epithelial tight junctions via the p65/MMP-9 signaling pathway. They also demonstrate that RBMX2 facilitates EMT and is overexpressed in human lung cancers, suggesting a potential link between chronic infection and tumor progression. The study highlights RBMX2 as a novel host factor that could serve as a therapeutic target for both TB pathogenesis and infection-related cancer risk.

Strengths:

The major strengths lie in its multi-omics integration (transcriptomics, proteomics, metabolomics) to map RBMX2's impact on host pathways, combined with rigorous functional assays (knockout/knockdown, adhesion/invasion, barrier tests) that establish causality through the p65/MMP-9 axis. Validation across bovine and human cell models and in clinical tissue samples enhances translational relevance. Finally, identifying RBMX2 as a novel regulator linking mycobacterial infection to EMT and cancer progression opens exciting therapeutic avenues.

Weaknesses:

Although it's a solid study, there are a few weaknesses noted below.

(1) In the transcriptomics analysis, the authors performed (GO/KEGG) to explore biological functions. Did they perform the search locally or globally? If the search was performed with a global reference, then I would recommend doing a local search. That would give more relevant results. What is the logic behind highlighting some of the enriched pathways (in red), and how are they relevant to the current study?

We appreciate the reviewer's thoughtful questions regarding our transcriptomic analysis. In this study, we employed a localized enrichment approach focusing specifically on gene expression profiles from our bovine lung epithelial cell system. This cell-type-specific analysis provides more biologically relevant results than global database searches alone.

Regarding the highlighted pathways, these represent:

(1) Temporally significant pathways showing strongest enrichment at each stage:

• 0h: Cell junction organization (immediate barrier response)

• 24h: ECM-receptor interaction (early EMT initiation)

• 48h: TGF-β signaling (chronic remodeling)

(2) Mechanistically linked to our core findings about RBMX2's role in:

• Epithelial barrier disruption

• Mesenchymal transition

• Chronic infection outcomes

We selected these particular pathways because they:

(1) Showed the most statistically significant changes (FDR <0.001)

(2) Formed a coherent biological narrative across infection stages

(3) Were independently validated in our functional assays

This targeted approach allows us to focus on the most infection-relevant pathways while maintaining statistical rigor.

(2) While the authors show that RBMX2 expression correlates with EMT-related gene expression and barrier dysfunction, the evidence for direct association remains limited in this study. How does RBMX2 activate p65? Does it bind directly to p65 or modulate any upstream kinases? Could ChIP-seq or CLIP-seq provide further evidence for direct RNA or DNA targets of RBMX2 that drive EMT or NF-κB signaling?

We sincerely appreciate the reviewer's in-depth questions regarding the mechanisms by which RBMX2 activates p65 and its association with EMT. Although the molecular mechanism remains to be fully elucidated, our study has provided experimental evidence supporting a direct regulatory relationship between RBMX2 and the p65 subunit of the NF-κB pathway. Specifically, we investigated whether the transcription factor p65 could directly bind to the promoter region of RBMX2 using CHIP experiments. The results demonstrated that the transcription factor p65 can physically bind to the RBMX2 region.

Furthermore, dual-luciferase reporter assays were conducted, showing that p65 significantly enhances the transcriptional activity of the RBMX2 promoter, indicating a direct regulatory effect of RBMX2 on p65 expression.

These findings support our hypothesis that RBMX2 activates the NF-κB signaling pathway through direct interaction with the p65 protein, thereby participating in the regulation of EMT progression and barrier function.

In our subsequent work papers, we will also employ experiments such as CLIP to further investigate the specific mechanisms through which RBMX2 exerts its regulatory functions.

(3) The manuscript suggests that RBMX2 enhances adhesion/invasion of several bacterial species (e.g., E. coli, Salmonella), not just M. bovis. This raises questions about the specificity of RBMX2's role in Mycobacterium-specific pathogenesis. Is RBMX2 a general epithelial barrier regulator or does it exhibit preferential effects in mycobacterial infection contexts? How does this generality affect its potential as a TB-specific therapeutic target?

Thank you for your valuable comments. When we initially designed this experiment, we were interested in whether the RBMX2 knockout cell line could confer effective resistance not only against Mycobacterium bovis but also against Gram-negative and Gram-positive bacteria. Surprisingly, we indeed observed resistance to the invasion of these pathogens, albeit weaker compared to that against Mycobacterium bovis.

Nevertheless, we believe these findings merit publication in eLife. Moreover, RBMX2 knockout does not affect the phenotype of epithelial barrier disruption under normal conditions; its significant regulatory effect on barrier function is only evident upon infection with Mycobacterium bovis.

Importantly, during our genome-wide knockout library screening, RBMX2 was not identified in the screening models for Salmonella or Escherichia coli, but was consistently detected across multiple rounds of screening in the Mycobacterium bovis model.

(4) The quality of the figures is very poor. High-resolution images should be provided.

Thank you for your feedback; we provided higher-resolution images.

(5) The methods are not very descriptive, particularly the omics section.

Thank you for your comments; we have revised the description of the sequencing section.

(6) The manuscript is too dense, with extensive multi-omics data (transcriptomics, proteomics, metabolomics) but relatively little mechanistic integration. The authors should have focused on the key mechanistic pathways in the figures. Improving the narratives in the Results and Discussion section could help readers follow the logic of the experimental design and conclusions.

Thank you for your valuable comments. We have streamlined the figures and revised the description of the results section accordingly.

Reviewer #2 (Public review):

Summary:

I am not familiar with cancer biology, so my review mainly focuses on the infection part of the manuscript. Wang et al identified an RNA-binding protein RBMX2 that links the Mycobacterium bovis infection to the epithelial-Mesenchymal transition and lung cancer progression. Upon mycobacterium infection, the expression of RBMX2 was moderately increased in multiple bovine and human cell lines, as well as bovine lung and liver tissues. Using global approaches, including RNA-seq and proteomics, the authors identified differential gene expression caused by the RBMX2 knockout during M. bovis infection. Knockout of RBMX2 led to significant upregulations of tight-junction related genes such as CLDN-5, OCLN, ZO-1, whereas M. bovis infection affects the integrity of epithelial cell tight junctions and inflammatory responses. This study establishes that RBMX2 is an important host factor that modulates the infection process of M. bovis.

Strengths:

(1) This study tested multiple types of bovine and human cells, including macrophages, epithelial cells, and clinical tissues at multiple timepoints, and firmly confirmed the induced expression of RBMX2 upon M. bovis infection.

(2) The authors have generated the monoclonal RBMX2 knockout cell lines and comprehensively characterized the RBMX2-dependent gene expression changes using a combination of global omics approaches. The study has validated the impact of RBMX2 knockout on the tight-junction pathway and on the M. bovis infection, establishing RBMX2 as a crucial host factor.

Weaknesses:

(1) The RBMX2 was only moderately induced (less than 2-fold) upon M. bovis infection, arguing its contribution may be small. Its value as a therapeutic target is not justified. How RBMX2 was activated by M. bovis infection was unclear.

(2) Although multiple time points have been included in the study, most analyses lack temporal resolution. It is difficult to appreciate the impact/consequence of M. bovis infection on the analyzed pathways and processes.

Reviewer #3 (Public review):

Summary:

This study investigates the role of the host protein RBMX2 in regulating the response to Mycobacterium bovis infection and its connection to epithelial-mesenchymal transition (EMT), a key pathway in cancer progression. Using bovine and human cell models, the authors have wisely shown that RBMX2 expression is upregulated following M. bovis infection and promotes bacterial adhesion, invasion, and survival by disrupting epithelial tight junctions via the p65/MMP-9 signaling pathway. They also demonstrate that RBMX2 facilitates EMT and is overexpressed in human lung cancers, suggesting a potential link between chronic infection and tumor progression. The study highlights RBMX2 as a novel host factor that could serve as a therapeutic target for both TB pathogenesis and infection-related cancer risk.

Strengths:

The major strengths lie in its multi-omics integration (transcriptomics, proteomics, metabolomics) to map RBMX2's impact on host pathways, combined with rigorous functional assays (knockout/knockdown, adhesion/invasion, barrier tests) that establish causality through the p65/MMP-9 axis. Validation across bovine and human cell models and in clinical tissue samples enhances translational relevance. Finally, identifying RBMX2 as a novel regulator linking mycobacterial infection to EMT and cancer progression opens exciting therapeutic avenues.

Weaknesses:

Although it's a solid study, there are a few weaknesses noted below.

(1) In the transcriptomics analysis, the authors performed (GO/KEGG) to explore biological functions. Did they perform the search locally or globally? If the search was performed with a global reference, then I would recommend doing a local search. That would give more relevant results. What is the logic behind highlighting some of the enriched pathways (in red), and how are they relevant to the current study?

(2) While the authors show that RBMX2 expression correlates with EMT-related gene expression and barrier dysfunction, the evidence for direct association remains limited in this study. How does RBMX2 activate p65? Does it bind directly to p65 or modulate any upstream kinases? Could ChIP-seq or CLIP-seq provide further evidence for direct RNA or DNA targets of RBMX2 that drive EMT or NF-κB signaling?

(3) The manuscript suggests that RBMX2 enhances adhesion/invasion of several bacterial species (e.g., E. coli, Salmonella), not just M. bovis. This raises questions about the specificity of RBMX2's role in Mycobacterium-specific pathogenesis. Is RBMX2 a general epithelial barrier regulator or does it exhibit preferential effects in mycobacterial infection contexts? How does this generality affect its potential as a TB-specific therapeutic target?

(4) The quality of the figures is very poor. High-resolution images should be provided.

(5) The methods are not very descriptive, particularly the omics section.

(6) The manuscript is too dense, with extensive multi-omics data (transcriptomics, proteomics, metabolomics) but relatively little mechanistic integration. The authors should have focused on the key mechanistic pathways in the figures. Improving the narratives in the Results and Discussion section could help readers follow the logic of the experimental design and conclusions.

eLife Assessment

This work describes an inference technique for extracting information about relative contributions of excitatory and inhibitory synaptic drive onto single neurons in neural networks. The electrophysiological techniques and results are of high quality, and the analytical work is novel and potentially powerful, yet with several untested assumptions underlying the approach. This is nevertheless solid work that will be valuable to neuroscience labs interested in exploring alternative approaches to studies of integrated synaptic connectivity.

Reviewer #2 (Public review):

Summary:

By measuring intracellular changes in membrane voltage from a single neuron of the medulla the authors attempted to develop a method for determining the balance of excitatory and inhibitory synaptic drive onto a single neuron.

Strengths:

This data-driven approach to explore neural circuits is described well in this study and could be valuable in identifying microcircuits that generate rhythms. Importantly, perhaps, this inference method could enable microcircuits to be studied without the need for time-consuming anatomical tracing or other more involved electrophysiological techniques. Therefore, I can see the value in developing an approach of this type.

Weaknesses:

The implications of several assumptions associated with this inference technique have been considered by the authors.

Most importantly, it is my understanding that this approach assumes a linear I-V when extracting information about the excitatory and inhibitory synaptic conductances (see equations 6 and 7). In Figure 6, the authors explore the impact of varying the reversal potential for the extraction of information about synaptic drive, but this still assumes that the underlying conductance is linear. However, open rectification will be a feature of any conductance generated by asymmetric distributions of ions (see the GHK current equation) and will therefore be a particular issue for the inhibition resulting from asymmetrical Cl- ion gradients across GABA-A receptors as well as the K+ conductance indirectly activated by GABA-B receptor activation. The mixed cation conductance that underlies most synaptic excitation will also generate a non-linear I-V relationship due to the inward rectification associated with polyamine block of AMPA receptors. The authors present evidence that the I-V relationship is linear over most of the voltage range examined, and this is a helpful addition. The authors have discussed the absence of active conductances contributing to the I-V, but I still wonder how the extraction of information concerning the excitatory and inhibitory conductances relies on the assumption of a linear I-V for these conductances.

This approach has similarities to earlier studies undertaken in the visual cortex that estimated the excitatory and inhibitory synaptic conductance changes that contributed to membrane voltage changes during receptive field stimulation. However, these approaches also involved the recording of transmembrane current changes during visual stimulation that were undertaken in voltage-clamp at various command voltages to estimate the underlying conductance changes. Molkov et al have attempted to essentially deconvolve the underlying conductance changes without this information and I am concerned that this simply may not be possible. However, I appreciate the efforts taken by the authors to address this issue.

The current balance equation (1) cited in this study is based upon the parallel conductance model developed by Hodgkin & Huxley. One key element of the HH equations is the inclusion of an estimate of the capacitive current generated due to the change in voltage across the membrane capacitance. While the present study considers the impact of membrane capacitance, a deeper discussion on how variations in capacitance across different neuron types might affect inference accuracy would be useful. Differences in capacitance could introduce variability in inferred conductances, potentially influencing model predictions.

Studies using acute slicing preparations to examine circuit effects have often been limited to the study of small microcircuits, especially feedforward and feedback interneuron circuits. It is widely accepted that any information gained from this approach will always be compromised by the absence of patterned afferent input from outside the brain region being studied. In this study, descending control from the Pons and the neocortex will not be contributing much to the synaptic drive and ascending information from respiratory muscles will also be absent completely. This may not have been such a major concern if this study had been limited to demonstrating the feasibility of a methodological approach. However, this limitation does need to be considered when using an approach of this type to speculate on the prevalence of specific circuit motifs within the medulla (Figure 4). Therefore, I would argue that some discussion of this limitation should be included in this manuscript.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public review):

Summary:

The study aims to create a comprehensive repository about the changes in protein abundance and their modification during oocyte maturation in Xenopus laevis.

Strengths:

The results contribute meaningfully to the field.

Weaknesses:

The manuscript could have benefitted from more comprehensive analyses and clearer writing. Nonetheless, the key findings are robust and offer a valuable resource for the scientific community.

We would like to thank the reviewer for his/her positive feedback on our article. The public review points out that "The manuscript could have benefitted from more comprehensive analyses and clearer writing." We have rewritten several sections and provided more detailed explanations of the analysis and interpretation of some data (see below for details). We have also followed all of the reviewer's recommendations, some of which specifically highlighted areas lacking clarity. We would also like to thank the reviewer for pointing out some errors, for which we apologize, and which have now been corrected. We sincerely appreciate the reviewer's thorough work, as it has greatly enhanced the clarity and precision of the manuscript.

Reviewer #2 (Public review):

Summary:

The authors analyzed Xenopus oocytes at different stages of meiosis using quantitative phosphoproteomics. Their advanced methods and analyses revealed changes in protein abundances and phosphorylation states to an unprecedented depth and quantitative detail. In the manuscript they provide an excellent interpretation of these findings putting them in the context of past literature in Xenopus as well as in other model systems.

Strengths:

High quality data, careful and detailed analysis, outstanding interpretation in the context of the large body of the literature.

Weaknesses:

Merely a resource, none of the findings are tested in functional experiments.

I am very impressed by the quality of the data and the careful and detailed interpretation of the findings. In this form the manuscript will be an excellent resource to the cell division community in general, and it presents a very large number of hypotheses that can be tested in future experiments. Xenopus has been and still is a popular and powerful model system that led to critical discoveries around countless cellular processes, including the spindle, nuclear envelope, translational regulation, just to name a few. This also includes a huge body of literature on the cell cycle describing its phosphoregulation. It is indeed somewhat frustrating to see that these earlier studies using phosphomutants and phospho-antibodies were just scratching the surface. The phosphoproteomics analysis presented here reveals much more extensive and much more dynamic changes in phosphorylation states. Thereby, in my opinion, this manuscript opens a completely new chapter in this line of research, setting the stage for more systematic future studies.

We thank the reviewer for his/her extremely positive comments. The public review points out that "none of the findings are tested in functional experiments." This is entirely accurate. We focused our work on obtaining the highest quality proteomic and phosphoproteomic data possible, and then sought to highlight these data by connecting them with existing functional data from the literature. This approach has opened up research avenues with enormous, previously unforeseen potential, in a wide range of biological fields (cell cycle, meiosis, oogenesis, embryonic development, cell biology, cellular physiology, signaling, evolution, etc.). We chose not to delay publication by experimentally investigating the narrow area in which we are specialists (meiotic maturation), while our data offer a vast array of research opportunities across various fields. Our goal was, therefore, to present this extensive dataset as a resource for different scientific communities, who can explore their specific biological questions using our data. This is why we submitted our article to the "Repository" section of eLife. Nevertheless, in the context of the comparative analysis of the mouse and Xenopus phosphoproteomes performed at the reviewer’s request, we felt it was important to complement this new section with functional experiments that not only validate the proteomic data but also provide new insights into certain proteins and their regulation by Cdk1 (new paragraph lines 824-860 and new Figure 9).

We are also grateful to the reviewer for the recommendation to improve the manuscript by including more comparisons between our Xenopus data and those from other systems. We have followed this suggestion (see below), which has significantly enriched the article (new paragraph lines 824-860 and new Figure 9).

Reviewer #3 (Public review):

Summary:

The authors performed time-resolved proteomics and phospho-proteomics in Xenopus oocytes from prophase I through the MII arrest of the unfertilized egg. The data contains protein abundance and phosphorylation sites of a large number set of proteins at different stages of oocyte maturation. The large sets of the data are of high quality. In addition, the authors discussed several key pathways critical for the maturation. The data is very useful for the researchers not only researchers in Xenopus oocytes but also those in oocyte biology in other organisms.

Strengths:

The data of proteomics and phospho-proteomics in Xenopus oocyte maturation is very useful for future studies to understand molecular networks in oocyte maturation.

Weaknesses:

Although the authors offered molecular pathways of the phosphorylation in the translation, protein degradation, cell cycle regulation, and chromosome segregation. The author did not check the validity of the molecular pathways based on their proteomic data by the experimentation.

We thank the reviewer for his/her positive comments. The public review points out that "The author did not check the validity of the molecular pathways based on their proteomic data by the experimentation." This is entirely accurate. We focused our work on obtaining the highest quality proteomic and phosphoproteomic data possible, and then sought to highlight these data by connecting them with existing functional data from the literature. This approach has opened up research avenues with enormous, previously unforeseen potential, in a wide range of biological fields (cell cycle, meiosis, oogenesis, embryonic development, cell biology, cellular physiology, signaling, evolution, etc.). We chose not to delay publication by experimentally investigating the very narrow area in which we are specialists (meiotic maturation), while our data offer a vast array of research opportunities across various fields. Our goal was, therefore, to present this extensive dataset as a resource for different scientific communities, who can explore their specific biological questions using our data. This is why we submitted our article to the "Repository" section of eLife. Nevertheless, in the context of the comparative analysis of the mouse and Xenopus phosphoproteomes performed at the reviewer’s request, we felt it was important to complement this new section with functional experiments that not only validate the proteomic data but also provide new insights into certain proteins and their regulation by Cdk1 (new paragraph lines 824-860 and new Figure 9).

We have also followed all of the reviewer's recommendations and thank him/her, as the suggestions have significantly enhanced the manuscript.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

(1) Fig. 1 -> In the Figure legend "mPRβ" is called "mPRb". In the Figure, it is indicated that PKA substrates are always activated by the phosphorylation. As the relevant substrates and the mode-of-action of the Arpp19 phosphorylation are not clear at the moment, this seems to be preliminary. It could for example also be conceivable that PKA phosphorylation inhibits a translation activator. In addition, the PG-dependent translation of RINGO/Speedy should be included in the model.

We fully agree with the reviewer. PKA substrates can either be activators of the Cdk1 activation pathway, which are inhibited by phosphorylation by PKA, or repressors of the same pathway, which are activated by phosphorylation by PKA. This is now illustrated in the new Fig. 1. In addition, we have also included RINGO/Speedy in the model and in the text (lines 78-79) and corrected "mPRb" in the legend.

(2) Lane 51-52 -> it is questionable if the meiotic divisions can be called "embryonic processes"

We agree with the reviewer comment, and we have removed the word “embryonic”.

(3) Lane 53 and lane 106-107 -> recent data have indicated that transcription already starts during cell cycle 12 and 13 in most cells (e.g. Blitz and Cho: Control of zygotic genome activation in Xenopus (2021))

We apologize for this mistake. The text has been corrected and the reference added (lines 53 and 107).

(4) Lane 61-62 -> "MI" and "MII" are given as abbreviation for "first and second meiotic spindle"

The text has been clarified to explain that MI is referred to metaphase I and MII stands for metaphase II (lines 61-64).

(%) Lane 131-132 -> "single-cell" is mentioned redundantly in this sentence.

The sentence has been corrected (lines 131-132).

(6) Fig. 2B -> it is not explained what is plotted as "Average levels" on the x-Axis. Is it the average of expression over all samples or at a given time point? Are the values given as a concentration or are the values normalized? If so, how were they normalized?

We agree with the reviewer comment that “Average levels” may have been unclear. In the new Fig. 2B, we have re-plotted the graph using the average protein concentration during meiosis, measured as described in the Methods section.

(7) In Fig. 2-supplement 3E -> from the descriptions it is not entirely clear to me what the difference to the data in Fig. 2B is?

We thank the reviewer for his/her question regarding the relationship between the data in Fig. 2B and Fig. 2-supplement 3E. We confirm that the raw data visualized in Fig. 2-supplement 3E are the same as those in Fig. 2B. However, in Fig. 2-supplement 3E, the data are color-coded differently to highlight the number of proteins whose concentrations change during meiotic divisions, based on the threshold adopted. The legend of Fig. 2-supplement 3E has been modified to clarify this point.

(8) Lane 225-226 -> Kifc1 is a minus-end directed motor

This mistake has been corrected (lines 232-233).

(9) Lane 271 -> Serbp1, here mentioned to be involved in stabilization of mRNAs, has also been implicated in the regulation of ribosomes (e.g. Leesch et al. 2023). Regarding the overall topic of this manuscript, this could be mentioned as well.

We agree with the referee that the important role of Serbp1 in the control of ribosome hibernation needs to be mentioned. We have included this point in the revised manuscript together with the reference (lines 277-279).

(10) Lane 360-363 -> it is mentioned that APPL1 and Akt2 act "to induce meiosis". Furthermore, in the Nader et al. 2020 paper, Akt2 phosphorylation is reported to happen within 30min after PG treatment. In the present work, they only seem to get phosphorylated when Cdk1 is activated. Is there an explanation for this discrepancy?

Indeed, Nader et al. (2020) indicate that Akt2 is phosphorylated on Ser473 (actually, they should have mentioned Ser474, which is the phosphorylated residue on Akt2; Ser473 corresponds to the numbering of Akt1) between 5 and 30 minutes post-Pg, which supports their hypothesis of an early role for this kinase. However, these conclusions should be taken with caution, considering that their functional experiment using antisense against Akt2 depletes only 25% of the protein, the antibody used to visualize Akt2 phosphorylation also recognizes phosphorylated Akt1 and Akt3, and they did not analyze phosphorylation of the protein after 30 minutes. Therefore, we cannot determine whether the level observed at 30 minutes represents a maximum or if it is just the onset of the phosphorylation that peaks later, possibly after activation of Cdk1, for example.

Regarding our measurements: we clearly observe phosphorylation of Akt2 following Cdk1 activation on Ser131. We did not detect Akt2 phosphorylation on Ser474, but since our measurements started 1 hour post-Pg, this protein may have returned to a dephosphorylated state on Ser474.

Therefore, the observations of Nader et al. and ours involve different residues and different phosphorylation kinetics, Nader et al. limiting their analysis to the first 30 minutes, whereas we started at 1 hour.

We have revised the manuscript text to make these aspects clearer (lines 387-392).

(11) Fig. 3B -> it could be made clearer in the Figure that all these sites belong to class I

A title “Class I proteins” has been added in Fig. 3B to clarify it.

(12) Lane 433-434 -> the authors write that the proteomic data of this study confirm that PATL1 is accumulating during meiotic maturation. However, in Fig. 2B PATL1 is not among the significantly enriched proteins.

We apologize for this error. Indeed, PATL1 protein is not significantly enriched. The text has been corrected (lines 461-465).

(13) Fig. 4B -> Zar2 is color-coded to increase in abundance. This is clearly different to published results and what is shown in Fig. 2B of this manuscript.

Indeed, our dataset shows that the quantity of Zar2 decreases. This does not appear anymore in Figure 2B since Zar2 average concentration cannot be estimated. We made an error in the color coding, which has now been corrected in Figure 4B.

(14) Lane 442-444 -> it might be worth mentioning that the interaction between CPEB1 and Maskin, and thus probably its role in regulation of translation, could not be reproduced in other studies (Minshall et al.: CPEB interacts with an ovary-specific eIF4E and 4E-T in early Xenopus oocytes (2007) or Duran-Arque et al.: Comparative analyses of vertebrate CPEB proteins define two subfamilies with coordinated yet distinct functions in post-transcriptional gene regulation (2022)).

This clarification is now mentioned in the text, supported by the two references that have been added (lines 471-477).

(15) Lane 483-485 -> The meaning of these sentences is not entirely clear to me. What exactly is the similarity with the function of Emi1? What does "...binding of Cyclin B1..." mean (binding to which other protein?). What is the similarity between Emi1 and CPEB1/BTG4, both of which are regulators of mRNA stability/polyadenylation?

We apologize if these sentences were unclear. Our intention was to emphasize the central role of ubiquitin ligases in regulating multiple events during meiotic divisions. We used SCF^βTrCP, a wellstudied ubiquitin ligase in Xenopus and mouse oocytes during meiosis, as an example. SCF^βTrCP regulates the degradation of several substrates, including Emi1, Emi2, CPEB1, and Btg4, whose degradation or stabilization is essential for the proper progression of meiosis. Lastly, we highlighted that these regulatory processes, mediated by protein degradation, may be conserved in mitosis, as for example the destruction of Emi1. We have rewritten this paragraph for clarity (lines 513-518).

(16) Lane 521-522 and 572-573 -> the authors write that Myt1 was not detected in their proteome. However, in Fig. 6A they list "pkmyt1" as a class II protein. On Xenbase, "pkmyt1" is the Cdk1 kinase, "Myt1" is a transcription factor, so the authors might have been looking for the wrong protein.

We thank the reviewer for this accurate observation. We have modified the text to correct this error (lines 554 and 607).

(17) Lane 564-565 -> The authors state that Cdk1 activity can be measured by analyzing Cdc27 S428 phosphorylation. However, in vivo the net phosphorylation of a site is always depending on the relevant kinase and phosphatase activities. As S428 is a Cdk1 site, it is not unlikely that it is dephosphorylated by PP2A-B55, which by itself is under the control of Cdk1. Do the authors have direct evidence that the change in phosphorylation of S428 can only be attributed to the changes in Cdk1 activity?

There is evidence in the literature that Cdc27 is dephosphorylated by PP2A (Torres et al., 2010). In Xenopus oocytes, PP2A activity is high during prophase (Lemonnier et al., 2021) and decreases at the time of Cdk1 activation, mediated by the Greatwall-ENSA/Arpp19 system, remaining low until MII (Labbé et al., 2021). Therefore, the period where fluctuations in Cdk1 activity are difficult to assess, from NEBD to MII, corresponds to a phase of inhibited PP2A activity. As a result, the phosphorylation level of Cdc27 reflects primarily the activity of Cdk1. We have added this clarification in the text (lines 597-600).

(18) Fig. 7C and 7D -> in 7C, for Nup35/Nup53 there is a phospho-peptide GIMEVRS(60)PPLHSGG. In Fig. 7D phosphorylation of GVMEMRS(59)PLFSGG is analyzed. Is this the same phosphosite/region of Nup35/Nup53? How can there be a slightly different version of the same peptide in one protein? Are these the L- and S-version of Nup35/Nup53? It is also very surprising that the two phosphosites belong to different classes, class III and class II, respectively.

We thank the reviewer for this observation. The peptides GIMEVRS(60)PPLHSGG and GVMEMRS(59)PLFSGG correspond to the same phosphorylation site in the L and S versions of Xenopus laevis Nup35, respectively. The L version peptide was classified as Class III, while the S version was not assigned to any class due to its high phosphorylation level in prophase, which prevented it from meeting the log₂ fold-change threshold of 1 required by our analysis to detect significant differences.

(19) Table 1 -> second last column is headed "Whur, 2014"

The typo has been corrected.

(20) Fig. 8 -> Why are all the traces starting at t=1h after PG?

The labeling of the graphs in Fig. 8 has been corrected, and the traces now begin at t0.

(21) Lane 754 -> Although a minority, there are also some minus-end directed kinesins, e.g. Kifc1

We agree with the reviewer. We should have mentioned that, in addition to dyneins, some kinesins are minus-end directed motors, especially since one of them, Kifc1, is regulated at the level of its accumulation. We have rephrased the relevant sentences to incorporate this observation (lines 790-793).

(22) Section "Assembly of microtubule spindles and microtubule dynamics" -> Although this section clearly has a strong focus on phosphorylation, it might be worth mentioning again that many regulators of the microtubule spindle, e.g. TXP2, are among the upregulated proteins in Fig. 2B/C

We have already discussed that the protein levels of certain key regulators of the mitotic spindle (Tpx2, PRC1, SSX2IP, Kif11/Eg5 among others) are subject to control during meiotic maturation in a previous chapter “Protein accumulation: the machinery of cell division and DNA replication” (lines 230-239). We agree with the reviewer that this important observation can be mentioned again at the beginning of this chapter on phosphorylation control. We have added a sentence regarding this at the start of the paragraph (lines 774-775).

Reviewer #2 (Recommendations for the authors):

While I find the manuscript excellent and detailed already in its current form, I would appreciate including even more comparisons to other systems. In particular, a similar phosphoproteomics experiment has been performed in starfish oocytes undergoing meiosis (Swartz et al, eLife, 2021), and there are several studies on mitosis of diverse mammalian cells. It would be very exciting to see to what extent changes are conserved.

We thank the reviewer for this recommendation, which we have attempted to follow. We have matched our dataset of mass spectrometry using the the phosphor-occupancy_matlab package, available as part of our code repository (https://github.com/elizabeth-van-itallie) previously described in (Van Itallie et al, 2025). Unfortunately, we were unable to match our dataset with the data from Swartz et al. (2021) on starfish oocyte due to the low sequence conservation. However, we have compared our dataset with the dataset from Sun et al. (2024) on mouse oocyte maturation. We identified a total of 408 conserved phosphorylation sites, which mapped to 320 proteins in Xenopus and 277 in mice (refer to a new paragraph: lines 824-860, new Figure 9, Methods: lines 1011-1032 and 1060-1065, and Appendix 7). The phosphorylation patterns during meiosis showed a significant crossspecies correlation (Pearson r = 0.39, p < 0.0001; see new Figure 9A), demonstrating the evolutionary conservation of phosphoproteomic regulation. Important phosphorylation events, including Plk1 at T201, Gwl at S467, and Erk2 at T188, were upregulated in both species, in line with the activation of the Cdk1 and MAPK signaling cascades (Figure 6B, new Figure 9A-B). We validated several of these phosphorylation sites by western blotting and demonstrated their dependency on Cdk1 activation (new Figure 9C). Together, these findings reinforce the notion that fundamental phospho-regulatory pathways are conserved during oocyte maturation in vertebrates.

Reviewer #3 (Recommendations for the authors):

(1) Page 6, the first paragraph of Results section: Please describe the method on how the authors measured and quantified the proteomes in different stages of Xenopus oocyte maturation briefly. Without the experimental design, it is very hard to evaluate the results in the following paragraphs.

As requested by the reviewer, we added a few sentences describing the method of proteomics and phosphoproteomics measurements in oocytes resuming meiosis (lines 151-158).

(2) In the phospho-proteome, it is better to classify the amino acids for the phosphorylation such as Ser, Thr, and Tyr. Particularly how many tyrosine phosphorylations are in the list.

Our phosphosites dataset contains 80% Ser, 19.9% Thr, and 0.01% Tyr. Phospho-Tyr are slightly less abundant than what has been described in the literature (in most cells “roughly 85-90% of protein phosphorylation happens on Ser, ~10% on Thr, and less than 0.05% on Tyr" after Sharma et al., 2014. The same observation was made regarding the distribution of phosphorylated amino acids in mouse oocytes, where phospho-Tyr abundance is relatively diminished in oocytes compared to mouse organs (Sun et al., 2024). These observations are now reported in the manuscript (lines 309-313).

(3) In class II (Figure 3), when Cdk1 (line 326) is a major kinase, how many phosphorylation sites are a target of Cdk1 (with the Cdk1-motif)? Moreover, do the authors find any other consensus sequences for the phosphorylation? Those are either known or unknown. This information would be useful for the readers.

We thank the reviewer for this valuable comment. To address it, we used the kinase prediction server (https://kinase-library.phosphosite.org/kinase-library/score-site) to analyze Class II phosphosites. These new results are mentioned in lines 340-349 and illustrated in a new Figure (Figure 3—figure supplement 1A). We identified 303 sites predicted to be phosphorylated by Cdk1. Of these, 166 were also predicted as Erk1/2 targets, reflecting the similarity between Cdk1 and Erk1/2 consensus motifs.

Cdk1 substrate phosphorylation is governed by more than just the presence of a consensus sequence. In addition to its preference for the (S/T)P×(K/R) motif, Cdk1/cyclin complexes achieve specificity through docking interactions with short linear motifs (SLiMs) recognized by the cyclin subunit (as LxF motifs)(Loog & Morgan, 2005), and via the Cdk-binding subunits Cks1 or Cks2, which interact with phosphorylated threonine residues in primed substrates (Örd et al, 2019). These mechanisms promote processive multisite phosphorylation and allow Cdk1 to target substrates even at non-canonical sites. Our motif-based analysis captures only part of this complexity and may underestimate the number of true Cdk1 targets.

To further explore kinase involvement across phosphosite classes, we extended the analysis to all clusters and identified the most enriched kinase predictions for each (lines 360-365, new Figure 3— figure supplement 1B). In Class II, the most enriched kinases included Cdk1, Erk2, and Plk1, supporting the conclusions derived from the identification of the phosphosites of this Class. But others such as Cdk2, Cdk3, Cdk5, Cdk16, KIS, JNK1, and JNK3 were also identified.

(4) Figure 3B: Why do the authors show this kind of Table only for Class I, not Classes II-V? It would be informative to show candidate proteins in other classes.

We chose to present the candidate proteins from Class I in a table format because the number of phosphosites (136) was too small to allow a meaningful Gene Ontology (GO) enrichment analysis. Therefore, we manually curated the data and highlighted proteins whose Class I phosphosites are associated with specific biological processes. For Classes II–V, the higher number of phosphosites allowed us to perform GO enrichment analyses. Since several of the enriched processes were shared across different classes, and some proteins have phosphosites in multiple classes, we opted to organize the results by biological processes rather than by class. We agree with the reviewer that it is indeed valuable to highlight interesting proteins with Class II–V phosphosites. We have done so in Figures 4 through 8, using graphical representations instead of tables, in order to make the data more accessible and avoid long tables. Additionally, the Supplementary Figures provide detailed phosphorylation trends for many of the proteins discussed in the main figures.

(5) It would be nice if the authors compare this phospho-proteome in Xenopus oocyte maturation with that in mouse oocyte maturation (Sun et al. 2024) in terms of evolutional conservation of the phospho-proteomes.

We thank the reviewer for this suggestion. As now detailed in the manuscript, we compared our Xenopus phosphoproteome with the dataset from Sun et al. (2024) on mouse oocyte maturation using the the phospho_occupancy_matlab package, available as part of our code repository (https://github.com/elizabeth-van-itallie) previously described in (Van Itallie et al, 2025). We identified 408 conserved phosphorylation sites corresponding to 320 Xenopus and 277 mouse proteins (see new paragraph: lines 824-860, new Figure 9, Methods: lines 1011-1032 and 1060-1065, and Appendix 7). Phosphorylation dynamics across meiosis were significantly correlated between the species (Pearson r = 0.39, p < 0.0001; new Figure 9A), highlighting evolutionary conservation of the phosphoproteomes. Key phosphorylation events such as Plk1 at T201, Gwl at S467, and Erk2 at T188 increased in both species, consistent with activation of the Cdk1 and MAPK pathways (Figure 6B, new Figure 9A–B). We validated experimentally several of these phosphorylation sites by western blot (Erk2, Plk1, Fak1 and Akts1) and demonstrated their dependency on Cdk1 activation (new Figure 9C). Together, these new findings support the conservation of key phospho-regulatory mechanisms across vertebrate oocyte maturation.

Minor points:

(1) Reference lists: Please add Sun et al (2024) shown in line 115.

This important reference has been added (lines 115, 134, 313 and 826).

(2) Figure 1, red arrows for the inhibition: This should be "T" shape for a better understanding of these complicated pathways.

We agree with the reviewer’s remark, and we have modified Figure 1.

(3) Line 236-238: The authors referred to the absence of Cdc6 in oocyte maturation in Xenopus. However, Figure 2C shows that Cdc6 belongs to a list of accumulating proteins with Orc1 and Ocr2 etc. and the authors did not discuss this discrepancy in the text. Please clarity the claim.

We apologize for the unclear wording in our text. The section of the manuscript regarding the pre-RC components may have been misleading. The text has been revised to clarify that Cdc6 was not detected in prophase-arrested oocytes by western blot and that it accumulates during meiotic maturation after MI, enabling oocytes to replicate DNA (lines 243-250).

(4) Line 306: Please add the link to phosphosite.org.

The link has been added (line 319).

eLife Assessment

This important paper describes a comprehensive quantitative phospho-proteomic analysis of the meiotic progression of Xenopus oocytes. Using time-resolved proteomic analyses, the authors provide insights into changes in protein levels and phosphorylation states to an unprecedented depth, quality, and quantitative detail. The key findings are compelling and offer a helpful resource for the scientific community.

Reviewer #1 (Public review):

In the revised version of the manuscript, the authors have adequately addressed all our concerns. The authors should spell check their manuscript, e.g., correct phosphor-site to phospho-site etc.

Summary:

The study aims to create a comprehensive repository about the changes in protein abundance and their modification during oocyte maturation in Xenopus laevis.

Reviewer #2 (Public review):

Summary:

The authors analyzed Xenopus oocytes at different stages of meiosis using quantitative phosphoproteomics. Their advanced methods and analyses revealed changes in protein abundances and phosphorylation states to an unprecedented depth and quantitative detail. In the manuscript they provide an excellent interpretation of these findings putting them in the context of past literature in Xenopus as well as in other model systems. The clarity of these explanations improved significantly in the revised version of the manuscript, and several minor imprecisions have been corrected as well.

Strengths:

High-quality data, careful and detailed analysis, and outstanding interpretation in the context of the large body of literature.

Weaknesses:

Merely a resource, none of the findings are tested in functional experiments.

I am very impressed by the quality of the data and the careful and detailed interpretation of the findings. In this form, the manuscript will be an excellent resource to the cell division community in general, and it presents a very large number of hypotheses that can be tested in future experiments. Xenopus has been and still is a popular and powerful model system that led to critical discoveries around countless cellular processes, including the spindle, nuclear envelope, and translational regulation, just to name a few. This also includes a huge body of literature on the cell cycle describing its phosphoregulation. It is indeed somewhat frustrating to see that these earlier studies using phospho-mutants and phospho-antibodies were just scratching the surface. The phosphoproteomics analysis presented here reveals much more extensive and much more dynamic changes in phosphorylation states. Thereby, in my opinion, this manuscript opens a completely new chapter in this line of research, setting the stage for more systematic future studies.

Reviewer #3 (Public review):

Summary:

The authors performed time-resolved proteomics and phospho-proteomics in Xenopus oocytes from prophase I through the MII arrest of the unfertilized egg. The data contains protein abundance and phosphorylation sites of a large number set of proteins at different stages of oocyte maturation. The large sets of data are of high quality. In addition, the authors discussed several key pathways critical for the maturation. The data is very useful for researchers, not only researchers in Xenopus oocytes but also those in oocyte biology in other organisms.

Strengths:

The data of proteomics and phospho-proteomics in Xenopus oocyte maturation is very useful for future studies to understand molecular networks in oocyte maturation.

Weaknesses:

Although the authors offered molecular pathways of the phosphorylation in translation, protein degradation, cell cycle regulation, and chromosome segregation. The authors did not check the validity of the molecular pathways based on their proteomic data by experimentation. But this is not essential since this is a resource paper.

eLife Assessment

The authors quantified intentions and knowledge gaps in scientists' use of sex as a biological variable in their work, and used a workshop intervention to show that while willingness was high, pressure points centered on statistical knowledge and perceived additional monetary costs to research. These important findings demonstrate the difficulty in changing understanding - while interventions can improve knowledge and decrease perceived barriers, the impact was small. The evidence was solid, although the sample size was small for the intervention.

Reviewer #1 (Public review):

Summary:

The authors use the theory of planned behavior to understand whether or not intentions to use sex as a biological variable (SABV), as well as attitude (value), subjective norm (social pressure), and behavioral control (ability to conduct behavior), across scientists at a pharmacological conference. They also used an intervention (workshop) to determine the value of this workshop in changing perceptions and misconceptions. Attempts to understand the knowledge gaps were made.

Strengths:

The use of SABV is limited in terms of researchers using sex in the analysis as a variable of interest in the models (and not a variable to control). To understand how we can improve on the number of researchers examining the data with sex in the analyses, it is vital we understand the pressure points that researchers consider in their work. The authors identify likely culprits in their analyses. The authors also test an intervention (workshop) to address the main bias or impediments for researchers' use of sex in their analyses.

Weaknesses:

There are a number of assumptions the authors make that could be revisited:

(1) that all studies should contain across sex analyses or investigations. It is important to acknowledge that part of the impetus for SABV is to gain more scientific knowledge on females. This will require within sex analyses and dedicated research to uncover how unique characteristics for females can influence physiology and health outcomes. This will only be achieved with the use of female-only studies. The overemphasis on investigations of sex influences limits the work done for women's health, for example, as within-sex analyses are equally important.

(2) It should be acknowledged that although the variability within each sex is not different on a number of characteristics (as indicated by meta-analyses in rats and mice), this was not done on all variables, and behavioral variables were not included. In addition, across-sex variability may very well be different, which, in turn, would result in statistical sex significance. In addition, on some measures, there are sex differences in variability, as human males have more variability in grey matter volume than females. PMID: 33044802.

(3) The authors need to acknowledge that it can be important that the sample size is increased when examining more than one sex. If the sample size is too low for biological research, it will not be possible to determine whether or not a difference exists. Using statistical modelling, researchers have found that depending on the effect size, the sample size does need to increase. It is important to bare this in mind as exploratory analyses with small sample size will be extremely limiting and may also discourage further study in this area (or indeed as seen the literature - an exploratory first study with the use of males and females with limited sample size, only to show there is no "significance" and to justify this as an reason to only use males for the further studies in the work.

Reviewer #2 (Public review):

Summary:

The investigators tested a workshop intervention to improve knowledge and decrease misconceptions about sex inclusive research. There were important findings that demonstrate the difficulty in changing opinions and knowledge about the importance of studying both males and females. While interventions can improve knowledge and decrease perceived barriers, the impact was small.

Strengths:

The investigators included control groups and replicated the study in a second population of scientists. The results appear to be well substantiated. These are valuable findings that have practical implications for fields where sex is included as a biological variable to improve rigor and reproducibility.

Weaknesses:

I found the figures difficult to understand and would have appreciated more explanation of what is depicted, as well as greater space between the bars representing different categories.

Reviewer #3 (Public review):

Summary:

This manuscript aims to determine cultural biases and misconceptions in inclusive sex research and evaluate the efficacy of interventions to improve knowledge and shift perceptions to decrease perceived barriers for including both sexes in basic research.

Overall, this study demonstrates that despite the intention to include both sexes and a general belief in the importance of doing so, relatively few people routinely include both sexes. Further, the perceptions of barriers to doing so are high, including misconceptions surrounding sample size, disaggregation, and variability of females. There was also a substantial number of individuals without the statistical knowledge to appropriately analyze data in studies inclusive of sex. Interventions increased knowledge and decreased perception of barriers.

Strengths:

(1) This manuscript provides evidence for the efficacy of interventions for changing attitudes and perceptions of research.

(2) This manuscript also provides a training manual for expanding this intervention to broader groups of researchers.

Weaknesses:

The major weakness here is that the post-workshop assessment is a single time point, soon after the intervention. As this paper shows, intention for these individuals is already high, so does decreasing perception of barriers and increasing knowledge change behavior, and increase the number of studies that include both sexes?

Similarly, does the intervention start to shift cultural factors? Do these contribute to a change in behavior?

Author response:

Reviewer #1 (Public review):

Summary:

The authors use the theory of planned behavior to understand whether or not intentions to use sex as a biological variable (SABV), as well as attitude (value), subjective norm (social pressure), and behavioral control (ability to conduct behavior), across scientists at a pharmacological conference. They also used an intervention (workshop) to determine the value of this workshop in changing perceptions and misconceptions. Attempts to understand the knowledge gaps were made.

Strengths:

The use of SABV is limited in terms of researchers using sex in the analysis as a variable of interest in the models (and not a variable to control). To understand how we can improve on the number of researchers examining the data with sex in the analyses, it is vital we understand the pressure points that researchers consider in their work. The authors identify likely culprits in their analyses. The authors also test an intervention (workshop) to address the main bias or impediments for researchers' use of sex in their analyses.

Weaknesses:

There are a number of assumptions the authors make that could be revisited:

(1) that all studies should contain across sex analyses or investigations. It is important to acknowledge that part of the impetus for SABV is to gain more scientific knowledge on females. This will require within sex analyses and dedicated research to uncover how unique characteristics for females can influence physiology and health outcomes. This will only be achieved with the use of female-only studies. The overemphasis on investigations of sex influences limits the work done for women's health, for example, as within-sex analyses are equally important.

The Sex and Gender Equity in Research (SAGER) guidelines (1) provide guidance that “Where the subjects of research comprise organisms capable of differentiation by sex, the research should be designed and conducted in a way that can reveal sex-related differences in the results, even if these were not initially expected.”. This is a default position of inclusion where the sex can be determined and analysis assessing for sex related variability in response. This position underpins many of the funding bodies new policies on inclusion.

However, we need to place this in the context of the driver of inclusion. The most common reason for including male and female samples is for those studies that are exploring the effect of a treatment and then the goal of inclusion is to assess the generalisability of the treatment effect (exploratory sex inclusion)(2). The second scenario is where sex is included because sex is one of the variables of interest and this situation will arise because there is a hypothesized sex difference of interest (confirmatory sex inclusion).

We would argue that the SABV concept was introduced to address the systematic bias of only studying one sex when assessing treatment effect to improve the generalisability of the research. Therefore, it isn’t directly to gain more scientific knowledge on females. However, this strategy will highlight when the effect is very different between male and female subjects which will potentially generate sex specific hypotheses.

Where research has a hypothesis that is specific to a sex (e.g. it is related to oestrogen levels) it would be appropriate to study only the sex of interest, in this case females. The recently published Sex Inclusive Research Framework gives some guidance here and allows an exemption for such a scenario classifying such proposals “Single sex study justified” (3).

We plan to add an additional paragraph to the introduction to clarify the objectives behind inclusion and how this assists the research process.

(2) It should be acknowledged that although the variability within each sex is not different on a number of characteristics (as indicated by meta-analyses in rats and mice), this was not done on all variables, and behavioral variables were not included. In addition, across-sex variability may very well be different, which, in turn, would result in statistical sex significance. In addition, on some measures, there are sex differences in variability, as human males have more variability in grey matter volume than females. PMID: 33044802.

The manuscript was highlighting the common argument used to exclude the use of females, which is that females are inherently more variable as an absolute truth. We agree there might be situations, where the variance is higher in one sex or another depending on the biology. We will extend the discussion here to reflect this, and we will also link to the Sex Inclusive Research Framework (3) which highlights that in these situations researchers can utlise this argument provided it is supported with data for the biology of interest.

(3) The authors need to acknowledge that it can be important that the sample size is increased when examining more than one sex. If the sample size is too low for biological research, it will not be possible to determine whether or not a difference exists. Using statistical modelling, researchers have found that depending on the effect size, the sample size does need to increase. It is important to bare this in mind as exploratory analyses with small sample size will be extremely limiting and may also discourage further study in this area (or indeed as seen the literature - an exploratory first study with the use of males and females with limited sample size, only to show there is no "significance" and to justify this as an reason to only use males for the further studies in the work.

The reviewer raises a common problem: where researchers have frequently argued that if they find no sex differences in a pilot then they can proceed to study only one sex. The SAGER guidelines (1), and now funder guidelines (4, 5), challenge that position. Instead, the expectation is for inclusion as the default in all experiments (exploratory inclusion strategy) to allow generalisable results to be obtained. When the results are very different between the male and female samples, then this can be determined. This perspective shift (2) requires a change in mindset and understanding that the driver behind inclusion is of generalisability not exploration of sex differences. This will be added to the introduction as an additional paragraph exploring the drivers behind inclusion.

We agree with the reviewer that if the researcher is interested in sex differences in an effect (confirmatory inclusion strategy, aka sex as a primary variable) then the N will need to be higher. However, in this situation, one, of course, must have male and female samples in the same experiment to allow the simultaneous exploration to assess the dependency on sex.

Reviewer #2 (Public review):

Summary:

The investigators tested a workshop intervention to improve knowledge and decrease misconceptions about sex inclusive research. There were important findings that demonstrate the difficulty in changing opinions and knowledge about the importance of studying both males and females. While interventions can improve knowledge and decrease perceived barriers, the impact was small.

Strengths:

The investigators included control groups and replicated the study in a second population of scientists. The results appear to be well substantiated. These are valuable findings that have practical implications for fields where sex is included as a biological variable to improve rigor and reproducibility.

Thank you for assessment and highlighting these strengths. We appreciate your recognition of the value and practical implications of this work.

Weaknesses:

I found the figures difficult to understand and would have appreciated more explanation of what is depicted, as well as greater space between the bars representing different categories.

We plan to review the figures and figure legends to improve clarity of the data.

Reviewer #3 (Public review):

Summary:

This manuscript aims to determine cultural biases and misconceptions in inclusive sex research and evaluate the efficacy of interventions to improve knowledge and shift perceptions to decrease perceived barriers for including both sexes in basic research.

Overall, this study demonstrates that despite the intention to include both sexes and a general belief in the importance of doing so, relatively few people routinely include both sexes. Further, the perceptions of barriers to doing so are high, including misconceptions surrounding sample size, disaggregation, and variability of females. There was also a substantial number of individuals without the statistical knowledge to appropriately analyze data in studies inclusive of sex. Interventions increased knowledge and decreased perception of barriers. Strengths:

(1) This manuscript provides evidence for the efficacy of interventions for changing attitudes and perceptions of research.

(2) This manuscript also provides a training manual for expanding this intervention to broader groups of researchers.

Thank you for highlighting these strengths. We appreciate your recognition that the intervention was effect in changing attitudes and perception. We deliberately chose to share the material to provide the resources to allow a wider engagement.

Weaknesses:

The major weakness here is that the post-workshop assessment is a single time point, soon after the intervention. As this paper shows, intention for these individuals is already high, so does decreasing perception of barriers and increasing knowledge change behavior, and increase the number of studies that include both sexes? Similarly, does the intervention start to shift cultural factors? Do these contribute to a change in behavior?

Measuring change in behaviour following an intervention is challenging and hence we had implemented an intention score as a proxy for behaviour. We appreciate the benefit of a long-term analysis, but it was beyond the scope of this study and would need a larger dataset size to allow for attrition. We agree that the strategy implemented has weaknesses. We plan to extend the limitation section in the discussion to include these.

References

(1) Heidari S, Babor TF, De Castro P, Tort S, Curno M. Sex and Gender Equity in Research: rationale for the SAGER guidelines and recommended use. Res Integr Peer Rev. 2016;1:2.

(2) Karp NA. Navigating the paradigm shift of sex inclusive preclinical research and lessons learnt. Commun Biol. 2025;8(1):681.

(3) Karp NA, Berdoy M, Gray K, Hunt L, Jennings M, Kerton A, et al. The Sex Inclusive Research Framework to address sex bias in preclinical research proposals. Nat Commun. 2025;16(1):3763.

(4) MRC. Sex in experimental design - Guidance on new requirements https://www.ukri.org/councils/mrc/guidance-for-applicants/policies-and-guidance-for-researchers/sex-in-experimental-design/: UK Research and Innovation; 2022

(5) Clayton JA, Collins FS. Policy: NIH to balance sex in cell and animal studies. Nature. 2014;509(7500):282-3.

Author response:

Public Reviews:

Reviewer #1 (Public review):

Summary:

In this interesting and original paper, the authors examine the effect that heat stress can have on the ability of bacterial cells to evade infection by lytic bacteriophages. Briefly, the authors show that heat stress increases the tolerance of Klebsiella pneumoniae to infection by the lytic phage Kp11. They also argue that this increased tolerance facilitates the evolution of genetically encoded resistance to the phage. In addition, they show that heat can reduce the efficacy of phage therapy. Moreover, they define a likely mechanistic reason for both tolerance and genetically encoded resistance. Both lead to a reorganization of the bacterial cell envelope, which reduces the likelihood that phage can successfully inject their DNA.

Strengths:

I found large parts of this paper well-written and clearly presented. I also found many of the experiments simple yet compelling. For example, the experiments described in Figure 3 clearly show that prior heat exposure can affect the efficacy of phage therapy. In addition, the experiments shown in Figures 4 and 6 clearly demonstrate the likely mechanistic cause of this effect. The conceptual Figure 7 is clear and illustrates the main ideas well. I think this paper would work even without its central claim, namely that tolerance facilitates the evolution of resistance. The reason is that the effect of environmental stressors on stress tolerance has to my knowledge so far only been shown for drug tolerance, not for tolerance to an antagonistic species.

Weaknesses:

I did not detect any weaknesses that would require a major reorganization of the paper, or that may require crucial new experiments. However, the paper needs some work in clarifying specific and central conclusions that the authors draw. More specifically, it needs to improve the connection between what is shown in some figures, how these figures are described in the caption, and how they are discussed in the main text. This is especially glaring with respect to the central claim of the paper from the title, namely that tolerance facilitates the evolution of resistance. I am sympathetic to that claim, especially because this has been shown elsewhere, not for phage resistance but for antibiotic resistance. However, in the description of the results, this is perhaps the weakest aspect of the paper, so I'm a bit mystified as to why the authors focus on this claim. As I mentioned above, the paper could stand on its own even without this claim.

Thank you for your feedback. We understand your concern regarding the central claim that tolerance facilitates the evolution of resistance, while the paper can stand on its own without this claim, we think it provides an important layer to the interpretation of our findings. Considering your comments, we plan to revise the title and adjust to “Heat Stress Induces Phage Tolerance in Bacteria”.

More specific examples where clarification is needed:

(1) A key figure of the paper seems to be Figure 2D, yet it was one of the most confusing figures. This results from a mismatch between the accompanying text starting on line 92 and the figure itself. The first thing that the reader notices in the figure itself is the huge discrepancy between the number of viable colonies in the absence of phage infection at the two-hour time point. Yet this observation is not even mentioned in the main text. The exclusive focus of the main text seems to be on the right-hand side of the figure, labeled "+Phage". It is from this right-hand panel that the authors seem to conclude that heat stress facilitates the evolution of resistance. I find this confusing, because there is no difference between the heat-treated and non-treated cells in survivorship, and it is not clear from this data that survivorship is caused by resistance, not by tolerance/persistence. (The difference between tolerance and resistance has only been shown in the independent experiments of Figure 1B.)

Thank you for your helpful comment. Figure 2d presents colony counts from a plating assay following the phage killing experiment in Figure 2c. Bacteria collected after 0 and 2 hours of phage exposure were plated on both phage-free (−phage) and phage-containing (+phage) plates. The “−phage” condition reflects total survivors, while the “+phage” condition indicates the resistant subset.

As seen in Figure 2d (left part), heat-treated bacteria showed markedly higher survival on phage-free plates than untreated cells, which were largely eliminated by phage. However, resistant colony counts on phage-containing plates were similar between two groups (as shown in figure 2d right part), suggesting that heat stress increased survival but did not promote resistance.

To clarify, we have revised the labels in Figure 2d as follows: “Total” will replace “-phage” to indicate the total survivors from the phage killing assay, and “Resisters” will replace “+phage” to indicate the resistant survivors, which are detected on phage-containing plates. This adjustment should eliminate any confusion and better reflect the experimental design.

Figure 2F supports the resistance claim, but it is not one of the strongest experiments of the paper, because the author simply only used "turbidity" as an indicator of resistance. In addition, the authors performed the experiments described therein at small population sizes to avoid the presence of resistance mutations. But how do we know that the turbidity they describe does not result from persisters?

I see three possibilities to address these issues. First, perhaps this is all a matter of explaining and motivating this particular experiment better. Second, the central claim of the paper may require additional experiments. For example, is it possible to block heat induced tolerance through specific mutations, and show that phage resistance does not evolve as rapidly if tolerance is blocked? A third possibility is to tone down the claim of the paper and make it about heat tolerance rather than the evolution of heat resistance.

Thank you for your thoughtful comment. We appreciate the opportunity to clarify the interpretation of Figure 2f and the rationale behind the experimental design. We agree that turbidity alone cannot fully distinguish resistance from persistence. However, our earlier experiments (Figures 2d and 2e) demonstrated that heat-treated survivors remained largely susceptible to phage, indicating that heat stress does not directly induce resistance. This led us to hypothesize that heat enhances phage tolerance, which in turn increases the likelihood of resistance emergence during subsequent infection.

To test this, we used a low initial bacterial population (~10³ CFU per well) to minimize the chance of pre-existing resistance. Bacteria were exposed to phages at MOIs of 1, 10, and 100 and incubated for 24 hours in 100 µL volumes. This setup ensured:

(1) The low initial population minimizes the presence of pre-existing resistant mutants, ensuring that any phage-resistant bacteria observed arise during the infection process.

(2) The high MOI (≥ 1) ensures that each bacterial cell has a high probability of infection by at least one phage.

(3) The small volume (100 µL per well) maximizes the interaction between bacteria and phages, ensuring rapid infection of susceptible bacteria, which leads to clear wells. If resistant mutants arise, they will grow and cause turbidity.

Thus, the turbidity observed in heat-treated samples reflects de novo emergence and outgrowth of resistant mutants from a tolerant population. This assay supports the idea that heat-induced tolerance increases the probability of resistance evolution, rather than directly causing resistance.

We have revised the text to better explain this experimental logic and adjust the framing of our conclusions accordingly.

A minor but general point here is that in Figure 2D and in other figures, the labels "-phage" and "+phage" do not facilitate understanding, because they suggest that cells in the "-phage" treatment have not been exposed to phage at all, but that is not the case. They have survived previous phage treatment and are then replated on media lacking phage.

Thank you for your valuable comment. To clarify, we have revised the labels in Figure 2d as follows: “Total” will replace “-phage” to indicate the total survivors from the phage killing assay, and “Resisters” will replace “+phage” to indicate the resistant survivors, which are detected on phage-containing plates.

(2) Another figure with a mismatch between text and visual materials is Figure 5, specifically Figures 5B-F. The figure is about two different mutants, and it is not even mentioned in the text how these mutants were identified, for example in different or the same replicate populations. What is more, the two mutants are not discussed at all in the main text. That is, the text, starting on line 221 discusses these experiments as if there was only one mutant. This is especially striking as the two mutants behave very differently, as, for example, in Figure 5C. Implicitly, the text talks about the mutant ending in "...C2", and not the one ending in "...C1". To add to the confusion, the text states that the (C2) mutant shows a change in the pspA gene, but in Figure 5f, it is the other (undiscussed) mutant that has a mutation in this gene. Only pspA is discussed further, so what about the other mutants? More generally, it is hard to believe that these were the only mutants that occurred in the genome during experimental evolution. It would be useful to give the reader a 2-3 sentence summary of the genetic diversity that experimental evolution generated.

Thank you for your thoughtful comment. In our heat treatment evolutionary experiment, we isolated six distinct bacterial clones, of which two are highlighted in the manuscript as representative examples. One clone, BC2G11C1, acquired both heat tolerance and phage resistance, while another clone, BC3G11C2, became heat-tolerant but did not develop resistance to phage infection. This variation highlights the inherent diversity in evolutionary responses when exposed to selective pressures. It demonstrates that not all evolutionary pathways lead to the same outcome, even under similar stress conditions. This variability is a key observation in our study, illustrating that different genetic adaptations may arise depending on the specific mutations or genetic context, and not every strain will evolve phage resistance in parallel with heat tolerance. We have updated the manuscript to better reflect this diversity in the evolutionary trajectories observed.

Reviewer #2 (Public review):

Summary:

An initial screening of pretreatment with different stress treatments of K. pneumoniae allowed the identification of heat stress as a protection factor against the infection of the lytic phage Kp11. Then experiments prove that this is mediated not by an increase of phage-resistant bacteria but due to an increase in phage transient tolerant population, which the authors identified as bacteriophage persistence in analogy to antibiotic persistence. Then they proved that phage persistence mediated by heat shock enhanced the evolution of bacterial resistance against the phage. The same trait was observed using other lytic phages, their combinations, and two clinical strains, as well as E. coli and two T phages, hence the phenomenon may be widespread in enterobacteria.

Next, the elucidation of heat-induced phage persistence was done, determining that phage adsorption was not affected but phage DNA internalization was impaired by the heat pretreatment, likely due to alterations in the bacterial envelope, including the downregulation of envelope proteins and of LPS; furthermore, heat treated bacteria were less sensitive to polymyxins due to the decrease in LPS.

Finally, cyclic exposure to heat stress allowed the isolation of a mutant that was both resistant to heat treatment, polymyxins, and lytic phage, that mutant had alterations in PspA protein that allowed a gain of function and that promoted the reduction of capsule production and loss of its structure; nevertheless this mutant was severely impaired in immune evasion as it was easily cleared from mice blood, evidencing the tradeoffs between phage/heat and antibiotic resistance and the ability to counteract the immune response.

Strengths:

The experimental design and the sequence in which they are presented are ideal for the understanding of their study and the conclusions are supported by the findings, also the discussion points out the relevance of their work particularly in the effectiveness of phage therapy and allows the design of strategies to improve their effectiveness.

Weaknesses:

In its present form, it lacks the incorporation of some relevant previous work that explored the role of heat stress in phage susceptibility, antibiotic susceptibility, tradeoffs between phage resistance and resistance against other kinds of stress, virulence, etc., and the fact that exposure to lytic phages induces antibiotic persistence.

Thank you for your insightful comments. I appreciate your suggestion regarding the inclusion of relevant previous works. I have now incorporated additional citations to discuss these points, including studies on the relationship between heat stress and antibiotic resistance, as well as the tradeoffs between phage resistance and other stress factors.

Reviewer #3 (Public review):

PspA, a key regulator in the phage shock protein system, functions as part of the envelope stress response system in bacteria, preventing membrane depolarization and ensuring the envelope stability. This protein has been associated in the Quorum Sensing network and biofilm formation. (Moscoso M., Garcia E., Lopez R. 2006. Biofilm formation by Streptococcus pneumoniae: role of choline, extracellular DNA, and capsular polysaccharide in microbial accretion. J. Bacteriol. 188:7785-7795; Vidal JE, Ludewick HP, Kunkel RM, Zähner D, Klugman KP. The LuxS-dependent quorum-sensing system regulates early biofilm formation by Streptococcus pneumoniae strain D39. Infect Immun. 2011 Oct;79(10):4050-60.)

It is interesting and very well-developed.

(1) Could the authors develop experiments about the relationship between Quorum Sensing and this protein?

(2) It would be interesting to analyze the link to phage infection and heat stress in relation to Quorum. The authors could study QS regulators or AI2 molecules.

Thank you for your insightful comments and for bringing up the role of PspA in quorum sensing and biofilm formation. However, we would like to clarify a potential misunderstanding: the PspA discussed in our manuscript refers to phage-shock protein A, a key regulator in the bacterial envelope stress response system. This is distinct from the pneumococcal surface protein A, which has been associated with quorum sensing and biofilm formation in Streptococcus pneumoniae (as referenced in your comment).

To avoid any confusion for readers, we will ensure that our manuscript explicitly states “phage-shock protein A (PspA)” at its first mention. We appreciate your feedback and hope this clarification addresses your concern.

(3) Include the proteins or genes in a table or figure from lytic phage Kp11 (GenBank: ON148528.1).

Thank you for your helpful suggestion. We have now included a figure, as appropriate summarizing the proteins of the lytic phage Kp11 (GenBank: ON148528.1) in supplementary Figure S1.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

Issues unrelated to those discussed in the public review

(1) Figure 4a and its caption describe an evolution experiment, but they do not mention how many cycles of high-temperature treatment and growth this experiment lasted. I assume it lasted for more than one cycle, because the methods section mentions "cycles", but the number is not provided.

Thank you for pointing this out. The evolutionary experiment shown in Figure 5a involved 11 cycles of high-temperature treatment and growth. We have now explicitly stated this in the figure legend to ensure clarity: BC: Batch culture, G: Evolution cycle number, C: Colony. BC2G11C1 refers to the first colony from batvh culture 2 after 11 rounds of heat treatment.

(2) It is not clear what Figure 5F is supposed to show. What are the gray boxes? The caption claims that the figure shows non-synonymous mutations, but the only information it contains is about genes that seem to be affected by mutation. Judging from the mismatch between the main text and the figure, the mutants with these mutations may actually be mislabeled.

Thank you for your careful review. Figure 5f highlights the non-synonymous mutations identified in the evolved strains. The gray boxes represent the ancestral strain’s whole genome without mutations, serving as a control. The corresponding labels indicate the specific mutations found in each evolved strain. We have clarified this in the figure caption to improve clarity. Additionally, we have carefully reviewed the labeling to ensure accuracy and consistency between the figure, main text, and sequencing data.

(3) I think that the acronym NC, which is used in just about every figure, is explained nowhere in the paper. Spell out all acronyms at first use.

Thank you for pointing this out. We have rivewed ensure that NC is clearly defined at its first mention in the text and figure legends to improve clarity. Additionally, we have reviewed the manuscript to ensure that all acronyms are properly introduced when first used.

(4) The same holds for the acronym N.D. This is an especially important oversight because N.D. could mean "not determined" or "not detectable", which would lead to very different interpretations of the same figure.

Thank you for your careful review. We have clarified the meaning of N.D., which stands for non-detectable, at its first use to avoid ambiguity and ensure accurate interpretation in the figure legend. Additionally, we have reviewed the manuscript to ensure that all acronyms are clearly defined.

(5) The panel labels (a,b, etc.) in all figure captions are very difficult to distinguish from the rest of the text, and should be better highlighted, for example by using a bold font. However, this is a matter of journal style and will probably be fixed during typesetting.

Thank you for your suggestion. We have adjusted the figure captions to better distinguish panel labels, such as using bold font, to improve readability and final formatting will follow the journal’s style during typesetting.

(6) Line 224: enhanced insusceptibility -> reduced susceptibility.

Thank you for your suggestion. We have revised “enhanced insusceptibility” to “reduced susceptibility” for clarity and precision.

(7) Line 259: mice -> mouse.

Thank you for catching this. We have corrected “mice” to “mouse”.

Reviewer #2 (Recommendations for the authors):

I have no concerns about the experimental design and conclusions of your work; however, I strongly recommend incorporating several relevant pieces of the literature related to your work, in the discussion of your manuscript, specifically:

(1) Previous studies about the role of heat stress in phage infections, see:

Greenrod STE, Cazares D, Johnson S, Hector TE, Stevens EJ, MacLean RC, King KC. Warming alters life-history traits and competition in a phage community. Appl Environ Microbiol. 2024 May 21;90(5):e0028624. doi: 10.1128/aem.00286-24. Epub 2024 Apr 16. PMID: 38624196; PMCID: PMC11107170.

Thank you for your thoughtful comment. We have ensured to incorporate the study by Greenrod et al. (2024) into the discussion to enrich the context of our findings. As this article pointed out, a temperature of 42°C can indeed limit phage infection in bacteria, acting as a barrier from the phage’s perspective. Our study builds on this by demonstrating that bacteria pre-treated with high temperatures exhibit tolerance to phage infection. These findings, together with the work you referenced, underscore the importance of heat stress or elevated temperature in host-phage interactions, with 42°C being particularly relevant in the context of fever. We will make sure to clarify this connection in our revised manuscript.

(2) The effect of heat stress and the tolerance/resistance against other antibiotics besides polymyxins, see:

Lv B, Huang X, Lijia C, Ma Y, Bian M, Li Z, Duan J, Zhou F, Yang B, Qie X, Song Y, Wood TK, Fu X. Heat shock potentiates aminoglycosides against gram-negative bacteria by enhancing antibiotic uptake, protein aggregation, and ROS. Proc Natl Acad Sci U S A. 2023 Mar 21;120(12):e2217254120. doi: 10.1073/pnas.2217254120. Epub 2023 Mar 14. PMID: 36917671; PMCID: PMC10041086.

Thank you for bringing this study to our attention. We have incorporated the findings from Lv et al. (2023) into the discussion of our manuscript, highlighting how sublethal temperatures may facilitate the killing of bacteria by antibiotics like kanamycin. This is consistent with our data showing enhanced susceptibility of heat-shocked bacteria to kanamycin. The study also provides insights into the potential role of PMF, which is relevant to our work on PspA, and strengthens the broader context of heat stress influencing both antibiotic resistance and tolerance.

(3) Perhaps the most relevant overlooked fact was that recently it was demonstrated for E. coli, Klebsiella and Pseudomonas that pretreatment with lytic phages induced antibiotic persistence! Please discuss this finding and its implications for your work, see:

Fernández-García L, Kirigo J, Huelgas-Méndez D, Benedik MJ, Tomás M, García-Contreras R, Wood TK. Phages produce persisters. Microb Biotechnol. 2024 Aug;17(8):e14543. doi: 10.1111/1751-7915.14543. PMID: 39096350; PMCID: PMC11297538.

Sanchez-Torres V, Kirigo J, Wood TK. Implications of lytic phage infections inducing persistence. Curr Opin Microbiol. 2024 Jun;79:102482. doi: 10.1016/j.mib.2024.102482. Epub 2024 May 6. PMID: 38714140.

Thank you for suggesting this important reference. We agree that the phenomenon of phage-induced bacterial persistence is highly relevant to our study. While our manuscript focuses on the role of heat stress in bacterial tolerance and resistance, we acknowledge that bacterial persistence against phages is an established concept. We have incorporated this finding into our discussion, emphasizing how persistence and tolerance can overlap in their effects on bacterial survival, especially under stress conditions like heat treatment. This will provide a more comprehensive understanding of how phage interactions with bacteria can lead to both persistence and resistance.

(4) Finally, you observed a tradeoff pf the pspA* mutant increased phage/heat/polymyxin resistance and decreased immune evasion (perhaps by being unable to counteract phagocytosis), those tradeoffs between gaining phage resistance but losing resistance to the immune system, virulence impairment and resistance against some antibiotics had been extensively documented, see:

Majkowska-Skrobek G, Markwitz P, Sosnowska E, Lood C, Lavigne R, Drulis-Kawa Z. The evolutionary trade-offs in phage-resistant Klebsiella pneumoniae entail cross-phage sensitization and loss of multidrug resistance. Environ Microbiol. 2021 Dec;23(12):7723-7740. doi: 10.1111/1462-2920.15476. Epub 2021 Mar 27. PMID: 33754440.

Gordillo Altamirano F, Forsyth JH, Patwa R, Kostoulias X, Trim M, Subedi D, Archer SK, Morris FC, Oliveira C, Kielty L, Korneev D, O'Bryan MK, Lithgow TJ, Peleg AY, Barr JJ. Bacteriophage-resistant Acinetobacter baumannii are resensitized to antimicrobials. Nat Microbiol. 2021 Feb;6(2):157-161. doi: 10.1038/s41564-020-00830-7. Epub 2021 Jan 11. PMID: 33432151.

García-Cruz JC, Rebollar-Juarez X, Limones-Martinez A, Santos-Lopez CS, Toya S, Maeda T, Ceapă CD, Blasco L, Tomás M, Díaz-Velásquez CE, Vaca-Paniagua F, Díaz-Guerrero M, Cazares D, Cazares A, Hernández-Durán M, López-Jácome LE, Franco-Cendejas R, Husain FM, Khan A, Arshad M, Morales-Espinosa R, Fernández-Presas AM, Cadet F, Wood TK, García-Contreras R. Resistance against two lytic phage variants attenuates virulence and antibiotic resistance in Pseudomonas aeruginosa. Front Cell Infect Microbiol. 2024 Jan 17;13:1280265. doi: 10.3389/fcimb.2023.1280265. Erratum in: Front Cell Infect Microbiol. 2024 Mar 06;14:1391783. doi: 10.3389/fcimb.2024.1391783. PMID: 38298921; PMCID: PMC10828002.

Thank you for highlighting these important studies. We have incorporated the work by Majkowska-Skrobek et al. (2021), Gordillo Altamirano et al. (2021), and García-Cruz et al. (2024) into the discussion to provide further context to the evolutionary trade-offs observed in our study. The findings in these studies, which describe the cross-sensitization to antimicrobials and the loss of multidrug resistance in phage-resistant bacteria, align with our observations of trade-offs in the pspA mutant. Specifically, our results show that while the pspA mutant exhibits increased resistance to phage, heat, and polymyxins, it also experiences a decrease in immune evasion and potential virulence. These trade-offs are significant in understanding the broader consequences of developing resistance to phages and other stressors.

eLife Assessment

This important study analyzes the effect of heat treatment on phage-bacterial interactions and convincingly shows that prior heat exposure alters the bacterial cell envelope, enhancing persistence and bacterial survival when exposed to lytic phages. The study will interest researchers working on antibiotic resistance, tolerance, and phage therapy.

Reviewer #1 (Public review):

Summary:

In this interesting and original paper, the authors examine the effect that heat stress can have on the ability of bacterial cells to evade infection by lytic bacteriophages. Briefly, the authors show that heat stress increases tolerance of Klebsiella pneumoniae to infection by the lytic phage Kp11. They also argue that this increased tolerance facilitates the evolution of genetically encoded resistance to the phage. In addition, they show that heat can reduce the efficacy of phage therapy. Moreover, they define a likely mechanistic reason for both tolerance and genetically encoded resistance. Both lead to a reorganization of the bacterial cell envelope, which reduces the likelihood that phage can successfully inject their DNA.

Strengths:

I found large parts of this paper well written and clearly presented. I also found many of the experiments simple yet compelling. For example, the experiments described in figure 3 clearly show that prior heat exposure can affect the efficacy of phage therapy. In addition the experiments shown in figure 4 and 6 clearly demonstrate the likely mechanistic cause of this effect. The conceptual figure 7 is clear and illustrates the main ideas well. I think this paper would be publishable even without its central claim, namely that tolerance facilitates the evolution of resistance. The reason is that the effect of environmental stressors on stress tolerance has to my knowledge so far only been shown for drug tolerance, not for tolerance to an antagonistic species.

Weaknesses:

I did not detect any weaknesses that would require a major reorganization of the paper, or that may require crucial new experiments without which the paper should not be published. The originally submitted paper needed some work in clarifying specific and central conclusions that the authors draw, which the authors have done during revision.

Reviewer #2 (Public review):

Summary:

An initial screening of pretreatment with different stress treatments of K. pneumonia allowed the identification of heat stress as a protection factor against the infection of the lytic phage Kp11. Then experiments prove that this is mediated not by an increase of phage resistant bacteria but due to an increase in phage transient tolerant population, that the authors identified as bacteriophage persistence in analogy to antibiotic persistence. Then they proved that phage persistence mediated by heath shock enhanced the evolution of bacterial resistance against the phage. The same trait was observed using other lytic phages, their combinations and two clinical strains, as well as E. coli and two T phages, hence the phenomenon may be widespread in enterobacteria.

Next, the elucidation of heat induced phage persistence was done, determining that phage adsorption was not affected but phage DNA internalization was impaired by the heat pretreatment, likely to alterations in the bacterial envelope, including the downregulation of envelope proteins and of LPS; furthermore, heat treated bacteria were less sensitive to polymyxins due to the decrease in LPS.

Finally, cyclic exposure to heat stress allowed the isolation of a mutant that was both resistant to heat treatment, polymyxins and lytic phage, that mutant had alterations in PspA protein that allowed a gain of function and that promoted the reduction of capsule production and loss of its structure; nevertheless this mutant was severely impaired in immune evasion as it was easily cleared from mice blood, evidencing the trade-off's between phage/heat and antibiotic resistance and the ability to counteract the immune response.

Strengths:

The experimental design and the sequence in which they are presented is ideal for the understanding of their study and the conclusions are supported by the findings, also the discussion points out the relevance of their work particularly in the effectiveness of phage therapy and allow the design of strategies to improve their effectiveness.

Weaknesses:

In its present form it lacks the incorporation of some relevant previous work that explored the role of heat stress in phage susceptibility, antibiotic susceptibility, trade offs between phage resistance and resistance against other kinds of stress, virulence, etc. and the fact that exposure to lytic phages induces antibiotic persistence.

Comments on revised version:

Thanks for addressing most of my comments; however, although you replied this in the rebuttal:

"Thank you for highlighting these important studies. We have incorporated the work by Majkowska-Skrobek et al. (2021), Gordillo Altamirano et al. (2021), and García-Cruz et al. (2024) into the discussion "

I was not able to find the new section in the discussion of the manuscript.

eLife Assessment

This manuscript presents important findings on how structural color can be manipulated through a specific single-gene mutation in a motile bacterium. Compelling data provide a promising model to identify genes and molecular mechanisms supporting this widespread optical phenomenon. This work will be of interest to biophysicists and microbiologists working on structural colors and Flavobacterium.

Reviewer #1 (Public review):

Structural colors (SC) are based on nanostructures reflecting and scattering light and producing optical wave interference. All kinds of living organisms exhibit SC. However, understanding the molecular mechanisms and genes involved may be complicated due to the complexity of these organisms. Hence, bacteria that exhibit SC in colonies, such as Flavobacterium IR1, can be good models.

Based on previous genomic mining and co-occurrence with SC in flavobacterial strains, this article focuses on the role of a specific gene, moeA, in SC of Flavobacterium IR1 strain colonies on an agar plate. moeA is involved in the synthesis of the molybdenum cofactor, which is necessary for the activity of key metabolic enzymes in diverse pathways.

The authors clearly showed that the absence of moeA shifts SC properties in a way that depends on the nutritional conditions. They further bring evidence that this effect was related to several properties of the colony, all impacted by the moeA mutant: cell-cell organization, cell motility and colony spreading, and metabolism of complex carbohydrates. Hence, by linking SC to a single gene in appearance, this work points to cellular organization (as a result of cell-cell arrangement and motility) and metabolism of polysaccharides as key factors for SC in a gliding bacterium. This may prove useful for designing molecular strategies to control SC in bacterial-based biomaterials.

Reviewer #2 (Public review):

The authors constructed an in-frame deletion of moeA gene, which is involved in molybdopterin cofactor (MoCo) biosynthesis, and investigated its role in structural colors in Flavobacterium IR1. The deletion of moeA shifted colony color from green to blue, reduced colony spreading, and increased starch degradation, which was attributed to the upregulation of various proteins in polysaccharide utilization loci. This study lays the ground for developing new colorants by modifying genes involved in structural colors.

Overall, this is a well-written paper in which the authors effectively address their research questions through proper experimentation. This work will help us understand the genetic basis of structural colors in Flavobacterium and open new avenues to study the roles of additional genes and proteins in structural colors.

eLife Assessment

This manuscript offers important insights into how polyphosphate (polyP) influences protein phase separation differently from DNA. The authors present compelling evidence that polyP distinguishes among protein conformational ensembles, leading to divergent condensate maturation behaviors that include unfolding and polyproline II formation. In response to reviewer feedback, the authors addressed key concerns by incorporating charge-equivalent DNA controls and extending structural analysis to FruR variants, further reinforcing the polymer-specific effects of polyP. While some discrepancies between protein systems remain unresolved, the study enhances our understanding of how biopolymers influence protein assembly and conformational transitions.

Reviewer #1 (Public review):

In the article Goyal and colleagues investigate the role of negatively charged biopolymers, i.e., polyphosphate (polyP) and DNA, play in phase separation of cytidine repressor (CytR) and fructose repressor (FruR). The authors find that both negative polymers drive the formation of metastable protein/polymer condensates. However, polyP-driven condensates form more gel- or solid-like structures over time while DNA-driven condensates tend to dissipate over time. The authors link this disparate condensate behavior to polyP-induced structures within the enzymes. Specifically, they observe the formation of polyproline II-like structures within two tested enzyme variants in the presence of polyP. Together, their results provide a unique insight into the physical and structural mechanism by which two unique negatively charged polymers can induce distinct phase transitions with the same protein. This study will be a welcomed addition to the condensate field and provide new molecular insights into how binding partner-induced structural changes within a given protein can affect the mesoscale behavior of condensates.

Reviewer #2 (Public review):

Summary:

In the article Goyal et al. investigate how protein/polymer phase transition behavior is modulated by different binding partners-specifically, DNA and polyphosphate (PolyP). The authors show that while both DNA and PolyP can induce metastable condensates, only PolyP drives unique phase transition behaviors by effectively discriminating among initial protein ensembles with varying degrees of conformational heterogeneity, compactness, and plasticity. This selectivity is attributed to PolyP's ability to unfold the enzyme during condensate formation, supported by the observation of polyproline II-rich structures in two tested variants (CytR WT and DM). Overall, this work offers valuable insights into the mechanistic factors underlying condensation assembly and advances our understanding of how molecular interactions influence phase behavior.

Strengths:

The authors employed a well-designed and technically sound experimental approach to investigate how the initial protein conformational ensemble influences phase transition behavior in the presence of two charged polymers. Specifically, they examined phase transitions of CytR and FruR variants in the context of either polyphosphate (PolyP) or DNA, enabling a direct comparison that effectively highlights key differences. This study provides mechanistic insights into the role of PolyP in driving condensation and may contribute to a broader understanding of assembly processes involving PolyP, particularly in the context of bacterial stress responses.

Weaknesses:

The primary weakness of this manuscript lies in the lack of a consistent trend linking the unique phase transitions observed in protein/PolyP systems to the initial protein conformational ensemble. The observed differences in assembly and maturation behavior do not consistently correlate with conformational heterogeneity, plasticity, or compactness of the starting ensemble. This is particularly evident in the divergent outcomes between the CytR/PolyP and FruR/PolyP systems. Consequently, the phase behavior of protein/PolyP condensates does not reliably reflect the composition of the initial conformational ensemble, limiting its effectiveness as a probe for conformational state characterization.

eLife Assessment

This important study offers a powerful empirical test of a highly influential hypothesis in population genetics. It incorporates a large number of animal genomes spanning a broad phylogenetic spectrum and treats them in a rigorous unified pipeline, providing the convincing negative result that effective population size scales neither with the content of transposable elements nor with overall genome size. These observations demonstrate that there is still no simple, global hypothesis that can explain the observed variation in transposable element content and genome size in animals.

Reviewer #1 (Public review):

Summary:

One enduring mystery involving the evolution of genomes is the remarkable variation they exhibit with respect to size. Much of that variation is due to differences in the number of transposable elements, which often (but not always) correlates with the overall quantity of DNA. Amplification of TEs is nearly always either selectively neutral or negative with respect to host fitness. Given that larger effective population sizes are more efficient at removing these mutations, it has been hypothesized that TE content, and thus overall genome size, may be a function of effective population size. The authors of this manuscript test this hypothesis by using a uniform approach to analysis of several hundred animal genomes, using the ration of synonymous to nonsynonymous mutations in coding sequence as a measure of overall strength of purifying selection, which serves as a proxy for effective population size over time. The data convincingly demonstrates that it is unlikely that effective population size has a strong effect on TE content and, by extension, overall genome size (except for birds, which are weird).

Strengths:

Although this ground has been covered before in many other papers, the strength of this analysis is that it is comprehensive and treats all the genomes with the same pipeline, making comparisons more convincing. Although this is a negative result, it is important because it is relatively comprehensive and indicates that there will be no simple, global hypothesis that can explain the observed variation.

Weaknesses:

In the first draft, the authors slipped between assertions of correlation and assertions of cause-effect relationships not established in the results. However, they have corrected the language so that it more carefully makes this distinction.

Reviewer #3 (Public review):

The Mutational Hazard Hypothesis (MHH) suggests that lineages with smaller effective population sizes should accumulate slightly deleterious transposable elements leading to larger genome size. Marino and colleagues tested the MHH using a set of 807 vertebrate, mollusc and insect species. The authors mined repeats de novo and estimated dN/dS for each genome. Then, they used dN/dS and life history traits as reliable proxies for effective population size and tested for correlations between these proxies and repeat content while accounting for phylogenetic nonindependence. The results suggest that overall, lineages with lower effective population sizes do not exhibit increases in repeat content or genome size. This contrasts with expectations from the MHH. The authors speculate that changes in genome size may be driven by lineage-specific host-TE conflicts rather than effective population size.

Strengths:

The general conclusions of this paper are supported by a powerful dataset of phylogenetically diverse species. Furthermore, the hypothesis tested is important and has proved challenging to test in the past due to technical challenges and confounding factors. The use of C-values rather than assembly size for many species (when available) helps to mitigate the challenges associated with underrepresentation of repetitive regions in short-read based genome assemblies. Overall, both the phylogenetic breadth of species considered and the approaches employed make the results highly convincing.

Weaknesses:

My primary concerns were related to possible biases in the author's data due to their approach to TE annotation. The authors have sufficiently acknowledged and addressed these concerns in their revised manuscript. I note no further weaknesses.

eLife Assessment

This manuscript reports valuable findings on the role of the Srs2 protein in turning off the DNA damage signaling response initiated by Mec1 (human ATR) kinase. The data provide convincing evidence that Srs2 interaction with PCNA and ensuing SUMO modification is required for checkpoint downregulation. However, while the model that Srs2 acts at gaps after camptothecin-induced DNA damage is reasonable, direct experimental evidence for this is currently lacking. The work will be of interest to cell biologists studying genome integrity.

Reviewer #1 (Public review):

Overall, the data presented in this manuscript is of good quality. Understanding how cells control RPA loading on ssDNA is crucial to understanding DNA damage responses and genome maintenance mechanisms. The authors used genetic approaches to show that disrupting PCNA binding and SUMOylation of Srs2 can rescue the CPT sensitivity of rfa1 mutants with reduced affinity for ssDNA. In addition, the authors find that SUMOylation of Srs2 depends on binding to PCNA and the presence of Mec1.

Comments on previous revisions:

I am satisfied with the revisions made by the authors, which helped clarify some points that were confusing in the initial submission.

Reviewer #2 (Public review):

This is an interesting paper that delves into the post-translational modifications of the yeast Srs2 helicase and proteins with which it interacts in coping with DNA damage. The authors use mutants in some interaction domains with RPA and Srs2 to argue for a model in which there is a balance between RPA binding to ssDNA and Srs2's removal of RPA.

The manuscript mostly addresses previous concerns by doubling down on the model without providing additional direct evidence of interactions between Srs2 and PCNA, and that "precise sites of Srs2 actions in the genome remain to be determined." One additional Srs2 allele has been examined, showing some effect in combination with rfa1-zm2.

Reviewer #3 (Public review):

The superfamily I 3'-5' DNA helicase Srs2 is well known for its role as an anti-recombinase, stripping Rad51 from ssDNA, as well as an anti-crossover factor, dissociating extended D-loops and favoring non-crossover outcome during recombination. In addition, Srs2 plays a key role in in ribonucleotide excision repair. Besides DNA repair defects, srs2 mutants also show a reduced recovery after DNA damage that is related to its role in downregulating the DNA damage signaling or checkpoint response. Recent work from the Zhao laboratory (PMID: 33602817) identified a role of Srs2 in downregulating the DNA damage signaling response by removing RPA from ssDNA. This manuscript reports further mechanistic insights into the signaling downregulation function of Srs2.

Using the genetic interaction with mutations in RPA1, mainly rfa1-zm2, the authors test a panel of mutations in Srs2 that affect CDK sites (srs2-7AV), potential Mec1 sites (srs2-2SA), known sumoylation sites (srs2-3KR), Rad51 binding (delta 875-902), PCNA interaction (delta 1159-1163), and SUMO interaction (srs2-SIMmut). All mutants were generated by genomic replacement and the expression level of the mutant proteins was found to be unchanged. This alleviates some concern about the use of deletion mutants compared to point mutations. Double mutant analysis identified that PCNA interaction and SUMO sites were required for the Srs2 checkpoint dampening function, at least in the context of the rfa1-zm2 mutant. There was no effect of this mutants in a RFA1 wild type background. This latter result is likely explained by the activity of the parallel pathway of checkpoint dampening mediated by Slx4, and genetic data with an Slx4 point mutation affecting Rtt107 interaction and checkpoint downregulation support this notion. Further analysis of Srs2 sumoylation showed that Srs2 sumoylation depended on PCNA interaction, suggesting sequential events of Srs2 recruitment by PCNA and subsequent sumoylation. Kinetic analysis showed that sumoylation peaks after maximal Mec1 induction by DNA damage (using the Top1 poison camptothecin (CPT)) and depended on Mec1. This data are consistent with a model that Mec1 hyperactivation is ultimately leading to signaling downregulation by Srs2 through Srs2 sumoylation. Mec1-S1964 phosphorylation, a marker for Mec1 hyperactivation and a site found to be needed for checkpoint downregulation after DSB induction, did not appear to be involved in checkpoint downregulation after CPT damage. The data are in support of the model that Mec1 hyperactivation when targeted to RPA-covered ssDNA by its Ddc2 (human ATRIP) targeting factor, favors Srs2 sumoylation after Srs2 recruitment to PCNA to disrupt the RPA-Ddc2-Mec1 signaling complex. Presumably, this allows gap filling and disappearance of long-lived ssDNA as the initiator of checkpoint signaling, although the study does not extend to this step.

Strengths:

(1) The manuscript focuses on the novel function of Srs2 to downregulate the DNA damage signaling response and provide new mechanistic insights.

(2) The conclusions that PCNA interaction and ensuing Srs2-sumoylation are involved in checkpoint downregulation are well supported by the data.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Overall, the data presented in this manuscript is of good quality. Understanding how cells control RPA loading on ssDNA is crucial to understanding DNA damage responses and genome maintenance mechanisms. The authors used genetic approaches to show that disrupting PCNA binding and SUMOylation of Srs2 can rescue the CPT sensitivity of rfa1 mutants with reduced affinity for ssDNA. In addition, the authors find that SUMOylation of Srs2 depends on binding to PCNA and the presence of Mec1.

Comments on revisions:

I am satisfied with the revisions made by the authors, which helped clarify some points that were confusing in the initial submission.

Thank you.

Reviewer #2 (Public Review):

This revised manuscript mostly addresses previous concerns by doubling down on the model without providing additional direct evidence of interactions between Srs2 and PCNA, and that "precise sites of Srs2 actions in the genome remain to be determined." One additional Srs2 allele has been examined, showing some effect in combination with rfa1-zm2. Many of the conclusions are based on reasonable assumptions about the consequences of various mutations, but direct evidence of changes in Srs2 association with PNCA or other interactors is still missing. There is an assumption that a deletion of a Rad51-interacting domain or a PCNA-interacting domain have no pleiotropic effects, which may not be the case. How SLX4 might interact with Srs2 is unclear to me, again assuming that the SLX4 defect is "surgical" - removing only one of its many interactions.

Previous studies have already provided direct evidence for the interaction between Srs2 and PCNA through the Srs2’s PIM region (Armstrong et al, 2012; Papouli et al, 2005); we have added these citations in the text. Similarly. Srs2 associations with SUMO and Rad51 have also been demonstrated (Colavito et al, 2009; Kolesar et al, 2016; Kolesar et al., 2012), and these studies were cited in the text.

We did not state that a deletion of a Rad51-interacting domain or a PCNA-interacting domain have no pleiotropic effects. We only assessed whether these previously characterized mutant alleles could mimic srs2∆ in rescuing rfa1-zm2 defects.

We assessed the genetic interaction between slx4-RIM and srs2-∆PIM mutants, and not the physical interaction between the two proteins. As we described in the text, our rationale for this genetic test is based on that the reports that both slx4 and srs2 mutants impair recovery from the Mec1 induced checkpoint, thus they may affect parallel pathways of checkpoint dampening.

One point of concern is the use of t-tests without some sort of correction for multiple comparisons - in several figures. I'm quite sceptical about some of the p < 0.05 calls surviving a Bonferroni correction. Also in 4B, which comparison is **? Also, admittedly by eye, the changes in "active" Rad53 seem much greater than 5x. (also in Fig. 3, normalizing to a non-WT sample seems odd).

Claims made in this work were based only on pairwise comparison not multi-comparison. We have now made this point clearer in the graphs and in Method. As the values were compared between a wild-type strain and a specific mutant strain, or between two mutants, we believe that t-test is suitable for statistical analysis.

Figure 4B, ** indicates that the WT value is significantly different from that of the slx4-RIM srs2-∆PIM double mutant and from that of srs2-∆PIM single mutant. We have modified the graph to indicate the pair-wide comparison. The 5-fold change of active Rad53 levels was derived by comparing the values between the srs2∆ PIM slx4^RIM-TAP double mutant and wild-type Slx4-TAP. In Figure 3, normalization to the lowest value affords better visualization. This is rather a stylish issue; we would like to maintain it as the other reviewers had no issues.

What is the WT doubling time for this strain? From the FACS it seems as if in 2 h the cells have completed more than 1 complete cell cycle. Also in 5D. Seems fast...

Wild-type W303 strain has less than 90 min doubling time as shown by many labs, and our data are consistent with this. The FACS profiles for wild-type cells shown in Figures 3C, 4C, and 5C are consistent with each other, showing that after G1 cells entered the cell cycle, they were in G2 phase at the 1-hour time points, and then a percentage of the cells exited the first cell cycle by two hours.

I have one over-arching confusion. Srs2 was shown initially to remove Rad51 from ssDNA and the suppression of some of srs2's defects by deleting rad51 made a nice, compact story, though exactly how srs2's "suppression of rad6" fit in isn't so clear (since Rad6 ties into Rad18 and into PCNA ubiquitylation and into PCNA SUMOylation). Now Srs2 is invoked to remove RPA. It seems to me that any model needs to explain how Srs2 can be doing both. I assume that if RPA and Rad51 are both removed from the same ssDNA, the ssDNA will be "trashed" as suggested by Symington's RPA depletion experiments. So building a model that accounts for selective Srs2 action at only some ssDNA regions might be enhanced by also explaining how Rad51 fits into this scheme.

While the anti-recombinase function of Srs2 was better studied, its “anti-RPA” role in checkpoint dampening was recently described by us (Dhingra et al, 2021) following the initial report by the Haber group some time ago (Vaze et al, 2002). A better understanding of this new role is required before we can generate a comprehensive picture of how Srs2 integrates the two functions (and possibly other functions). Our current work addresses this issue by providing a more detailed understanding of this new role of Srs2.

Single molecular data showed that Srs2 strips both RPA and Rad51 from ssDNA, but this effect is highly dynamic (i.e. RPA and Rad51 can rebind ssDNA after being displaced) (De Tullio et al, 2017). As such, generation of “deserted” ssDNA regions lacking RPA and Rad51 in cells can be an unlikely event. Rather, Srs2 can foster RPA and Rad51 dynamics on ssDNA. Additional studies will be needed to generate a model that integrates the anti-recombinase and the anti-RPA roles of Srs2.

As a previous reviewer has pointed out, CPT creates multiple forms of damage. Foiani showed that 4NQO would activate the Mec1/Rad53 checkpoint in G1- arrested cells, presumably because there would be singlestrand gaps but no DSBs. Whether this would be a way to look specifically at one type of damage is worth considering; but UV might be a simpler way to look. As also noted, the effects on the checkpoint and on viability are quite modest. Because it isn't clear (at least to me) why rfa1 mutants are so sensitive to CPT, it's hard for me to understand how srs2-zm2 has a modest suppressive effect: is it by changing the checkpoint response or facilitating repair or both? Or how srs2-3KR or srs2-dPIM differ from rfa1-zm2 in this respect. The authors seem to lump all these small suppressions under the rubric of "proper levels of RPA-ssDNA" but there are no assays that directly get at this. This is the biggest limitation.

CPT treatment is an ideal condition to examine how cells dampen the DNA damage checkpoint, because while most genotoxic conditions (e.g. 4NQO, MMS) induce both the DNA replication checkpoint and the DNA damage checkpoint, CPT was shown to only induced the latter (Menin et al, 2018; Minca & Kowalski, 2011; Redon et al, 2003; Tercero et al, 2003). Future studies examining 4NQO and UV conditions can further expand our understanding of checkpoint dampening in different conditions.

We have previously provided evidence to support the conclusion that srs2 suppression of rfa1-zm is partly mediated by changing checkpoint levels (Dhingra et al., 2021). We cannot exclude the possibility that the suppression may also be related to changes of DNA repair; we have now added this note in the text.

Regarding direct testing RPA levels on DNA, we have previously shown that srs2∆ increased the levels of chromatin associated Rfa1 and this is suppressed by rfa1-zm2 (Dhingra et al., 2021). We have now included chromatin fractionation data to show that srs2-∆PIM also led to an increase of Rfa1 on chromatin, and this was suppressed by rfa1-zm2 (new Fig. S2).

Srs2 has also been implicated as a helicase in dissolving "toxic joint molecules" (Elango et al. 2017). Whether this activity is changed by any of the mutants (or by mutations in Rfa1) is unclear. In their paper, Elango writes: "Rare survivors in the absence of Srs2 rely on structure-specific endonucleases, Mus81 and Yen1, that resolve toxic joint-molecules" Given the involvement of SLX4, perhaps the authors should examine the roles of structure-specific nucleases in CPT survival?

Srs2 has several roles, and its role in RPA antagonism can be genetically separated from its role in Rad51 regulation as we have shown in our previous work (Dhingra et al., 2021) and this notion is further supported by evidence presented in the current work. Srs2’s role in dissolving "toxic joint molecules” was mainly observed during BIR (Elango et al, 2017). Whether it is related to checkpoint dampening will be interesting to address in the future but is beyond of the scope of the current work that seeks to answer the question how Srs2 regulates RPA during checkpoint dampening. Similarly, determining the roles of Mus81 and Yen1 and other structural nucleases in CPT survival is a worthwhile task but it is a research topic well separated from the focus of this work.

Experiments that might clarify some of these ambiguities are proposed to be done in the future. For now, we have a number of very interesting interactions that may be understood in terms of a model that supposes discriminating among gaps and ssDNA extensions by the presence of PCNA, perhaps modified by SUMO. As noted above, it would be useful to think about the relation to Rad6.

Several studies have shown that Srs2’s functional interaction with Rad6 is based on Srs2-mediated recombination regulation (reviewed by (Niu & Klein, 2017). Given that recombinational regulation by Srs2 is genetically separable from the Srs2 and RPA antagonism (Dhingra et al., 2021), we do not see a strong rationale to examine Rad6 in this work, which addresses how Srs2 regulates RPA. With this said, this study has provided basis for future studies of possible cross-talks among different Srs2-mediated pathways.

Reviewer #3 (Public Review):

The superfamily I 3'-5' DNA helicase Srs2 is well known for its role as an anti-recombinase, stripping Rad51 from ssDNA, as well as an anti-crossover factor, dissociating extended D-loops and favoring non-crossover outcome during recombination. In addition, Srs2 plays a key role in in ribonucleotide excision repair. Besides DNA repair defects, srs2 mutants also show a reduced recovery after DNA damage that is related to its role in downregulating the DNA damage signaling or checkpoint response. Recent work from the Zhao laboratory (PMID: 33602817) identified a role of Srs2 in downregulating the DNA damage signaling response by removing RPA from ssDNA. This manuscript reports further mechanistic insights into the signaling downregulation function of Srs2.

Using the genetic interaction with mutations in RPA1, mainly rfa1-zm2, the authors test a panel of mutations in Srs2 that affect CDK sites (srs2-7AV), potential Mec1 sites (srs2-2SA), known sumoylation sites (srs2-3KR), Rad51 binding (delta 875-902), PCNA interaction (delta 1159-1163), and SUMO interaction (srs2SIMmut). All mutants were generated by genomic replacement and the expression level of the mutant proteins was found to be unchanged. This alleviates some concern about the use of deletion mutants compared to point mutations. Double mutant analysis identified that PCNA interaction and SUMO sites were required for the Srs2 checkpoint dampening function, at least in the context of the rfa1-zm2 mutant. There was no effect of this mutants in a RFA1 wild type background. This latter result is likely explained by the activity of the parallel pathway of checkpoint dampening mediated by Slx4, and genetic data with an Slx4 point mutation affecting Rtt107 interaction and checkpoint downregulation support this notion. Further analysis of Srs2 sumoylation showed that Srs2 sumoylation depended on PCNA interaction, suggesting sequential events of Srs2 recruitment by PCNA and subsequent sumoylation. Kinetic analysis showed that sumoylation peaks after maximal Mec1 induction by DNA damage (using the Top1 poison camptothecin (CPT)) and depended on Mec1. This data are consistent with a model that Mec1 hyperactivation is ultimately leading to signaling downregulation by Srs2 through Srs2 sumoylation. Mec1-S1964 phosphorylation, a marker for Mec1 hyperactivation and a site found to be needed for checkpoint downregulation after DSB induction, did not appear to be involved in checkpoint downregulation after CPT damage. The data are in support of the model that Mec1 hyperactivation when targeted to RPA-covered ssDNA by its Ddc2 (human ATRIP) targeting factor, favors Srs2 sumoylation after Srs2 recruitment to PCNA to disrupt the RPA-Ddc2-Mec1 signaling complex. Presumably, this allows gap filling and disappearance of long-lived ssDNA as the initiator of checkpoint signaling, although the study does not extend to this step.

Strengths:

(1) The manuscript focuses on the novel function of Srs2 to downregulate the DNA damage signaling response and provide new mechanistic insights.

(2) The conclusions that PCNA interaction and ensuing Srs2-sumoylation are involved in checkpoint downregulation are well supported by the data.

Weaknesses:

(1) Additional mutants of interest could have been tested, such as the recently reported Pin mutant, srs2-Y775A (PMID: 38065943), and the Rad51 interaction point mutant, srs2-F891A (PMID: 31142613).

(2) The use of deletion mutants for PCNA and RAD51 interaction is inferior to using specific point mutants, as done for the SUMO interaction and the sites for post-translational modifications.

(3) Figure 4D and Figure 5A report data with standard deviations, which is unusual for n=2. Maybe the individual data points could be plotted with a color for each independent experiment to allow the reader to evaluate the reproducibility of the results.

Comments on revisions:

In this revision, the authors adequately addressed my concerns. The only issue I see remaining is the site of Srs2 action. The authors argue in favor of gaps and against R-loops and ssDNA resulting from excessive supercoiling. The authors do not discuss ssDNA resulting from processing of onesided DSBs, which are expected to result from replication run-off after CPT damage but are not expected to provide the 3'-junction for preferred PCNA loading. Can the authors exclude PCNA at the 5'-junction at a resected DSB?

We have now added a sentence stating that we cannot exclude the possibility that PCNA may be positioned at a 5’-junction, as this can be observed in vitro, albert that PCNA loading was seen exclusively at a 3’-junction in the presence of RPA (Ellison & Stillman, 2003; Majka et al, 2006).

Recommendations For the authors:

Reviewer #2 (Recommendations For the authors):

A Bonferroni correction should be made for the multiple comparisons in several figures.

Specific comments:

l. 41. This is a too long and confusing sentence.

Sentence shortened: “These data suggest that Srs2 recruitment to PCNA proximal ssDNA-RPA filaments followed by its sumoylation can promote checkpoint recovery, whereas Srs2 action is minimized at regions with no proximal PCNA to permit RPA-mediated ssDNA protection”.

l. 60. Identify Ddc2 and Mec1 as ATRIP and ATR.

Done.

l. 125 "fails to downregulate RPA levels on chromatin and Mec1-mediated DDC..." fails to downregulate RPA and fails to reduce Mec1-mediated DDC?

Sentence modified: “fails to downregulate both the RPA levels on chromatin and the Mec1-mediated DDC”

l. 204 "consistent with the notion that Srs2 has roles beyond RPA regulation"... What other roles? It's stripping of Rad51? Removing toxic joint molecules? Something else?

Sentence modified: “consistent with the notion that Srs2 has roles beyond RPA regulation, such as in Rad51 regulation and removing DNA joint molecules”.

l. 249 "Significantly, srs2-ΔPIM and -3KR increased the percentage of rfa1-zm2 cells transitioning into the G1 phase" No. Just back to normal. As stated in l. 258: "258 We found that srs2-ΔPIM and srs2-3KR mutants on their own behaved normally in the two DDC assays described above." All of these effects are quite small.

Sentence modified: “Compared with rfa1-zm2 cells, srs2-∆PIM rfa1-zm2 and srs2-3KR rfa1-zm2 cells showed increased percentages of cells transitioning into the G1 phase”.

l. 468 "Our previous work has provided several lines of evidence to support that Rad51 removal by Srs2 is separable from the Srs2-RPA antagonism (Dhingra et al., 2021). What evidence? See my comment above about not having both proteins removed at the same time.

We have addressed this point in our initial rebuttal and some key points are summarized below. In our previous report (Dhingra et al., 2021), we provided several lines of evidence to support the conclusion that Rad51 is not relevant to the Srs2-RPA antagonism. For example, while rad51∆ rescues the hyper-recombination phenotype of srs2∆ cells, rad51∆ did not affect the hyper-checkpoint phenotype of srs2∆. In contrast, rfa1-zm1/zm2 have the opposite effects, that is, rfa1zm1/zm2 suppressed the hyper-checkpoint, but not the hyper-recombination, phenotype of srs2∆ cells. The differential effects of rad51∆ and rfa1-zm1/zm2 were also seen for the ATPase dead allele of Srs2 (srs2K41A). For example, rfa1-zm2 rescued hyper-checkpoint and CPT sensitivity of srs2-K41A cells, while rad51∆ had neither effect. These and other data described by Dhingra et al (2021) suggest that Srs2’s effects on checkpoint vs. recombination can be separated genetically. Consistent with our conclusion summarized above, deleting the Rad51 binding domain in Srs2 (srs2-∆Rad51BD) has no effect on rfa1-zm2 phenotype in CPT (Fig. 2D). This data provides yet another evidence that Srs2 regulation of Rad51 is separable from the Srs2RPA antagonism.

l. 525 "possibility, we tested the separation pin of Srs2 (Y775), which was shown to enables its in vitro helicase activity during the revision of our work..." ?? there was helicase activity during the revision of your work? Please fix the sentence.

Sentence modified: “we tested the separation pin of Srs2 (Y775). This residue was shown to be key for the Srs2’s helicase activity in vitro in a report that was published during the revision of our work (Meir et al, 2023).”

Fig. 3. "srs2-ΔPIM and -3KR allow better G1 entry of rfa1-zm2 cells." is it better entry or less arrest at G2/M? One implies better turning off of a checkpoint, the other suggests less activation of the checkpoint.

This is a correct statement. For all strains examined in Figure 3, cells were seen in G2/M phase after 1-hour CPT treatment, suggesting proper arrest.

References:

Armstrong AA, Mohideen F, Lima CD (2012) Recognition of SUMO-modified PCNA requires tandem receptor motifs in Srs2. Nature 483: 59-63

Colavito S, Macris-Kiss M, Seong C, Gleeson O, Greene EC, Klein HL, Krejci L, Sung P (2009) Functional significance of the Rad51-Srs2 complex in Rad51 presynaptic filament disruption. Nucleic Acids Res 37: 6754-6764.

De Tullio L, Kaniecki K, Kwon Y, Crickard JB, Sung P, Greene EC (2017) Yeast Srs2 helicase promotes redistribution of single-stranded DNA-bound RPA and Rad52 in homologous recombination regulation. Cell Rep 21: 570-577

Dhingra N, Kuppa S, Wei L, Pokhrel N, Baburyan S, Meng X, Antony E, Zhao X (2021) The Srs2 helicase dampens DNA damage checkpoint by recycling RPA from chromatin. Proc Natl Acad Sci U S A 118: e2020185118

Elango R, Sheng Z, Jackson J, DeCata J, Ibrahim Y, Pham NT, Liang DH, Sakofsky CJ, Vindigni A, Lobachev KS et al (2017) Break-induced replication promotes formation of lethal joint molecules dissolved by Srs2. Nat Commun 8: 1790

Ellison V, Stillman B (2003) Biochemical characterization of DNA damage checkpoint complexes: clamp loader and clamp complexes with specificity for 5' recessed DNA. PLoS Biol 1: E33

Kolesar P, Altmannova V, Silva S, Lisby M, Krejci L (2016) Pro-recombination Role of Srs2 Protein Requires SUMO (Small Ubiquitin-like Modifier) but Is Independent of PCNA (Proliferating Cell Nuclear Antigen) Interaction. J Biol Chem 291: 7594-7607.

Kolesar P, Sarangi P, Altmannova V, Zhao X, Krejci L (2012) Dual roles of the SUMO-interacting motif in the regulation of Srs2 sumoylation. Nucleic Acids Res 40: 7831-7843.

Majka J, Binz SK, Wold MS, Burgers PM (2006) Replication protein A directs loading of the DNA damage checkpoint clamp to 5'-DNA junctions. J Biol Chem 281: 27855-27861

Meir A, Raina VB, Rivera CE, Marie L, Symington LS, Greene EC (2023) The separation pin distinguishes the pro- and anti-recombinogenic functions of Saccharomyces cerevisiae Srs2. Nat Commun 14: 8144

Menin L, Ursich S, Trovesi C, Zellweger R, Lopes M, Longhese MP, Clerici M (2018) Tel1/ATM prevents degradation of replication forks that reverse after Topoisomerase poisoning. EMBO Rep 19: e45535

Minca EC, Kowalski D (2011) Replication fork stalling by bulky DNA damage: localization at active origins and checkpoint modulation. Nucleic Acids Res 39: 2610-2623

Niu H, Klein HL (2017) Multifunctional roles of Saccharomyces cerevisiae Srs2 protein in replication, recombination and repair. FEMS Yeast Res 17: fow111

Papouli E, Chen S, Davies AA, Huttner D, Krejci L, Sung P, Ulrich HD (2005) Crosstalk between SUMO and ubiquitin on PCNA is mediated by recruitment of the helicase Srs2p. Mol Cell 19: 123-133

Redon C, Pilch DR, Rogakou EP, Orr AH, Lowndes NF, Bonner WM (2003) Yeast histone 2A serine 129 is essential for the efficient repair of checkpoint-blind DNA damage. EMBO Rep 4: 678-684

Tercero JA, Longhese MP, Diffley JFX (2003) A central role for DNA replication forks in checkpoint activation and response. Mol Cell 11: 1323-1336

Vaze MB, Pellicioli A, Lee SE, Ira G, Liberi G, Arbel-Eden A, Foiani M, Haber JE (2002) Recovery from checkpointmediated arrest after repair of a double-strand break requires Srs2 helicase. Mol Cell 10: 373-385

eLife Assessment

This study reports a fundamental observation concerning cell death regulation by the anti-apoptotic BCL2 family NOXA. The authors convincingly demonstrate that NOXA is destabilized through the interaction with WSB2, a substrate receptor in CRL5 ubiquitin ligase complex, sensitizing the cells to treatments. These are key findings for cell biologists and cancer researchers as they identified a new target impacting drug responsiveness in cancer therapies.

Reviewer #1 (Public Review):

Summary:

In this manuscript, Jiao D et al reported the induction of synthetic lethality by combined inhibition of anti-apoptotic BCL-2 family proteins and WSB2, a substrate receptor in CRL5 ubiquitin ligase complex. Mechanistically, WSB2 interacts with NOXA to promote its ubiquitylation and degradation. Cancer cells deficient in WSB2, as well as heart and liver tissues from Wsb2-/- mice exhibit high susceptibility to apoptosis induced by inhibitors of BCL-2 family proteins. The anti-apoptotic activity of WSB2 is partially dependent on NOXA.

Overall, the finding that WSB2 disruption triggers synthetic lethality to BCL-2 family protein inhibitors by destabilizing NOXA is rather novel. The manuscript is largely hypothesis-driven, with experiments that are adequately designed and executed. However, there are quite a few issues for the authors to address, including those listed below.

Specific comments from the previous round of review:

(1) At the beginning of the Results section, a clear statement is needed as to why the authors are interested in WSB2 and what brought them to analyze "the genetic co-dependency between WSB2 and other proteins".

(2) In general, the biochemical evidence supporting the role of WSB2 as a SOCS box-containing substrate-binding receptor of CRL5 E3 in promoting NOXA ubiquitylation and degradation is relatively weak. First, since NOXA2 binds to WSB2 on its SOCS box, which consists of a BC box for Elongin B/C binding and a CUL5 box for CUL5 binding, it is crucial to determine whether the binding of NOXA on the SOCS box affects the formation of CRL5WSB2 complex. The authors should demonstrate the endogenous binding between NOXA and the CRL5WSB2 complex. Additionally, the authors may also consider manipulating CUL5, SAG, or ElonginB/C to assess if it would affect NOXA protein turnover in two independent cell lines. Second, in all the experiments designed to detect NOXA ubiquitylation in cells, the authors utilized immunoprecipitation (IP) with FLAG-NOXA/NOXA, followed by immunoblotting (IB) with HA-Ub. However, it is possible that the observed poly-Ub bands could be partly attributed to the ubiquitylation of other NOXA binding proteins. Therefore, the authors need to consider performing IP with HA-Ub and subsequently IB with NOXA. Alternatively, they could use Ni-beads to pull down all His-Ub-tagged proteins under denaturing conditions, followed by the detection of FLAG-tagged NOXA using anti-FLAG Ab. The authors are encouraged to perform one of these suggested experiments to exclude the possibility of this concern. Furthermore, an in vitro ubiquitylation assay is crucial to conclusively demonstrate that the polyubiquitylation of NOXA is indeed mediated by the CRL5WSB2 complex.

(3) In their attempt to map the binding regions between NOXA and WSB2, the authors utilized exogenous proteins of both WSB2 and NOXA. To strengthen their findings, it would be more convincing to perform IP with exogenous wt/mutant WSB2 or NOXA and subsequently perform IB to detect endogenous NOXA or WSB2, respectively. Additionally, an in vitro binding assay using purified proteins would provide further evidence of a direct binding between NOXA and WSB2.

Comments on latest version:

The authors have adequately addressed my previous comments.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

I In this manuscript, Jiao D et al reported the induction of synthetic lethal by combined inhibition of anti-apoptotic BCL-2 family proteins and WSB2, a substrate receptor in CRL5 ubiquitin ligase complex. Mechanistically, WSB2 interacts with NOXA to promote its ubiquitylation and degradation. Cancer cells deficient in WSB2, as well as heart and liver tissues from Wsb2-/- mice exhibit high susceptibility to apoptosis induced by inhibitors of BCL-2 family proteins. The anti-apoptotic activity of WSB2 is partially dependent on NOXA.

Overall, the finding, that WSB2 disruption triggers synthetic lethality to BCL-2 family protein inhibitors by destabilizing NOXA, is rather novel. The manuscript is largely hypothesis-driven, with experiments that are adequately designed and executed. However, there are quite a few issues for the authors to address, including those listed below.

Specific comments:

(1) At the beginning of the Results section, a clear statement is needed as to why the authors are interested in WSB2 and what brought them to analyze "the genetic co-dependency between WSB2 and other proteins".

We thank the reviewer for raising this important point. We agree that a clear rationale should be provided at the beginning of the Results section. As reported in previous studies [Ref: 1, 2, 3], strong synthetic interactions have been observed between WSB2 and several mitochondrial apoptosis-related factors, including MCL-1, BCL-xL, and MARCH5. We have referenced these findings in the Discussion section. Motivated by these studies, we became interested in the role of WSB2 and aimed to investigate the specific mechanisms underlying its synthetic lethality with anti-apoptotic BCL-2 family members. We will revise the beginning of the Results section to clearly state this rationale.

(1) McDonald, E.R., 3rd et al. Project DRIVE: A Compendium of Cancer Dependencies and Synthetic Lethal Relationships Uncovered by Large-Scale, Deep RNAi Screening. Cell 170, 577-592 e510 (2017).

(2) DeWeirdt, P.C. et al. Genetic screens in isogenic mammalian cell lines without single cell cloning. Nat Commun 11, 752 (2020).

(3) DeWeirdt, P.C. et al. Optimization of AsCas12a for combinatorial genetic screens in human cells. Nat Biotechnol 39, 94-104 (2021).

(2) In general, the biochemical evidence supporting the role of WSB2 as a SOCS box-containing substrate-binding receptor of CRL5 E3 in promoting NOXA ubiquitylation and degradation is relatively weak. First, since NOXA binds to WSB2 on its SOCS box, which consists of a BC box for Elongin B/C binding and a CUL5 box for CUL5 binding, it is crucial to determine whether the binding of NOXA on the SOCS box affects the formation of CRL5WSB2 complex. The authors should demonstrate the endogenous binding between NOXA and the CRL5WSB2 complex. Additionally, the authors may also consider manipulating CUL5, SAG, or ElonginB/C to assess if it would affect NOXA protein turnover in two independent cell lines.

We thank the reviewer for raising this important point. To determine whether endogenous NOXA binds to the intact CRL5^WSB2 complex, we performed co-immunoprecipitation assays using an antibody against NOXA. Indeed, NOXA co-immunoprecipitated with all subunits of the CRL5^WSB2 complex (Figure 2—figure supplement 1D), suggesting that NOXA binding to WSB2 does not disrupt interactions between WSB2 and the other CRL5 subunits. Moreover, depletion of CRL5 complex components (RBX2/SAG, CUL5, ELOB, or ELOC) through siRNAs in C4-2B or Huh-7 cells also resulted in a marked increase in NOXA protein levels.

Second, in all the experiments designed to detect NOXA ubiquitylation in cells, the authors utilized immunoprecipitation (IP) with FLAG-NOXA/NOXA, followed by immunoblotting (IB) with HA-Ub. However, it is possible that the observed poly-Ub bands could be partly attributed to the ubiquitylation of other NOXA binding proteins. Therefore, the authors need to consider performing IP with HA-Ub and subsequently IB with NOXA. Alternatively, they could use Ni-beads to pull down all His-Ub-tagged proteins under denaturing conditions, followed by the detection of FLAG-tagged NOXA using anti-FLAG Ab. The authors are encouraged to perform one of these suggested experiments to exclude the possibility of this concern. Furthermore, an in vitro ubiquitylation assay is crucial to conclusively demonstrate that the polyubiquitylation of NOXA is indeed mediated by the CRL5WSB2 complex.

We appreciate the reviewer for raising these important considerations regarding our ubiquitylation assays. We fully acknowledge the reviewer's concern that classical ubiquitination assays could potentially detect ubiquitination of proteins interacting with NOXA. However, we would like to clarify that our experimental conditions effectively mitigate this issue. Specifically, cells were lysed using buffer containing 1% SDS followed by boiling at 105°C for 5 minutes. These rigorous denaturing conditions ensure disruption of non-covalent protein interactions, thereby effectively eliminating the possibility of detecting ubiquitination signals from NOXA-associated proteins.

Regarding the suggestion to perform an in vitro ubiquitination assay, we agree this experiment would indeed provide additional evidence. However, due to significant technical complexities associated with reconstituting CRL5-based E3 ubiquitin ligase activity in vitro—which would require the expression and purification of at least six recombinant proteins—such experiments are rarely performed in this context. Furthermore, NOXA is uniquely localized as a membrane protein on the mitochondrial outer membrane, posing additional significant challenges for protein expression and purification. Given the robustness of our current in vivo ubiquitylation assay under stringent denaturing conditions, we believe our existing data sufficiently and conclusively demonstrate NOXA ubiquitination mediated by the CRL5^WSB2 complex.

(3) In their attempt to map the binding regions between NOXA and WSB2, the authors utilized exogenous proteins of both WSB2 and NOXA. To strengthen their findings, it would be more convincing to perform IP with exogenous wt/mutant WSB2 or NOXA and subsequently perform IB to detect endogenous NOXA or WSB2, respectively. Additionally, an in vitro binding assay using purified proteins would provide further evidence of a direct binding between NOXA and WSB2.

We thank the reviewer for raising these important issues. In response to the reviewer’s suggestion to map the binding regions between NOXA and WSB2 more convincingly, we have indeed performed semi-endogenous Co-IP assays, which yielded results consistent with our exogenous protein experiments (Figure 3—figure supplement 1A, B). Concerning the recommendation to further validate direct interaction using purified recombinant proteins, we encountered substantial technical difficulties in obtaining pure and soluble recombinant WSB2 protein. Additionally, given that NOXA is an outer mitochondrial membrane protein and the interaction occurs on mitochondria, we believe that an in vitro binding assay may have limited physiological relevance. We hope the reviewer can appreciate these practical challenges and our current evidence supporting the strong interaction between NOXA and WSB2.

Reviewer #2 (Public Review):

Summary:

Exploring the DEP-MAP database and two drug-screen databases, the authors identify WSB2 as an interactor of several BCL2 proteins. In follow-up experiments, they show that CRL5/WSB2 controls NOXA protein levels via K48 ubiquitination following direct protein-protein interaction, and cell death sensitivity in the context of BH3 mimetic treatment, where WSB2 depletion synergizes with drug treatment.

Strengths:

The authors use a set of orthogonal methods across different model cell lines and a new WSB2 KO mouse model to confirm their findings. They also manage to correlate WSB2 expression with poor prognosis in prostate and liver cancer, supporting the idea that targeting WSB2 may sensitize cancers for treatment with BH3 mimetics.

Weaknesses:

The conclusions drawn based on the findings in cancer patients are very speculative, as regulation of NOXA cannot be the sole function of CRL5/WSB2 and it is hence unclear what causes correlation with patient survival. Moreover, the authors do not provide a clear mechanistic explanation of how exactly higher levels of NOXA promote apoptosis in the absence of WSB2. This would be important knowledge, as usually high NOXA levels correlate with high MCL1, as they are turned over together, but in situations like this, or loss of other E3 ligases, such as MARCH, the buffering capacity of MCL1 is outrun, allowing excess NOXA to kill (likely by neutralizing other BCL2 proteins it usually does not bind to, such as BCLX). Moreover, a necroptosis-inducing role of NOXA has been postulated. Neither of these options is interrogated here.

Recommendations For The Authors:

Reviewer #1 (Recommendations For The Authors):

(1) Figure 2J. The authors showed that "the mRNA levels of NOXA were even reduced in WSB2-KO cells compared to parental cells". What is the possible mechanism? This point should at least be discussed.

We thank the reviewer for raising these important issues. The underlying mechanisms for the significantly lower mRNA levels of NOXA following the KO of WSB2 are not fully understood at present. However, we propose that this could represent a form of negative feedback regulation at the level of gene expression. Specifically, when the protein levels of BNIP3/3L rise sharply, it may activate mechanisms that suppress their own mRNA synthesis or stability, serving as a buffering system to prevent further protein accumulation. Such negative feedback loops may be critical for maintaining cellular homeostasis and avoiding excessive protein production. Moreover, this phenomenon is frequently observed in other studies investigating substrates targeted by E3 ubiquitin ligases for degradation. We have elaborated on this point in the Discussion section.

(2) Figure 2M. A previous study has clearly demonstrated that NOXA is subjected to ubiquitylation and degradation by CRL5 E3 ligase (PMID: 27591266). This paper should be cited. Also, in that publication, NOXA ubiquitylation is via the K11 linkage, not the K48 linkage. The authors should include K11R mutant in their assay.

We thank the reviewer for raising this important issue. We thank the reviewer for suggesting the relevant reference (PMID: 27591266), which we have now cited accordingly. Additionally, we would like to clarify that our new in vivo ubiquitination assays included the K11R and K11-only ubiquitin mutants, and our data demonstrate that WSB2-mediated NOXA ubiquitination indeed involves the K11 linkage ubiquitination（Figure 2—figure supplement 1E).

(3) Figure 3H, J. The authors stated, "By mutating these lysine residues to arginine, we found that WSB2-mediated NOXA ubiquitination was completely abolished". Which one of the three lysine residues is playing the dominant role?

We thank the reviewer for raising this important issue. To address this, we generated FLAG-NOXA mutants individually substituting lysine residues K35, K41, and K48 with arginine. In vivo ubiquitination assays demonstrated that lysine 48 (K48) is the predominant residue responsible for WSB2-mediated NOXA ubiquitination (Figure 3—figure supplement 1C).

(4) Figure 3N. The authors need to show that the fusion peptide containing C-terminal NOXA peptide competitively inhibits the interaction between endogenous WSB2 and NOXA and extends the protein half-life of NOXA, leading to NOXA accumulation.

We sincerely thank the reviewer for raising these important issues. As suggested, we investigated whether the fusion peptide containing the C-terminal NOXA sequence competitively disrupts the interaction between endogenous WSB2 and NOXA, subsequently influencing NOXA stability. Our results demonstrated that treatment with this fusion peptide indeed significantly reduced the endogenous interaction between WSB2 and NOXA (Figure 3—figure supplement 1D). Furthermore, we observed that the peptide dose-dependently increased endogenous NOXA protein levels and prolonged its protein half-life, thereby resulting in the accumulation of NOXA (Figure 3N; Figure 3—figure supplement 1E, F). These findings collectively indicate that the fusion peptide competitively inhibits the WSB2-NOXA interaction, stabilizes NOXA protein, and enhances its accumulation.

(5) Figure 4. a) It would be better to investigate whether WSB2 knockdown can sensitize cancer cells to the treatment with ABT-737 or AZD5991, evidenced by a decrease in both IC50 values and clonogenic survival rates and whether such sensitization is dependent on NOXA. b) The authors need to show the levels of cleaved caspase-3/7/9 and the percentages of apoptotic cells in shNC cells upon silencing of WSB2 in Figure 4A-F. c) It will be more convincing to repeat the experiment to show synthetic lethality by WSB2 disruption and MCL-1 inhibitor AZD5991 treatment using another cell line, such as WSB2-deficient Huh-7 cells in Figure 4 I&J.

We sincerely thank the reviewer for these valuable and constructive suggestions. Regarding point (a): We believe that our current Western blot and flow cytometry data (Figure 4G–L) have already provided strong evidence that WSB2 depletion enhances apoptosis in response to ABT-737 and AZD5991. Therefore, we consider that additional IC50 and clonogenic survival assays, while informative, may not be essential for supporting our conclusion. Furthermore, as shown in Figure 5A–F, we found that silencing NOXA largely, though not completely, reversed the enhanced apoptosis triggered by these inhibitors in WSB2-deficient cells, suggesting that the sensitization effect is at least partially dependent on NOXA.

Regarding point (b): We have shown that WSB2 knockout alone had no impact on the levels of cleaved caspase-3/7/9 or the percentages of apoptotic cells in Huh-7 and C4-2B cells (Figure 4G-L and Figure 4—figure supplement 1A-D), indicating that WSB2 loss does not induce apoptosis on its own under basal conditions.

Regarding point (c): We appreciate the reviewer’s suggestion and have now repeated the experiment in WSB2 knockout Huh-7 cells. The new results further support the synthetic lethality between WSB2 loss and AZD5991 treatment (Figure 4—figure supplement 1C, D).

(6) Figure 5A/C/E. The effect of siNOXA is minor, if any, for cleavage of caspases. The same thing for Figure 6F/H.

We appreciate the reviewer’s insightful observation regarding the relatively modest effect of shNOXA on caspase cleavage in Figures 5A/C/E and Figures 6F/H. Indeed, we acknowledge that the reduction in caspase cleavage following NOXA knockdown is moderate. However, consistent with our discussions in the manuscript, NOXA knockdown significantly—but not completely—rescued the increased apoptosis observed in WSB2-deficient cells treated with BCL-2 family inhibitors. This suggests that while NOXA plays a notable role, additional mechanisms or unidentified targets may also be involved in WSB2-mediated regulation of apoptosis.

(7) Figure 5 I&J. The authors may consider performing IHC staining, immunofluorescence, or WB analysis to show the levels of NOXA and cleaved caspases or PARP in xenograft tumors. This would provide in vivo evidence of significant apoptosis induction resulting from the co-administration of ABT-737 and R8-C-terminal NOXA peptide.

We appreciate the reviewer's thoughtful suggestion regarding additional immunohistochemical or immunofluorescence analyses in xenograft tumors. However, due to current limitations in available antibodies suitable for reliable detection of NOXA by IHC and IF, we are unable to perform these experiments. We greatly appreciate the reviewer's understanding of this technical constraint. Nevertheless, our existing data collectively supports the conclusion that the combination of ABT-737 and R8-C-terminal NOXA peptide significantly enhances apoptosis in vivo.

(8) Figure 7. Does an inverse correlation exist between the protein levels of WSB2 and NOXA in RPAD or LIHC tissue microarrays? On page 12, in the first paragraph, Figure 7M-P was cited incorrectly.

We sincerely thank the reviewer for raising this important issue. As mentioned above, due to current limitations regarding the availability of suitable antibodies that can reliably detect NOXA by IHC, we regret that it is not feasible to experimentally address this question at this time.

Additionally, we have carefully corrected the citation error involving Figure 7M-P on page 12, as pointed out by the reviewer.

(9) Figure S1D. BCL-W levels were reduced upon WSB2 overexpression, which should be acknowledged.

We sincerely thank the reviewer for raising this important issue. We acknowledge that BCL-W protein levels were slightly reduced upon WSB2 overexpression in Figure S1D. However, this effect is distinct from the pronounced reduction observed in NOXA protein levels. We have revised the manuscript to clarify this point. Additionally, we recognize that transient overexpression systems may occasionally lead to non-specific or artifactual changes. Our exogenous expression and co-immunoprecipitation experiments did not support an interaction between BCL-W and WSB2. Therefore, the observed reduction of BCL-W under these conditions may not reflect a physiologically relevant regulation.

(10) Figure S4. Given WSB2 KO mice are viable; the authors may consider determining whether these mice are more sensitive to radiation-induced tissue damage or but more resistant to radiation-induced tumorigenesis?

We sincerely thank the reviewer for this insightful and biologically meaningful suggestion. We agree that investigating the potential role of WSB2 in radiation-induced tissue damage and tumorigenesis would be of great interest. However, conducting such experiments requires access to specialized irradiation facilities, which are currently unavailable to us. Nevertheless, we recognize the value of this line of investigation and plan to explore it in our future studies.

(11) All data were displayed as mean{plus minus}SD. However, for data from three independent experiments, it is more appropriate to present the results as mean{plus minus}SEM, not mean{plus minus}SD.

We sincerely thank the reviewer for highlighting this important issue. In line with the reviewer's suggestion, we have revised the manuscript accordingly and now present data from three independent experiments as mean ± SEM.

(12) The figure legends require careful review: i) The low dose of ABT-199 (Figure 6H) and the dose of ABT-199 used in Figure 6I are missing. ii) The legends for Figure S1D-E are incorrect. iii) The name of the antibody in the legend of Figure S3C is incorrect.

We sincerely thank the reviewer for raising these important issues. We have carefully corrected all the errors mentioned. In addition, we have thoroughly reviewed the manuscript to prevent similar errors.

Reviewer #2 (Recommendations For The Authors):

The authors focus on NOXA, after initially identifying WSB2 to interact with several BCL2 proteins. The rationale behind this is that WSB2 depletion or overexpression affects NOXA levels, but none of the other BCL2 proteins tested, as stated in the text. Yet, BCLW is also depleted upon overexpression of WSB2 (Supplementary Figure 1). How does this phenomenon relate to the sensitization noted, is BCL-W higher in WSB2 KO cells? It does not seem so though. This warrants discussion.

We appreciate the reviewer for raising this important issue. Our results showed that overexpression of WSB2 markedly reduced NOXA levels, while the levels of other BCL-2 family proteins remained unaffected or minimally affected, such as BCL-W (Figure 2—figure supplement 1A). Furthermore, depletion of WSB2 through shRNA-mediated KD or CRISPR/Cas9-mediated KO in C4-2B cells or Huh-7 cells led to a marked increase in the steady-state levels of endogenous NOXA, without affecting other BCL-2 family proteins examined, included BCL-W (Figure 2A-C, Figure 2—figure supplement 2A, B).

If WSB2 depletion does not affect MCL1 levels, how does excess NOXA actually kill? Does it bind to any (other) prosurvival proteins under conditions of WSB2 depletion? Is the MCL1 half-life changed?

We appreciate the reviewer for raising this important point. NOXA is a BH3-only protein known to promote apoptosis primarily by binding to and neutralizing anti-apoptotic BCL-2 family members, especially MCL-1, via its BH3 domain. It can inhibit MCL-1 either through competitive binding or by facilitating its ubiquitination and subsequent proteasomal degradation. In our system, the total protein levels of MCL-1 remained unchanged in WSB2 knockout cells, suggesting that NOXA may not be promoting apoptosis through enhanced MCL-1 degradation. Instead, we speculate that the accumulation of NOXA in WSB2-deficient cells enhances apoptosis by sequestering MCL-1 through direct binding, thereby freeing pro-apoptotic effectors such as BAK and BAX. In line with our observations, Nakao et al. reported that deletion of the mitochondrial E3 ligase MARCH5 led to a pronounced increase in NOXA expression, while leaving MCL-1 protein levels unchanged in leukemia cell lines (Leukemia. 2023 ;37:1028-1038., PMID: 36973350).

Additionally, NOXA has been reported to interact with other anti-apoptotic proteins, including BCL-XL. It is therefore possible that under conditions of WSB2 depletion, excess NOXA may also bind to BCL-XL and relieve its inhibition of BAX/BAK, further contributing to apoptosis. Future experiments assessing NOXA binding partners in WSB2-deficient cells would help clarify this mechanism.

I think some initial insights into the mechanism underlying the sensitization would add a lot to this study. Is there a role of BFL1/A1 in any of these cell lines, as it can also rather selectively bind to NOXA and is sometimes deregulated in cancer?

We appreciate the reviewer for raising this important issue. While BFL1/A1 is indeed another anti-apoptotic BCL-2 family member that can selectively bind to NOXA and has been implicated in cancer, our study primarily focuses on the WSB2-NOXA axis. However, given its potential involvement in apoptosis regulation, it would be an interesting direction for future studies to explore whether BFL1/A1 contributes to NOXA-mediated sensitization in specific cellular contexts.

Otherwise, this is a very nice and convincing study.

eLife Assessment

This paper reports on a correlation between diminished cardiolipin content and the severity of steatohepatitis in human subjects. This is supported further by experimental evidence from mice in which the gene encoding a key enzyme in cardiolipin synthesis has been compromised in the liver. The correlations established between lipidology, mitochondrial function, and the induction of respiration and oxidative stress are notable and will be useful to researchers in the field. However, given that the causal relationship between lipid perturbation and the progression of steatohepatitis implied in the title has not been tested experimentally, the evidence supporting the paper's key conclusion is incomplete.

Reviewer #1 (Public review):

Summary:

The manuscript by Brothwell and colleagues describes a central role for hepatic cardiolipin deficiency in MASH. The authors identify cardiolipin as a mediator of two long-standing problems in the field: how dysregulated lipid metabolism relates to altered mitochondrial metabolism during MASLD, and what the innate changes are in the steatotic liver that cause the increased respiration. The authors identified reduced liver cardiolipin in humans with MASH and in a variety of mouse models with MASH. When they knocked out hepatic cardiolipin synthesis, mice developed steatosis and inflammation. These mice also recapitulated the elevated hepatic oxidative metabolism and oxidative stress found in obese humans with MASLD. Some of the in vivo functional data related to glucose homeostasis and substrate metabolism could be stronger, and interpretation of the in vitro flux data needs some clarification, but in both cases, the data are not essential to the main conclusions of the manuscript. Overall, the study offers compelling evidence that cardiolipin is reduced in MASLD and that impaired cardiolipin synthesis is sufficient to recapitulate many features of MASLD.

Strengths:

The main strengths of the study are:

(1) The identification of reduced cardiolipin levels in the liver of humans with MASLD and in a variety of mouse models of MASLD.

(2) The finding that loss of cardiolipin synthesis recapitulates steatosis and inflammation in MASH.

(3) The finding that loss of cardiolipin increases mitochondrial respiration, ROS production, and fat oxidation (in a separate hepatocyte cell line), again recapitulates several previous studies in obese humans with MASLD.

(4) Evidence, though less definitive, that cardiolipin deficiency promotes electron leak by disrupting respiratory supercomplexes and preventing CoQ reduction.

Weaknesses:

(1) Figure 3A-D tries to make the point that liver CLS KO causes defects in substrate handling in vivo, based on glucose and pyruvate tolerance tests. The KO mice have a blunted response to a glucose tolerance test, but the pyruvate tolerance test showed very little (almost no) effect on glucose levels in either WT or LKO mice. The small blunting of the response in the LKO is impossible to interpret (if it's real), since the ability to clear glucose is also increased, and no tracers were used. It might be useful to monitor pyruvate and lactate levels during the experiment. However, this reviewer doesn't think the data is essential to prove the authors' main points.

(2) After presenting convincing evidence that respiration is elevated in isolated mitochondria from CLS KO liver, the authors follow up the findings by investigating whether 13C-palmitate and 13C-glucose oxidation are altered by CLS knockdown in murine Hepa1-6 cells (Figure 4). A few comments are worth mentioning about Figure 4:

a. It is not clear why the authors chose to use a hepatoma cell line rather than primary hepatocytes from LKO mice. The latter would be more convincing, since there could be important differences in metabolism between hepatoma cells and hepatocytes (e.g., preference for fatty acids vs glucose). Nevertheless, I think the approach is sufficient to test the general effect of loss of CLS on substrate metabolism.

b. The authors use the M+2 enrichments of TCA cycle intermediates to infer rates of oxidation of [U-13C]palmitate or [U-13C]glucose. It is important to note that this kind of data reports fractional carbon sources (i.e., substrate preference) rather than rates of oxidation. For example, data from the 13C-palmitate experiment indicates that the CLS KD cells increase the fractional contribution from 13C palmitate (compared to glucose, for example) to the TCA cycle, but the actual rate of palmitate oxidation is not implicit in the data. However, it is reasonable to suggest that, in combination with the increased rates of O2 consumption observed in isolated mitochondria, this data supports increased fat oxidation.

c. I have some concern that the [U-13C]glucose experiment is more complicated to interpret than the description implies. I'm not sure what happens in this cell line, but in the liver, most labeling from pyruvate (i.e., originating from glucose in this case) enters the TCA cycle via pyruvate carboxylase, with smaller amounts entering via PDH (depending on the nutritional state). Since one could expect pyruvate carboxylase to contribute M+3 labeled TCA cycle intermediates initially, and M+2 on the first turn of the cycle, it's hard to conclude what the data indicates about glucose oxidation. The authors could generalize the conclusion by framing the TCA cycle enrichment data as the contribution of glucose carbons and noting in Figure 4A that pyruvate carbons can enter the TCA cycle via PDH or pyruvate carboxylase, without attempting to assign their relative contributions. There are better ways to do it, but it's a small nuance here since the authors aren't making a critical point about the pathways.

Reviewer #2 (Public review):

In this study, the authors show that alterations in the lipid composition of the inner mitochondrial membrane, particularly changes in cardiolipin (CL) content, lead to defects in electron transport, supercomplex formation, and oxidative stress. Using liver-specific CLS knockout mice, which are characterized by dysfunctional capacity for cardiolipin synthesis, the authors highlight an underappreciated role for CL in MASH pathology. Overall, this is an interesting study highlighting the importance of functional/physiological electron transport (and in this context, electron leakage) in MASH pathophysiology. Despite that, this manuscript has several weaknesses that require attention.

(1) For all LKO studies, it is stated that the decrease in hepatic CL is causal for the observed phenotype. However, it is evident that many other lipids are impacted by CLS KO, including a marked increase in hepatic PG. In this respect, the authors show no evidence that the observed metabolic phenotype is indeed due to the reduction in CL and not to other accompanying changes.

(2) In the results, the authors highlight that 'MASLD has been shown to alter the total cellular lipidome in liver.' Given that this study focused on CL, it would be useful to include specific studies that pointed to changes in hepatic CL content in MASLD/MASH/fibrosis.

(3) The initial human mitochondrial lipidomics studies show a reduction in mitochondrial CL and PG content. What was the content/expression of CL synthase and PGP synthase in these samples? If this cannot be assessed, is there any association of CLS or PGPS expression and MASLD/fibrosis (etc) in publicly available databases (e.g, GEP liver) that may explain the reduction in mitochondrial PG and CL content?

(4) The validation of MASH in patients (Figure 1B) is not convincing (ie., no quantification/scoring provided). NAS /fibrosis scoring (according to Kleiner) would help to define if all patients have indeed MASH, and what subset has fibrosis. Could the reduction in CL/PG content be (also) associated with fibrosis? In addition, Masson's Trichrome should be added to Figure 1B.

(5) In human lipidomics, the authors suggest that reductions are observed in tetralinoleoyl CL (Figure 1C). However, Figure 1C only shows the combined FA acyl chain length + unsaturation, therefore not allowing for FA-specific ID (unless such data are available from the LC/MS analysis).

(6) Figures 1 J/K/I. It is obvious that the background in all murine immunoblotting analysis has been altered. The authors should provide unaltered images for these immunoblots.

(7) For Figure 1, it is unclear what is meant by 'we performed all mitochondrial lipidomic analyses by quantifying lipids per mg of mitochondrial proteins'. Was the murine lipidomics carried out on fractionated mitochondria or whole liver? If whole liver, then how were the data corrected, particularly given that PG is not a mitochondria-specific lipid?

(8) While total CL content seems indeed decreased across the different mouse models, this is mostly due to 1-2 CL species showing a pronounced reduction, with the remainder being unaltered. This should at least be acknowledged in the results. This is similarly the case in the LKO livers.

(9) Figure 2. A secondary biochemical analysis of changes in lipid content should be provided, e.g., total triglyceride content, particularly given that the histology analysis does not show any major changes in hepatic lipid droplets/steatosis. In addition, the Masson's Trichrome staining shows almost no collagen deposition.

(10) Figure 3. 'CLS deletion modestly reduced glucose handling' should be reworded. The LKO mice show improved glucose tolerance (despite the MASH phenotype), which is not evident from the above wording.

(11) Looking at the mechanism behind the increase in hepatic steatosis, the authors state that lipid accumulation can occur due to increased lipogenesis, or dysfunctional VLDL secretion or beta oxidation, and subsequently assessed the relevant proteins/pathways. What about fatty acid uptake, which is also one of the four major pathways impacted in MASLD? This should be included in this assessment in Figure 3.

(12) For Figure 5A, it is simply stated 'CLS deletion promotes liver fibrosis in standard chow-fed condition', and it is unclear what is highlighted within the selected EM images and what the arrows refer to. The authors should clarify this within the text.

Reviewer #3 (Public review):

Summary:

Mitochondrial oxphos causes lipid accumulation, leading to MASH, although the mechanism has been poorly understood. In this study, Funai and colleagues identify that reductions in cardiolipin in the mitochondria cause disruptions in the electron transport chain. Knockout of cardiolipin synthase was sufficient to drive MASH phenotypes, increase respiratory capacity, and cause electron leak at complexes II and III. It is well established that loss of cardiolipin increases ROS. Studies to date have been performed on whole tissue lysates, but to rule out which changes in mitochondrial lipids are driven by changes in mitochondrial number versus lipid synthesis/turnover, the authors uniquely purified mitochondria from human and mouse livers in MASH and NASH models for this study. This study provides critical information to the field that will inevitably help us better understand the mechanisms underlying MASH and NASH onset. The evidence provided is both convincing and compelling. With further suggested revision experiments, this study has the potential to change our understanding of MASH and NASH pathogenesis.

Strengths:

The authors use a unique approach of lipidomics on purified mitochondria. They also analyze many distinct MASH models and provide a unique resource for the field of comprehensive lipidomics analysis of the different ways in which MASH can be induced. The use of human tissue elevates the impact/significance of the findings.

Weaknesses:

The data on the super complexes was the least compelling, and frankly, I do not think the authors needed those data to make a compelling argument! The authors should shift their focus more to the compelling electron leak data they have collected. If possible, it would also strengthen the work to include cardiolipin rescues on more of the experiments. Finally, expanding their explanations of the model systems would be very helpful for the readership.

Reviewer #4 (Public review):

Summary:

Here, the authors wish to shed light on factors that contribute to the development of liver disease in what used to be called 'the metabolic syndrome'. This is a human-health problem of considerable significance, and the insights they provide, namely the implication of a defect in mitochondrial cardiolipin (CL) content to the progression from metabolic dysfunction-associated steatotic liver disease to steatohepatitis, are plausible.

Strengths:

The experimental evidence proffered is derived from the observation of lower levels of (CL) in mitochondria from the liver of patients undergoing liver transplant or resection due to end-stage steatohepatitis compared with mitochondria derived from livers of patients with other conditions. This correlation is buttressed by observations made in mice with liver-selective compromise in CL synthesis and which suggest a pathological environment associated with mitochondrial dysfunction and enhanced oxidative stress, features deemed to play a role in the progression from steatotic liver disease to steatohepatitis.

The paper is well written, and the findings are well explained and superficially convincing.

Weaknesses:

It is unclear how much can be learned from compromising a key enzyme that produces a key mitochondrial lipid in a busy metabolic organ like the liver - isn't the discovery of a mitochondrial defect in such a context rather trivial? And how reliably can these findings be related to the human observations? Most importantly, the chain of causality implied by the title is unproven: the key question of whether or not (somehow) preventing the drop in cardiolipin content affects the course of steatohepatitis remains unanswered.

eLife Assessment

This important manuscript investigates the role of olfactory cues in Pieris brassicae larvae, focusing on their interactions with the host plant Brassica oleracea and the parasitoid wasp Cotesia glomerata. The authors' demonstration that impaired olfactory perception reduces caterpillar performance and increases susceptibility to parasitism is convincing. These findings highlight the ecological significance of olfaction in mediating feeding behavior and predator avoidance in herbivorous insects.

Reviewer #1 (Public review):

Summary:

The manuscript focuses on the olfactory system of Pieris brassicae larvae and the importance of olfactory information in their interactions with the host plant Brassica oleracea and the major parasitic wasp Cotesia glomerata. The authors used CRISPR/Cas9 to knockout odorant receptor co-receptors (Orco), and conducted a comparative study on the behavior and olfactory system of the mutant and wild-type larvae. The study found that Orco-expressing olfactory sensory neurons in antennae and maxillary palps of Orco knockout (KO) larvae disappeared, and the number of glomeruli in the brain decreased, which impairs the olfactory detection and primary processing in the brain. Orco KO caterpillars show weight loss and loss of preference for optimal food plants; KO larvae also lost weight when attacked by parasitoids with the ovipositor removed, and mortality increased when attacked by untreated parasitoids. On this basis, the authors further studied the responses of caterpillars to volatiles from plants attacked by the larvae of the same species and volatiles from plants on which the caterpillars were themselves attacked by parasitic wasps. Lack of OR mediated olfactory inputs prevents caterpillars from finding suitable food sources and from choosing spaces free of enemies.

Strengths:

The findings help to understand the important role of olfaction in caterpillar feeding and predator avoidance, highlighting the importance of odorant receptor genes in shaping ecological interactions.

Weaknesses:

There are the following major concerns:

(1) Possible non-targeted effects of Orco knockout using CRISPR/Cas9 should be analyzed and evaluated in Materials and Methods and Results

(2) Figure 1E: Only one olfactory receptor neuron was marked in WT. There are at least three olfactory sensilla at the top of the maxillary palp. Therefore, to explain the loss of Orco expressing neurons in the mutant (Figure 1F), a more rigorous explanation of the photo is required.

(3) In Figure 1G, H, the four glomeruli circled by dotted lines: their corresponding relationship between the two figures needs to be further clarified.

(4) Line 130: Since the main topic in this study is the olfactory system of larvae, the experimental results of this part are all about antennal electrophysiological responses, mating frequency and egg production of female and male adults of wild type and Orco KO mutant, it may be considered to include this part in the supplementary files. It is better to include some data about the olfactory responses of larvae.

(5) Line 166: The sentences in the text is about the choice test between " healthy plant vs. infested plant", while in Fig 3C, it is "infested plant vs. no plant". The content in the text does not match the figure.

(6) Lines 174-178: Fig 3A showed that the body weight of Orco KO larvae in the absence of parasitic wasps also decreased compared with that of WT. Therefore, in the experiments of Fig 3A and E, the difference in the body weight of Orco KO larvae in the presence or absence of parasitic wasps without ovipositors should also be compared. The current data cannot determine the reduced weight of KO mutant is due to the Orco knockout or the presence of parasitic wasps.

(7) Lines 179-181: Fig 3F show that the survival rate of larvae of Orco KO mutant decreased in the presence of parasitic wasps, and the difference in survival rate of larvae of WT and Orco KO mutant in the absence of parasitic wasps should also be compared. The current data cannot determine whether the reduced survival of the KO mutant is due to the Orco knockout or the presence of parasitic wasps.

(8) In Figure 4B, why do the compounds tested had no volatiles derived from plants? Cruciferous plants have the well-known mustard bomb. In the behavioral experiments the larvae responses to ITC compounds were not included, which is suggested to be explained in the discussion section.

(9) The custom-made setup and the relevant behavioral experiments in Fig 4C needs to be described in detail (Line 545).

(10) Materials and Methods Line 448: 10 μL paraffin oil should be used for negative control.

Comments on revised version:

The authors have replied my concerns and made revisions accordingly.

Author response:

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

Public Reviews:

Reviewer #1 (Public review):

Summary:

The manuscript focuses on the olfactory system of Pieris brassicae larvae and the importance of olfactory information in their interactions with the host plant Brassica oleracea and the major parasitic wasp Cotesia glomerata. The authors used CRISPR/Cas9 to knockout odorant receptor coreceptors (Orco), and conducted a comparative study on the behavior and olfactory system of the mutant and wild-type larvae. The study found that Orco-expressing olfactory sensory neurons in antennae and maxillary palps of Orco knockout (KO) larvae disappeared, and the number of glomeruli in the brain decreased, which impairs the olfactory detection and primary processing in the brain. Orco KO caterpillars show weight loss and loss of preference for optimal food plants; KO larvae also lost weight when attacked by parasitoids with the ovipositor removed, and mortality increased when attacked by untreated parasitoids. On this basis, the authors further studied the responses of caterpillars to volatiles from plants attacked by the larvae of the same species and volatiles from plants on which the caterpillars were themselves attacked by parasitic wasps. Lack of OR-mediated olfactory inputs prevents caterpillars from finding suitable food sources and from choosing spaces free of enemies.

Strengths:

The findings help to understand the important role of olfaction in caterpillar feeding and predator avoidance, highlighting the importance of odorant receptor genes in shaping ecological interactions.

Weaknesses:

There are the following major concerns:

(1) Possible non-targeted effects of Orco knockout using CRISPR/Cas9 should be analyzed and evaluated in Materials and Methods and Results.

Thank you for your suggestion. In the Materials and Methods, we mention how we selected the target region and evaluated potential off-target sites by Exonerate and CHOPCHOP. Neither of these methods found potential off-target sites with a more-than-17-nt alignment identity. Therefore, we assumed no off-target effect in our Orco knockout. Furthermore, we did not find any developmental differences between wildtype and knockout caterpillars when these were reared on leaf discs in Petri dishes (Fig S4). We will further highlight this information on the off-target evaluation in the Results section.

(2) Figure 1E: Only one olfactory receptor neuron was marked in WT. There are at least three olfactory sensilla at the top of the maxillary palp. Therefore, to explain the loss of Orcoexpressing neurons in the mutant (Figure 1F), a more rigorous explanation of the photo is required.

Thank you for pointing this out. The figure shows only a qualitative comparison between WT and KO and we did not aim to determine the total number of Orco positive neurons in the maxillary palps or antennae of WT and KO caterpillars, but please see our previous work for the neuron numbers in the caterpillar antennae (Wang et al., 2024). We did indeed find more than one neuron in the maxillary palps, but as these were in very different image planes it was not possible to visualize them together. However, we will add a few sentences in the Results and Discussion section to explain the results of the maxillary palp Orco staining.

(3) In Figure 1G, H, the four glomeruli are circled by dotted lines: their corresponding relationship between the two figures needs to be further clarified.

Thank you for pointing this out. The four glomeruli in Figure 1G and 1H are not strictly corresponding. We circled these glomeruli to highlight them, as they are the best visualized and clearly shown in this view. In this study, we only counted the number of glomeruli in both WT and KO, however, we did not clarify which glomeruli are missing in the KO caterpillar brain. We will further clarify this in the figure legend.

(4) Line 130: Since the main topic in this study is the olfactory system of larvae, the experimental results of this part are all about antennal electrophysiological responses, mating frequency, and egg production of female and male adults of wild type and Orco KO mutant, it may be considered to include this part in the supplementary files. It is better to include some data about the olfactory responses of larvae.

Thank you for your suggestion. We do agree with your suggestion, and we will consider moving this part to the supplementary information. Regarding larval olfactory response, we unfortunately failed to record any spikes using single sensillum recordings due to the difficult nature of the preparation; however we do believe that this would be an interesting avenue for further research.

(5)Line 166: The sentences in the text are about the choice test between " healthy plant vs. infested plant", while in Fig 3C, it is "infested plant vs. no plant". The content in the text does not match the figure.

Thank you for pointing this out. The sentence is “We compared the behaviors of both WT and Orco KO caterpillars in response to clean air, a healthy plant and a caterpillar-infested plant”. We tested these three stimuli in two comparisons: healthy plant vs no plant, infested plant vs no plant. The two comparisons are shown in Figure 3C separately. We will aim to describe this more clearly in the revised version of this manuscript.

(6) Lines 174-178: Figure 3A showed that the body weight of Orco KO larvae in the absence of parasitic wasps also decreased compared with that of WT. Therefore, in the experiments of Figure 3A and E, the difference in the body weight of Orco KO larvae in the presence or absence of parasitic wasps without ovipositors should also be compared. The current data cannot determine the reduced weight of KO mutant is due to the Orco knockout or the presence of parasitic wasps.

Thank you for pointing this out. We did not make a comparison between the data of Figures 3A and 3E since the two experiments were not conducted at the same time due to the limited space in our BioSafety III greenhouse. We do agree that the weight decrease in Figure 3E is partly due to the reduced caterpillar growth shown in Figure 3A. However, we are confident that the additional decrease in caterpillar weight shown in Figure 3E is mainly driven by the presence of disarmed parasitoids. To be specific, the average weight in Figure 3A is 0.4544 g for WT and 0.4230 g for KO, KO weight is 93.1% of WT caterpillars. While in Figure 3E, the average weight is 0.4273 g for WT and 0.3637 g for KO, KO weight is 85.1% of WT caterpillars. We will discuss this interaction between caterpillar growth and the effect of the parasitoid attacks more extensively in the revised version of the manuscript.

(7) Lines 179-181: Figure 3F shows that the survival rate of larvae of Orco KO mutant decreased in the presence of parasitic wasps, and the difference in survival rate of larvae of WT and Orco KO mutant in the absence of parasitic wasps should also be compared. The current data cannot determine whether the reduced survival of the KO mutant is due to the Orco knockout or the presence of parasitic wasps.

We are happy that you highlight this point. When conducting these experiments, we selected groups of caterpillars and carefully placed them on a leaf with minimal disturbance of the caterpillars, which minimized hurting and mortality. We did test the survival of caterpillars in the absence of parasitoid wasps from the experiment presented in Figure 3A, although this was missing from the manuscript. There is no significant difference in the survival rate of caterpillars between the two genotypes in the absence of wasps (average mortality WT = 8.8 %, average mortality KO = 2.9 %; P = 0.088, Wilcoxon test), so the decreased survival rate is most likely due to the attack of the wasps. We will add this information to the revised version of the manuscript.

(8) In Figure 4B, why do the compounds tested have no volatiles derived from plants? Cruciferous plants have the well-known mustard bomb. In the behavioral experiments, the larvae responses to ITC compounds were not included, which is suggested to be explained in the discussion section.

Thank you for the suggestion. We assume you mean Figure 4D/4E instead of Figure 4B. In Figure 4B, many of the identified chemical compounds are essentially plant volatiles, especially those from caterpillar frass and caterpillar spit. In Figure 4D/4E, most of the tested chemicals are derived from plants. But indeed, we did not include ITCs, based on information from the EAG results in Figures 2A & 2B. Butterfly antennae did not respond strongly to ITCs, so we did not include ITCs in the larval behavioural tests. Instead, the tested chemicals in Figure 4D/4E either elicit high EAG responses of butterflies or have been identified as “important” by VIP scores in the chemical analyses. In the EAG results of Plutella xylostella (Liu et al., 2020), moths responded well to a few ITCs, the tested ITCs in our study are actually adopted from this study except for those that were not available to us. However, butterflies did not show a strong response to the tested ITCs; therefore, we did not include ITCs because we expected that Pieris brassicae caterpillars are not likely to show good responses to ITCs. We will add this explanation to the revised version of our manuscript.

(9) The custom-made setup and the relevant behavioral experiments in Figure 4C need to be described in detail (Line 545).

We will add more detailed descriptions for the setup and method in the Materials and Methods.

(10) Materials and Methods Line 448: 10 μL paraffin oil should be used for negative control.

Thank you for pointing this out. We used both clean filter paper and clean filter paper with 10 μL paraffin oil as negative controls, but we did not find a significant difference between the two controls. Therefore, in the EAG results of Figure 2A/2B, we presented paraffin oil as one of the tested chemicals. We will re-run our statistical tests with paraffin oil as negative control, although we do not expect any major differences to the previous tests.

Reviewer #2 (Public review):

Summary:

This manuscript investigated the effect of olfactory cues on caterpillar performance and parasitoid avoidance in Pieris brassicae. The authors knocked out Orco to produce caterpillars with significantly reduced olfactory perception. These caterpillars showed reduced performance and increased susceptibility to a parasitoid wasp.

Strengths:

This is an impressive piece of work and a well-written manuscript. The authors have used multiple techniques to investigate not only the effect of the loss of olfactory cues on host-parasitoid interactions, but also the mechanisms underlying this.

Weaknesses:

(1) I do have one major query regarding this manuscript - I agree that the results of the caterpillar choice tests in a y-maze give weight to the idea that olfactory cues may help them avoid areas with higher numbers of parasitoids. However, the experiments with parasitoids were carried out on a single plant. Given that caterpillars in these experiments were very limited in their potential movement and source of food - how likely is it that avoidance played a role in the results seen from these experiments, as opposed to simply the slower growth of the KO caterpillars extending their period of susceptibility? While the two mechanisms may well both take place in nature - only one suggests a direct role of olfaction in enemy avoidance at this life stage, while the other is an indirect effect, hence the distinction is important.

We do agree with your comment that both mechanisms may be at work in nature and we do address this in the Discussion section. In our study, we did find that wildtype caterpillars were more efficient in locating their food source and did grow faster on full plants than knockout caterpillars. This faster growth will enable wildtype caterpillars to more quickly outgrow the life-stages most vulnerable to the parasitoids (L1 and L2). The olfactory system therefore supports the escape from parasitoids indirectly by enhancing feeding efficiency directly.

Figure 3D shows that WT caterpillars prefer infested plants without parastioids to infested plants with parasitoids. In addition, we observed that caterpillars move frequently between different leaves. Therefore, we speculate that WT caterpillars make use of volatiles from the plant or from (parasitoid-exposed) conspecifics via their spit or faeces to avoid parts of the plant potentially attracting natural enemies. Knockout caterpillars are unable to use these volatile danger cues and therefore do not avoid plant parts that are most attractive to their natural enemies, making KO caterpillars more susceptible and leading to more natural enemy harassment. Through this, olfaction also directly impacts the ability of a caterpillar to find an enemy-free feeding site.

We think that olfaction supports the enemy avoidance of caterpillars via both these mechanisms, although at different time scales. Unfortunately, our analysis was not detailed enough to discern the relative importance of the two mechanisms we found. However, we feel that this would be an interesting avenue for further research. Moreover, we will sharpen our discussion on the potential importance of the two different mechanisms in the revised version of this manuscript.

(2) My other issue was determining sample sizes used from the text was sometimes a bit confusing. (This was much clearer from the figures).

We will revise the sample size in the text to make it more clear.

(3) I also couldn't find the test statistics for any of the statistical methods in the main text, or in the supplementary materials.

Thank you for pointing this out. We will provide more detailed test statistics in the main text and in the supplementary materials of the revised version of the manuscript.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

(1) Abstract

Line 24: "optimal food plant" should be changed to "optimal food plants"

Thank you for the suggestion, we will revise it.

(2) Introduction

Lines 44-46: The sentence should be rephrased.

Thank you for the suggestion, we will revise it.

Line 50: "are" should be changed to "is".

Thank you for the suggestion, we will revise it.

Lines 57 and 58: Please provide the Latin names of "brown planthoppers" and "striped stem borer".

Thank you for the suggestion, we will revise it.

Line 85: "investigate the influence of odor-guided behavior by this primary herbivore on the next trophic levels"; similarly, Line 160: "investigate if caterpillars could locate the optimal host-plant when supplied with differently treated plants". These sentences are not very accurate in describing the relevant experiments. A: Thank you for the suggestion, we will revise them.

Reviewer #2 (Recommendations for the authors):

(1) L53 Remove the "the" from "Under the strong selection pressure"

Thank you for the suggestion, we will revise it.

(2) L80 I suggest adding a reference for the spitting behaviour, e.g. Muller et al 2003.

Thank you for the suggestion, we will add it.

(3) L89 establishing a homozygous KO insect colony.

Thank you for the suggestion, we will revise it.

(4) L107 perhaps this goes against the journal style but I always like to see acronyms explained the first time they are used.

Thank you for the suggestion, we will try to make it more understandable.

(5) L146-148 sentence difficult to read - consider rephrasing.

Thank you for the suggestion, we will revise it.

(6) L230 do you mean still produce? Rather than still reproduce?

Thank you for the suggestion, we will revise it.

(7) L233 missing an and before "a greater vulnerability to the parasitoid wasp".

Thank you for pointing this out, we will revise it.

(8) L238 malfunctional is a strange word choice.

Thank you for pointing this out, we will revise it.

(9) L181 - can the authors confirm that this lower survival was due to parasitism by the wasps?

This question is similar to Q(7) of Reviewer 1, so we quote our answer for Q(7) here:

When conducting these experiments, we selected groups of caterpillars and carefully placed them on a leaf with minimal disturbance of the caterpillars, which minimized hurting and mortality. We did test the survival of caterpillars in the absence of parasitoid wasps from the experiment presented in Figure 3A, although this was missing from the manuscript. There is no significant difference in the survival rate of caterpillars between the two genotypes in the absence of wasp (average mortality WT = 8.8 %, average mortality KO = 2.9 %; P = 0.088, Wilcoxon test), so the decreased survival rate is most likely due to the attack of the wasps. We will add this information to the revised version of the manuscript.

(10) L474 - has it been tested if wasps still behave similarly after their ovipositor has been removed?

Thank you for pointing out this issue. We did not strictly compare if disarmed and untreated wasps have similar behaviors. However, we did observe if disarmed wasps can actively move or fly after recovering from anesthesia before releasing into a cage, otherwise we would replace with another active one.

eLife Assessment

This is an important study that characterized proteins associated with electrical synapses in zebrafish and mouse retinal neurons using proximity labeling approaches, complemented by biochemical and histological validations. The resulting protein interactome datasets are convincing and reveal novel scaffold proteins at the electrical synapse. Additional quantification and validation would strengthen the work further.

Reviewer #1 (Public review):

This study aims to identify the proteins that compose the electrical synapse, which are much less understood than those of the chemical synapse. Identifying these proteins is important to understand how synaptogenesis and conductance are regulated in these synapses.

Using a proteomics approach, the authors identified more than 50 new proteins and used immunoprecipitation and immunostaining to validate their interaction of localization. One new protein, a scaffolding protein (Sipa1l3), shows particularly strong evidence of being an integral component of the electrical synapse. The function of Sipa1l3 remains to be determined.

Another strength is the use of two different model organisms (zebrafish and mice) to determine which components are conserved across species. This approach also expands the utility of this work to benefit researchers working with both species.

The methodology is robust and there is compelling evidence supporting the findings.

Reviewer #2 (Public review):

Summary:

This study aimed to uncover the protein composition and evolutionary conservation of electrical synapses in retinal neurons. The authors employed two complementary BioID approaches: expressing a Cx35b-TurboID fusion protein in zebrafish photoreceptors and using GFP-directed TurboID in Cx36-EGFP-labeled mouse AII amacrine cells. They identified conserved ZO proteins and endocytosis components in both species, along with over 50 novel proteins related to adhesion, cytoskeleton remodeling, membrane trafficking, and chemical synapses. Through a series of validation studies¬-including immunohistochemistry, in vitro interaction assays, and immunoprecipitation-they demonstrate that novel scaffold protein SIPA1L3 interacts with both Cx36 and ZO proteins at electrical synapse. Furthermore, they identify and localize proteins ZO-1, ZO-2, CGN, SIPA1L3, Syt4, SJ2BP, and BAI1 at AII/cone bipolar cell gap junctions.

Strengths:

The study demonstrates several significant strengths in both experimental design and validation approaches. First, the dual-species approach provides valuable insights into the evolutionary conservation of electrical synapse components across vertebrates. Second, the authors compare two different TurboID strategies in mice and demonstrate that the HKamac promoter and GFP-directed approach can successfully target the electrical synapse proteome of mouse AII amacrine cells. Third, they employed multiple complementary validation approaches-including retinal section immunohistochemistry, in vitro interaction assays, and immunoprecipitation-providing evidence supporting the presence and interaction of these proteins at electrical synapses.

Weaknesses:

The major weakness of this paper is the insufficient number of replicates in the proteomics datasets. The zebrafish datasets include only two biological replicates, while the mouse dataset has only one high-quality replicate. Due to the limited number of replicates, it is not possible to determine which enriched proteins are statistically significant.

Additionally, the Neutravidin staining in the TurboID condition is not restricted to where Cx35 is expressed but is broadly distributed throughout the INL and IPL in the zebrafish retina (Figure 1B, bottom). Therefore, it is necessary to include NeutrAvidin staining in non-labeled retinas to verify whether the biotinylated proteins are specifically associated with Cx35 expression. Although the western blot results showed increased protein enrichment in the TurboID condition compared to non-labeled retinas, this does not confirm that the streptavidin pull-down proteins are associated with Cx35.

Similarly, it is important to include NeutrAvidin staining in both TurboID and non-labeled conditions in the mouse retina to verify that the biotinylated proteins are specifically associated with gap junctions.

Reviewer #3 (Public review):

Summary:

This study by Tetenborg S et al. identifies proteins that are physically closely associated with gap junctions in retinal neurons of mice and zebrafish using BioID, a technique that labels and isolates proteins in proximal to a protein of interest. These proteins include scaffold proteins, adhesion molecules, chemical synapse proteins, components of the endocytic machinery, and cytoskeleton-associated proteins. Using a combination of genetic tools and meticulously executed immunostaining, the authors further verified the colocalizations of some of the identified proteins with connexin-positive gap junctions. The findings in this study highlight the complexity of gap junctions. Electrical synapses are abundant in the nervous system, yet their regulatory mechanisms are far less understood than those of chemical synapses. This work will provide valuable information for future studies aiming to elucidate the regulatory mechanisms essential for the function of neural circuits.

Strengths:

A key strength of this work is the identification of novel gap junction-associated proteins in AII amacrine cells and photoreceptors using BioID in combination with various genetic tools. The well-studied functions of gap junctions in these neurons will facilitate future research into the functions of the identified proteins in regulating electrical synapses.

The authors have addressed my concerns in the revised manuscript.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public review):

Summary:

This study aims to identify the proteins that compose the electrical synapse, which are much less understood than those of the chemical synapse. Identifying these proteins is important to understand how synaptogenesis and conductance are regulated in these synapses. The authors identified more than 50 new proteins and used immunoprecipitation and immunostaining to validate their interaction of localization. One new protein, a scaffolding protein, shows particularly strong evidence of being an integral component of the electrical synapse. However, many key experimental details are missing (e.g. mass spectrometry), making it difficult to assess the strength of the evidence.

Strengths:

One newly identified protein, SIPA1L3, has been validated both by immunoprecipitation and immunohistochemistry. The localization at the electrical synapse is very striking.<br /> A large number of candidate interacting proteins were validated with immunostaining in vivo or in vitro.

Weaknesses:

There is no systematic comparison between the zebrafish and mouse proteome. The claim that there is "a high degree of evolutionary conservation" was not substantiated.

We have added a table as supplementary figure 3 that shows a comparison of all candidates. While there are differences in both proteomes, components such as ZO proteins and the endocytosis machinery are clearly conserved.

No description of how mass spectrometry was done and what type of validation was done.

We have contacted the mass spec facility we worked with and added a paragraph explaining the mass spec. procedure in the material and methods section.

The threshold for enrichment seems arbitrary.

Yes, the thresholds are somewhat arbitrary. This is due to the fact that experiments that captured larger total amounts of protein (mouse retina samples) had higher signal-to-noise ratio than those that captured smaller total amounts of protein (zebrafish retina). This allowed us to use a more stringent threshold in the mouse dataset to focus on high probability captured proteins.

Inconsistent nomenclature and punctuation usage.

We have scanned through the manuscript and updated terms that were used inconsistently in the interim revision of the manuscript.

The description of figures is very sparse and error-prone (e.g. Figure 6).

In Figure 1B, there is very broad non-specific labeling by avidin in zebrafish (In contrast to the more specific avidin binding in mice, Figure 2B). How are the authors certain that the enrichment is specific at the electrical synapse?

The enrichment of the proteins we identified is specific for electrical synapses because we compared the abundance of all candidates between Cx35b-V5-TurboID and wildtype retinas. Proteins that are components of electrical synapses, will only show up in the Cx35b-V5-TurboID condition. The western blot (Strep-HRP) in figure 1C shows the differences in the streptavidin labeling and hence the enrichment of proteins that are part of electrical synapses. Moreover, while the background appears to be quite abundant in sections, biotinylation is a rare posttranslational modification and mainly occurs in carboxylases: The two intense bands that show up above 50 and 75 kDa. The background mainly originates from these two proteins. Therefore, it is easy to distinguish specific hits from non-specific background.

In Figure 1E, there is very little colocalization between Cx35 and Cx34.7. More quantification is needed to show that it is indeed "frequently associated."

We agree that “frequently associated” is too strong as a statement. We corrected this and instead wrote “that Cx34.7 was only expressed in the outer plexiform layer (OPL) where it was associated with Cx35b at some gap junctions” in line 151. There are many gap junctions at which Cx35b is not colocalized with Cx34.7.

Expression of GFP in HCs would potentially be an issue, since GFP is fused to Cx36 (regardless of whether HC expresses Cx36 endogenously) and V5-TurboID-dGBP can bind to GFP and biotinylate any adjacent protein.

Thank you for this suggestion! There should be no Cx36-GFP expression in horizontal cells, which means that the nanobody cannot bind to anything in these cells. Moreover, to recognize specific signals from non-specific background, we included wild type retinas throughout the entire experiments. This condition controls for non-specific biotinylation.

Figure 7: the description does not match up with the figure regarding ZO-1 and ZO-2.

It appears that a portion of the figure legend was left out of the submitted version of the manuscript. We have put the legend for panels A through C back into the manuscript in the interim revision.

Reviewer #2 (Public review):

Summary:

This study aimed to uncover the protein composition and evolutionary conservation of electrical synapses in retinal neurons. The authors employed two complementary BioID approaches: expressing a Cx35b-TurboID fusion protein in zebrafish photoreceptors and using GFP-directed TurboID in Cx36-EGFP-labeled mouse AII amacrine cells. They identified conserved ZO proteins and endocytosis components in both species, along with over 50 novel proteins related to adhesion, cytoskeleton remodeling, membrane trafficking, and chemical synapses. Through a series of validation studies¬-including immunohistochemistry, in vitro interaction assays, and immunoprecipitation - they demonstrate that novel scaffold protein SIPA1L3 interacts with both Cx36 and ZO proteins at electrical synapse. Furthermore, they identify and localize proteins ZO-1, ZO-2, CGN, SIPA1L3, Syt4, SJ2BP, and BAI1 at AII/cone bipolar cell gap junctions.

Strengths:

The study demonstrates several significant strengths in both experimental design and validation approaches. First, the dual-species approach provides valuable insights into the evolutionary conservation of electrical synapse components across vertebrates. Second, the authors compare two different TurboID strategies in mice and demonstrate that the HKamac promoter and GFP-directed approach can successfully target the electrical synapse proteome of mouse AII amacrine cells. Third, they employed multiple complementary validation approaches - including retinal section immunohistochemistry, in vitro interaction assays, and immunoprecipitation-providing evidence supporting the presence and interaction of these proteins at electrical synapses.

Weaknesses:

The conclusions of this paper are supported by data; however, some aspects of the quantitative proteomics analysis require clarification and more detailed documented. The differential threshold criteria (>3 log2 fold for mouse vs >1 log2 fold for zebrafish) will benefit from biological justification, particularly given the cross-species comparison. Additionally, providing details on the number of biological or technical replicates used in this study, along with analyses of how these replicates compare to each other, would strengthen the confidence in the identification of candidate proteins. Furthermore, including negative controls for the histological validation of proteins interacting with Cx36 could increase the reliability of the staining results.

While the study successfully characterized the presence of candidate proteins at the electrical synapses between AII amacrine cells and cone bipolar cells, it did not compare protein compositions between the different types of electrical synapses within the circuit. Given that AII amacrine cells form both homologous (AII-AII) and heterologous (AII-cone bipolar cell) electrical synapses-connections that serve distinct functional roles in retinal signaling processing-a comparative analysis of their molecular compositions could have provided important insights into synapse specificity.

Reviewer #3 (Public review):

Summary:

This study by Tetenborg S et al. identifies proteins that are physically closely associated with gap junctions in retinal neurons of mice and zebrafish using BioID, a technique that labels and isolates proteins proximal to a protein of interest. These proteins include scaffold proteins, adhesion molecules, chemical synapse proteins, components of the endocytic machinery, and cytoskeleton-associated proteins. Using a combination of genetic tools and meticulously executed immunostaining, the authors further verified the colocalizations of some of the identified proteins with connexin-positive gap junctions. The findings in this study highlight the complexity of gap junctions. Electrical synapses are abundant in the nervous system, yet their regulatory mechanisms are far less understood than those of chemical synapses. This work will provide valuable information for future studies aiming to elucidate the regulatory mechanisms essential for the function of neural circuits.

Strengths:

A key strength of this work is the identification of novel gap junction-associated proteins in AII amacrine cells and photoreceptors using BioID in combination with various genetic tools. The well-studied functions of gap junctions in these neurons will facilitate future research into the functions of the identified proteins in regulating electrical synapses.

Thank you for these comments.

Weaknesses:

I do not see major weaknesses in this paper. A minor point is that, although the immunostaining in this study is beautifully executed, the quantification to verify the colocalization of the identified proteins with gap junctions is missing. In particular, endocytosis component proteins are abundant in the IPL, making it unclear whether their colocalization with gap junction is above chance level (e.g. EPS15l1, HIP1R, SNAP91, ITSN in Figure 3B).

Recommendations for the authors:

Reviewer #2 (Recommendations for the authors):

(1) It would be helpful to include a comprehensive summary of the results from the quantitative proteomics analyses, such as the number of proteins detected in each species and the number of proteins associated with each GO term. Additionally, a clear figure or table highlighting the specific proteins conserved between zebrafish and mice would strengthen the evidence for evolutionary conservation of proteins at electrical synapses.

We have added the raw data we received from our mass spec facility including a comparison of all the candidates for different species. Supplementary figure 3.

(2) A more detailed description of the number of experimental and/or technical replicates would improve the technical rigor of the study. For example, what was the rationale for using different log2 fold-change cutoffs in mice versus zebrafish? Are the replicates consistent in terms of protein enrichment?

We have added raw data from individual experiments as a supplement (Excel spreadsheet). We have two replicates from zebrafish and two from mice. The first experiment in mice was conducted with fewer retinas and a different promoter (human synapsin promoter) and didn’t yield nearly as many candidates. We are currently running a third experiment with 35 mouse retinas which will most likely detect more candidates as we have identified currently. We can update the proteome in this paper once the analysis is complete. It is not feasible to conduct these experiments with multiple replicates at the same time, since the number of animals that have to be used is simply too high, especially since very specific genotypes are required that are difficult obtain.

(3) It would be interesting to determine whether there are differences in the presence of candidate proteins between AII-AII gap junctions and AII-cone bipolar cell gap junctions. Given that the subcellular localization of AII-AII gap junctions differs from that of AII-cone bipolar cell gap junctions (with most AII-AII gap junctions located below AII-cone ones), histological validations of the proteins shown in Figure 6 can be repeated for AII-AII gap junctions. This would help reveal similarities or differences in the protein compositions of these two types of gap junctions.

Thank you for this suggestion. We had similar plans. However, we realized that homologous gap junctions are difficult to recognize with GFP. The dense GFP labeling in the proximal IPL, where AII-AII gap junctions are formed, does not allow us to clearly trace the location of individual dendrites from different cells. Detecting AII-AII gap junctions would require intracellular dye Injections of neighboring AII cells. Unfortunately, we don’t have a set up that would allow this. Bipolar cell terminals, on the contrary, are a lot easier to detect with markers such as SCGN, which is why we decided to focus on AII/ONCB gap junctions.

(4) In Figures 1 and 2, it would be helpful to clarify in the figure legends whether the proteins in the interaction networks represent all detected proteins or only those selected based on log2 fold-change or other criteria.

Thank you for this suggestion! We have added a description in lines 643 and 662.

(5) In Figure 1A (bottom panel), please include a negative control for the Neutravidin staining result from the non-labeling group.

We only tested the biotinylation for wild type retinas in cell lysates and western blots as shown in figure 1C, which shows an entirely different biotinylation pattern.

(6) In Figure 2B, please include the results of Neutravidin staining for both the labeling and non-labeling groups.

Same comment: We see the differences in the biotinylation pattern on western blots, which is distinct for Cx36-EGFP and wild type retinas, although both genotypes were injected with the same AAV construct and the same dose of biotin. We hope that this provides sufficient evidence for the specificity of our approach.

(7) In Figure 5B, the sizes of multiple proteins detected by Western blotting are inconsistent and confusing. For example, the size of Cx36 in the "FLAG-SJ2BP" panel differs from that in the other three panels. Additionally, in the "Myc-SIPA1L3+" panel, the size of SIPA1l3 appears different between the input and IP conditions.

Thank you for pointing this out! The differences in the molecular weight can be explained by dimerization. We have indicated the position of the dimer and the monomer bands with arrows. Especially, when larger amounts of Cx36 are coprecipitated Cx36 preferentially occurs as a dimer. This can also be seen in our previous publication:

S. Tetenborg et al., Regulation of Cx36 trafficking through the early secretory pathway by COPII cargo receptors and Grasp55. Cellular and Molecular Life Sciences 81, 1-17 (2024). Figure 1D

The band that occurs above 150kDa in the SIPA1L3 input is most likely a non-specific product. The specific band for SIPA1L3 can be seen in the IP sample, which has the appropriate molecular weight. We often see much better immuno reactivity for the protein of interest in IP samples, because the protein is concentrated in these experiments which facilitates its detection.

(8) How specific are the antibodies used for validating the proteins in this study? Given that many proteins, such as EPS15l1, HIP1R, SNAP91, GPrin1, SJ2BP, Syt4, show broad distribution in the IPL (Figure 3B, 4A, 6D), it is important to validate the specificity of these antibodies. Additionally, including negative controls in the histological validation would strengthen the reliability of the results.

We carefully selected the antibodies based on western blot data, that confirmed that each antibody detected an antigen of appropriate size. Moreover, the distribution of the proteins mentioned is consistent with function of each protein described in the literature. EPS15L1 and GPrin1 for instance are both membrane-associated, which is evident in Hek cells. Figure 5C.

A true negative control would require KO tissue and we don’t think that this is feasible at this point.

(9) In Figure 7F, the model could be improved by highlighting which components may be conserved between zebrafish and mice, as well as which components are conserved between the AII-AII junction and AII-cone bipolar cell junction?

Thank you for this suggestion. However, we don’t think that this is necessary as our study primarily focuses on the AII amacrine cell.

Currently we are unable to distinguish differences in the composition of AII-AII and AII-ONCB junctions as described above.

(10) Are there any functional measurements that could support the conclusion that "loss of Cx36 resulted in a quantitative defect in the formation of electrical synapse density complex"?

The loss of electrical synapse density proteins is shown by these immunostaining comparisons. Functional measurements necessarily depend on the function of the electrical synapse itself, which is gone in the case of the Cx36 KO. It is not clear that a different functional measurement can be devised.

Reviewer #3 (Recommendations for the authors):

(1) It would be very helpful if there were page and line numbers on the manuscript.

Line and page numbers have been added.

(2) Typos in the 3rd paragraph, the sentence 'which is triggered by the influx of Calcium though non-synaptic NMDA...'

Should it read '... Calcium THROUGH non-synaptic NMDA'?

We have corrected this typo.

(3) Figure 1B: please add a description of the top panels, 'Cx36 S293'.

A description of the top panels has been added to the figure legend in line. Line 639.

(4) Figure 1C: what do the arrows indicate?

We apologize for the confusion. The arrows in the western blot indicate the position of the Cx35-V5-TurboID construct, which can be detected with streptavidin-HRP and the V5 antibody. We have added a description for these arrows to the figure legend. See line 641.

(5) Related to the point in the 'Weakness', there are some descriptions of how well some of the gap junction-associated proteins colocalize with Cx36 in immunostaining. For example, 'In comparison to the scaffold proteins, however, the colocalization of Cx36 with each of these endocytic components, was clearly less frequent and more heterogenous, which appears to reflect different stages in the life cycle of Cx36' and 'All of these proteins showed considerable colocalization with Cx36 in AII amacrine cell dendrites'. It would be nice to see quantification data to support these claims.

Thank you for this suggestion. We have added a colocalization analysis to figure 3 (C & D). We quantified the colocalization for the endocytosis proteins Eps15l1 and Hip1r. This quantification included a flipped control to rule out random overlap. For both proteins we confirmed true colocalization (Figure 3D).

(6) In Figure 5B, it would be helpful if there were arrows or some kind in western blottings to indicate which bands are supposed to be the targeted proteins.

We have added arrows in IP samples to indicate bands representing the corresponding protein.

(7) In the sentence including 'for the PBM of Cx36, as it is the case for ZO-1', what is PBM?

The PBM means PDZ binding motif. We have added an explanation for this abbreviation in line 244.

(8) Please add a description of the Cx35b promoter construct in the Method section.

The Cx35b Promoter is a 6.5kb fragment. We will make the clone available via Addgene to ensure that all details of the clone can be accessed via snapgene or alternative software.

eLife Assessment

This valuable study explores changes in remote memory impairment in an amyloid pathology mouse model, demonstrating that progressive deficits coincide with inhibitory interneuron alterations. While the findings shed light on circuit remodeling in this model, the mechanistic links between heightened inhibition and memory loss are currently incomplete. Additional data and deeper analysis may be needed to fully substantiate the authors' interpretations.

Reviewer #1 (Public review):

This study presents evidence that remote memory in the APP/PS1 mouse model of Alzheimer's disease (AD) is associated with PV interneuron hyperexcitability and increased inhibition of cortical engram cells. Its strength lies in the fact that it explores a neglected aspect of memory research - remote memory impairments related to AD (for which the primary research focus is usually on recent memory impairments) -which has received minimal attention to date. While the findings are intriguing, the weakness of the paper hovers around purely correlational types of evidence and superficial data analyses, which require substantial revisions as outlined below.

Major concerns:

(1) In light of previous work, including that by the authors themselves, the data in Figure 1 should be complemented by measurements of recent memory recall in order to assess whether remote memories are exclusively impaired or whether remote memory recall merely represents a continuation of recent memory impairments.

(2) Figure 2 shows electrophysiological properties of PV cells in the mPFC that correlate with the behavior shown in Figure 1. However, the mice used in Figure 2 are different than the mice used in Figure 1. Thus, the data are correlative at best, and the authors need to confirm that behavioral impairments in the APP/PS1 mice crossed to PV-Cre (and SST-Cre mice) used in Figure 2 are similar to those of the APP/PS1 mice used in Figure 1. Without that, no conclusions between behavioral impairments and electrophysiological as well as engram reactivation properties can be made, and the central claims of the paper cannot be upheld.

(3) The reactivation data starting in Figure 3 should be analysed in much more depth: a) The authors restrict their analysis to intra-animal comparisons, but additional ones should be performed, such as inter-animal (WT vs APP/PS1) as well as inter-age (12-16w vs 16-20w). In doing so, reactivation data should be normalized to chance levels per animal, to account for differences in labelling efficiency - this is standard in the field (see original Tonegawa papers and for a reference). This could highlight differences in total reactivation that are already apparent, such as for instance in WT vs APP/PS1 at 20w (Figure 3o), and highlight a decrease in reactivation in AD mice at this age, contrary to what is stated in lines 213-214. b) Comparing the proportion of mcherry+ cells in PV- and PV+ is problematic, considering that the PV- population is not "pure" like the PV+, but rather likely to represent a mix of different pyramidal neurons (probably from several layers), other inhibitory neurons like SST and maybe even glial cells. Considering this, the statement on line 218 is misleading in saying that PVs are overrepresented. If anything, the same populations should be compared across ages or groups. c) A similar concern applies to the mcherry- population in Figure 4, which could represent different types of neurons that were never active, compared to the relatively homogeneous engram mcherry+ population. This could be elegantly fixed by restricting the comparison to mCherry+Fos+ vs mCherry+Fos- ensembles, and could indicate engram reactivation-specific differences in perisomatic inhibition by PV cells.

(4) At several instances, there are some doubts about the statistical measures having been employed: a) In Figure 4f, it is unclear why a repeated measurement ANOVA was used as opposed to a regular ANOVA. b) In Supplementary Figure 2b, a Mann-Whitney test was used, supposedly because the data were not normally distributed. However, when looking at the individual data points, the data does seem to be normally distributed. Thus, the authors need to provide the test details as to how they measured the normalcy of distribution.

Minor concerns:

(1) Line 117: The authors cite a recent memory impairment here, as shown by another paper. However, given the notorious difficulty in replicating behavioral findings, in particular in APP/PS1 mice (number of backcrossings, housing conditions, etc., might differ between laboratories), such a statement cannot be made. The authors should either show in their own hands that recent memory is indeed affected at 12 weeks of age, or they should omit this statement.

(2) Pertaining to Figure 3, low-resolution images of the mPFC should be provided to assess the spread of injection and the overall degree of double-positive cells.

Reviewer #2 (Public review):

This study presents a comprehensive investigation of remote memory deficits in the APP/PS1 mouse model of Alzheimer's disease. The authors convincingly show that these deficits emerge progressively and are paralleled by selective hyperexcitability of PV interneurons in the mPFC. Using viral-TRAP labeling and patch-clamp electrophysiology, they demonstrate that inhibitory input onto labeled engram cells is selectively increased in APP/PS1 mice, despite unaltered engram size or reactivation. These findings support the idea that alterations in inhibitory microcircuits may contribute to cognitive decline in AD.

However, several aspects of the study merit further clarification. Most critically, the central paradox, i.e., increased inhibitory input without an apparent change in engram reactivation, remains unresolved. The authors propose possible mechanisms involving altered synchrony or impaired output of engram cells, but these hypotheses require further empirical support. Additionally, the study employs multiple crossed transgenic lines without reporting the progression of amyloid pathology in the mPFC, which is important for interpreting the relationship between circuit dysfunction and disease stage. Finally, the potential contribution of broader network dysfunction, such as spontaneous epileptiform activity reported in APP/PS1 mice, is also not addressed.

eLife Assessment

This valuable study presents a novel approach to enhance the therapeutic potential of mesenchymal stromal cells (MSCs) by genetically modifying their glycogen synthesis pathway, resulting in increased glycogen accumulation and improved cell survival under starvation conditions, particularly in the context of experimental pulmonary fibrosis. The methods and findings are generally solid and could be strengthened by investigating the kinetics of persistence, the immunomodulatory effects, and the underlying improved mechanism of action of MSCs in this pulmonary fibrosis model. If confirmed, this approach could suggest potential methods to improve the therapeutic functionality of MSCs in cell therapy strategies.

Reviewer #1 (Public review):

Summary:

This study provides the first evidence that glucose availability, previously shown to support cell survival in other models, is also a key determinant for post-implantation MSC survival in the specific context of pulmonary fibrosis. To address glucose depletion in this context, the authors propose an original, elegant, and rational strategy: enhancing intracellular glycogen stores to provide transplanted MSCs with an internal energy reserve. This approach aims to prolong their viability and therapeutic functionality after implantation.

Strengths:

The efficacy of this metabolic engineering strategy is robustly demonstrated both in vitro and in an orthotopic mouse model of pulmonary fibrosis.

Comments and questions for clarification:

(1) Glycogen biosynthesis typically involves several enzymes. In this context, could the authors comment on the effect of overexpressing a single enzyme - especially a mutant version - on the structure or quality of the glycogen synthesized?

(2) Regarding the in vitro starvation experiments (Figure 2C), what oxygen conditions (pO₂) were used? Are these conditions physiologically relevant and representative of the in vivo lung microenvironment?

(3) In the in vitro model, how many hours does it take for the intracellular glycogen reserve to be completely depleted under starvation conditions?

(4) For the in vivo model, is there a quantitative analysis of the survival kinetics of the transplanted cells over time for each group? This would help to better assess the role and duration of glycogen stores as an energy buffer after implantation.

(5) Finally, the study was performed in male mice only. Could sex differences exist in the efficacy or metabolism of the engineered MSCs? It would be helpful to discuss whether the approach could be expected to be similarly effective in female subjects.

(6) The number of mice for each group and time point should be specified.

Reviewer #2 (Public review):

Summary:

In this article, the authors investigate enhancing the therapeutic and regenerative properties of mesenchymal stem cells (MSCs) through genetic modification, specifically by overexpressing genes involved in the glycogen synthesis pathway. By creating a non-phosphorylatable mutant form of glycogen synthase (GYSmut), the authors successfully increased glycogen accumulation in MSCs, leading to significantly improved cell survival under starvation conditions. The study highlights the potential of glycogen engineering to improve MSC function, especially in inflammatory or energy-deficient environments. However, critical gaps in the study's design, including the lack of validation of key findings, limited differentiation assessments, and missing data on MSC-GYSmut resistance to reactive oxygen species (ROS), necessitate further exploration.

Strengths:

(1) Novel Approach: The study introduces an innovative method of enhancing MSC function by manipulating glycogen metabolism.

(2) Increased Glycogen Storage: The genetic modification of GYS1, resulting in GYSmut, significantly increased glycogen accumulation, leading to improved MSC survival under starvation, which has strong implications for enhancing MSC therapeutic properties in energy-deficient environments.

(3) Potential Therapeutic Impact: The findings suggest significant therapeutic potential for MSCs in conditions that require improved survival, persistence, and immunomodulation, especially in inflammatory or energy-limited settings.

(4) In Vivo Validation: The in vivo murine model of pulmonary fibrosis demonstrated the improved survival and persistence of MSC-GYSmut, supporting the translational potential of the approach.

Weaknesses:

(1) Lack of Differentiation Assessments: The study did not evaluate key MSC differentiation pathways, including chondrogenic and osteogenic differentiation. The absence of analysis of classical MSC surface markers and multipotency limits the understanding of the full potential of MSC-GYSmut.

(2) Missing Validation of RNA Sequencing Data: Although RNA sequencing data revealed promising transcriptomic changes in chondrogenesis and metabolic pathways, these findings were not experimentally validated, limiting confidence.

(3) Lack of ROS Resistance Analysis: Resistance to reactive oxygen species (ROS), an important feature for MSCs under regenerative conditions, was not assessed, leaving out a critical aspect of MSC function.

(4) Inconsistencies in In Vivo Data: There is a discrepancy between the number of animals shown in the figures and the graph (three individuals vs. five animals), as well as missing details on how luciferase signal intensity was quantified, requiring further clarification.

(5) Limited Exploration of Immunosuppressive Properties: The study did not address the immunosuppressive functions of MSC-GYSmut, which are critical for MSC-based therapies in clinical settings.

Conclusion:

The study presents an exciting new direction for enhancing MSC function through glycogen metabolism engineering. While the results show promise, key experiments and validations are missing, and several areas, such as differentiation capacity, ROS resistance, and immunosuppressive properties, require further investigation. Addressing these gaps would solidify the conclusions and strengthen the potential clinical applications of MSC-GYSmut in regenerative medicine.

eLife Assessment

Valencia et al. combine elegant in vitro biochemical experiments with functional assays in cardiomyocytes to determine which properties of the FHOD3 formin are essential for sarcomere assembly. Using separation-of-function mutants, they show that FHOD3's elongation activity, rather than its nucleation, capping, or bundling activities, is key to its sarcomeric function. This is an important finding and the data presented in the manuscript are convincing; however, the presence of FHOD3 at filament barbed ends in the TIRF elongation assays should probably be verified directly in a future study.

Reviewer #1 (Public review):

Summary:

Formins are complex proteins with multiple effects on actin filament assembly, including nucleation, capping with processive elongation, and bundling. Determining which of these activities are important for a given biological process and normal cellular function is a major challenge.

Here, the authors study the formin FHOD3L, which is essential for normal sarcomere assembly in muscle cells. They identify point mutants of FHOD3L in which formin nucleation and elongation/bundling activities are functionally separated. Expression of these mutants in neonatal rat ventricular myocytes shows that the control of actin filament elongation by formin is the major activity required for normal assembly of functional sarcomeres.

Strengths:

The strength of this work is to combine sensitive biochemical assays with excellent work in neonatal rat ventricular myocytes. This combination of approaches is highly effective for analyzing the function of proteins with multiple activities in vitro. The authors have pushed the experiments and data analysis as far as possible with the technologies available to them.

Weaknesses:

FHOD3L is not the easiest formin to study because of its relatively weak nucleation activity and the short duration of capping events. This difficulty imposes rigorous biochemical analysis and careful interpretation of the data. As the authors acknowledge, it will be important in future to perform complementary multi-color TIRF experiments to confirm that the brief accelerations in the elongation of actin filaments are indeed due to FHOD3 binding.

Reviewer #3 (Public review):

Valencia et al. aim to elucidate the biochemical and cellular mechanisms through which the human formin FHOD3 drives sarcomere assembly in cardiomyocytes. To do so, they combined rigorous in vitro biochemical assays with comprehensive in vivo characterizations, evaluating two wild type FHOD3 isoforms and two function-separating mutants. Surprisingly, they found that both wild type FHOD3 isoforms can nucleate new actin filaments, as well as elongate existing actin filaments in conjunction with profilin following barbed-end capping. This is in addition to FHOD3's proposed role as an actin bundler. Next, the authors focused on the longer isoform FHOD3L due to its essential role in sarcomere assembly in cardiomyocytes. They asked whether FHOD3L promote sarcomere assembly through its activity in actin nucleation or rather elongation. To do so, the authors designed two function-separating mutants: the K1193L mutation in the FH2 domain, known for its importance in actin nucleation, and the glycine-serine linker substitution in the FH1 domain ("GS-FH1",) known for its requirement in actin elongation. They demonstrated that while K1193L maintains its elongation activity and greatly diminishes nucleation and bundling, in GS-FH1 keeps its nucleation activity while lose its capacity to drive elongation. Armed with these tools, the authors attempted to rescue FHOD3L siRNA-treated neonatal rat ventricular myocytes (NRVM) with transgenes carrying wild type, K1193L, or GS-FH1 mutant forms of human FHOD3. In each condition, they evaluated the numbers and morphology of sarcomeres, as well as their ability to beat and generate cardiac rhythm. The authors found that while the wild type FHOD3L and the K1193L mutant can rescue sarcomere morphology and physiology, the GS-FH1 mutant fails to do so. Given that in GS-FH1 mainly elongation activity is compromised, the authors concluded that the elongation activity of FHOD3 is essential for its role in sarcomere assembly in cardiomyocytes, while its nucleator activity is dispensable. Overall, this important study provided a broadened view on the biochemical activities of FHOD3, and a pioneering view on a possible cellular mechanism of how FHOD3L drives sarcomere assembly. If further validated, this can lead to new mechanistic models of sarcomere assembly and potentially new therapeutic targets of cardiomyopathy.

The conclusions of this paper are mostly well supported by the comprehensive biochemical analyses performed by the authors. In my original assessment, I raised the point that the extreme low level of GS-FH1 signal in transfected cells in Figure 6A may reflect a failure of actin-binding by this construct in vivo, rather than its inability of driving elongation. The authors have thoroughly addressed this concern by: 1) providing new images of the GS-FH1 rescue condition with HA-FHOD3L signal intensities matching that of the K1193L rescue condition, and 2) quantitatively demonstrating that the expression levels in the GS-FH1 rescue condition are comparable with that of wild type FHOD3L rescue condition. This is nicely complemented by the new phalloidin staining of the GS-FH1 rescue condition, which showcased additional details of actin puncta reminiscent of that present in muscle stress fibers or premyofibrils. Overall, I am now convinced that the GS-FH1 cannot rescue sarcomere formation even when expressed at comparable levels. Given that GS-FH1 demonstrates actin elongation defects in vitro, it is reasonable to conclude that the actin elongation function of FHOD3L is essential for sarcomere formation in vivo.

Author response:

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

Public Reviews:

Reviewer #1 (Public review):

Summary:

Formins are complex proteins with multiple effects on actin filament assembly, including nucleation, capping with processive elongation, and bundling. Determining which of these activities is important for a given biological process and normal cellular function is a major challenge.

Here, the authors study the formin FHOD3L, which is essential for normal sarcomere assembly in muscle cells. They identify point mutants of FHOD3L in which formin nucleation and elongation/bundling activities are functionally separated. Expression of these mutants in neonatal rat ventricular myocytes shows that the control of actin filament elongation by formin is the major activity required for the normal assembly of functional sarcomeres.

Strengths:

The strength of this work is to combine sensitive biochemical assays with excellent work in neonatal rat ventricular myocytes. This combination of approaches is highly effective for analyzing the function of proteins with multiple activities in vitro.

Weaknesses:

FHOD3L does not seem to be the easiest formin to study because of its relatively weak nucleation activity and the short duration of capping events. This difficulty imposes rigorous biochemical analysis and careful interpretation of the data, which should be improved in this work.

We thank the reviewer for their praise and appreciation of our work. Indeed, FHOD3L is a challenging formin to work with.

Important points are raised here and below regarding the brief elongation events we reported. As suggested, we performed more rigorous analysis of the data and present it in the revised manuscript. We now report that from 45 dim regions analyzed, in three independent experiments with wild type FHOD3L, we detected 40 bursts. (The remaining five could be formin falling off too quickly to detect or the dim spots could be regions of inhomogeneity in intensity, not due to formin.) For comparison to the presented data with FHOD3L-CT, we analyzed the filaments in TIRF assays with no formin present. As the reviewers point out, inhomogeneities in filament intensity are normal. Thus, we examined any dim spots for pauses and/or bursts. As is now reported in Figure 2G,H, the velocity of growth of these dim spots is indistinguishable from the velocity of the rest of the filament. We acknowledge that our numbers may not be perfectly accurate, due to the noise in our system, we believe that the difference of 3-4 fold increase versus no change in rate is substantial and convincing.

We also determined the number of dim spots per length of filament. We found a higher frequency when FHOD3L-CT or FHOD3S-CT was present vs no formin, as now shown in Figure 2 – supplements 1G and 2E.

We were asked about the pauses we observe before bursts of elongation and how we know they are functionally relevant. The short answer is that we do not know. We reported them because they were so common: Of the 40 bursts, pauses preceded the burst in 38 cases. We cannot rule out that this pause reflects an interaction with the surface but might expect the frequency to be lower if it were. We revise the text to make our conclusions about pauses more circumspect.

We are convinced that the brief dim events we observed in the presence of FHOD3L-CT, in fact, reflect formin-mediated elongation and worked hard to improve their presentation, in addition to the added analysis. We include new kymographs, including examples from FHOD3L, FHOD3S, K1193L, and actin alone. We hope that the reviewers are also convinced.

This does not preclude our interest in the microfluidics and two-color assays, which will be pursued in the future. We have reached out to a colleague who is set up to repeat these measurements with microfluidics-assisted TIRF. The noise should be greatly reduced and the system is also optimal for directly visualizing labeled FHOD3, as suggested. We expect these experimental approaches will provide additional insights.

Reviewer #2 (Public review):

This article elucidates the biochemical and cellular mechanisms by which the FHOD-family of formins, particularly FHOD3, contributes to sarcomere formation and contractility in cardiomyocytes. Formins are mainly known to nucleate and elongate actin filaments, with certain family members also exhibiting capping, severing, and bundling activities. Although FHOD3 has been well-established as essential for sarcomere assembly in cardiomyocytes, its precise biochemical functions and contributions to actin dynamics remain poorly understood.

In this study, the authors combine in vitro biochemical assays with cellular experiments to dissect FHOD3's roles in actin assembly and sarcomere formation. They demonstrate that FHOD3 nucleates actin filaments and acts as a transient elongator, pausing elongation after an initial burst of filament growth. Using separation-of-function mutants, they show thatFHOD3's elongation activity - rather than its nucleation, capping, or bundling capabilities - is key for its sarcomeric function.

The experiments have been conducted rigorously and well-analyzed, and the paper is clearly written. The data presented support the authors' conclusions. I appreciate the detailed description and rationale behind the FHOD3 constructs used in this study.

We are happy to hear others find paper to be clearly written and well described.

However, I was somewhat surprised and a bit disappointed that while the authors conducted single-color TIRF experiments to observe the effects of FHOD3 on single filaments, they did not use fluorescently labeled FHOD3 to directly visualize its behavior. Incorporating such experiments would significantly strengthen their conclusions regarding FHOD3's bursts of elongation interspersed with capping activity. While I understand this might require a few additional weeks of experiments, these data would add considerable value by directly testing the proposed mechanism.

We appreciate the suggestion and hope to incorporate a two-color approach soon. As noted, FHOD3L is not always easy to work with and we do not have a functional labeled copy of the protein at this time.

There is a typo in the word "required" in line number 30. The authors also use fit data to extract parameters in several panels (e.g., Figures 2b, 2d, 3a, and 3b). While these fit functions may be intuitive to actin experts, explicitly describing the fit functions in the figure legends or methods would greatly benefit the broader readership.

Thank you for these comments. We updated the indicated figures and described the analysis in greater detail.

Reviewer #3 (Public review):

Valencia et al. aim to elucidate the biochemical and cellular mechanisms through which the human formin FHOD3 drives sarcomere assembly in cardiomyocytes. To do so, they combined rigorous in vitro biochemical assays with comprehensive in vivo characterizations, evaluating two wild-type FHOD3 isoforms and two function-separating mutants. Surprisingly, they found that both wild-type FHOD3 isoforms can nucleate new actin filaments, as well as elongate existing actin filaments in conjunction with profilin following barbed-end capping. This is in addition to FHOD3's proposed role as an actin bundler. Next, the authors asked whether FHOD3L promotes sarcomere assembly in cardiomyocytes through its activity in actin nucleation or rather elongation. With two function-separating mutants, the authors evaluated the numbers and morphology of sarcomeres, as well as their ability to beat and generate cardiac rhythm. The authors found that while the wild-type FHOD3L and the K1193L mutant can rescue sarcomere morphology and physiology, the GS-FH1 mutant fails to do so. Given that in GS-FH1 mainly elongation activity is compromised, the authors concluded that the elongation activity of FHOD3 is essential for its role in sarcomere assembly in cardiomyocytes, while its nucleator activity is dispensable. Overall, this important study provided a broadened view on the biochemical activities of FHOD3, and a pioneering view on a possible cellular mechanism of how FHOD3L drives sarcomere assembly. If further validated, this can lead to new mechanistic models of sarcomere assembly and potentially new therapeutic targets of cardiomyopathy.

The conclusions of this paper are mostly well supported by the comprehensive biochemical analyses performed by the authors. However, the sarcomere assembly defect phenotype in the GS-FH1 rescue condition requires further investigation, as the extremely low level of GS-FH1 signal in transfected cells in Figure 6A may reflect a failure of actin-binding by this construct in vivo, rather than its inability to drive elongation. Though the authors do show in Figure 6 that GS-FH1 can bind to normal-looking sarcomeres when they are present, this may be due to a lack of siRNA activity in these cells, such that endogenous FHOD3L is still present. In this possible scenario, GS-FH1 may dimerize with endogenous FHOD3L. The authors should demonstrate that GS-FH1 alone can indeed interact with existing actin filaments in vivo. While this has been clearly demonstrated in vitro, given the more complex biochemical environment in vivo where additional unknown binding partners may present, cautions should be made when extrapolating findings from the former to the latter.

The reviewer is concerned about the low protein levels in the GS-FH1 rescue experiments as reflected in the HA fluorescence intensity distributions shown in Fig. 5 Supplement 2A. While the scenario proposed could explain our observations with the GSFH1 rescues it is quite complex. Nor does the scenario preclude the conclusion that the FH1 domain is critical. We agree that the observed sarcomeres are likely to be residual in cells with incomplete RNAi. We now include the image of a cell that is still full of sarcomeres and note that the GH-FH1 is expressed at a relatively high level and striated throughout the cell. We interpret this as evidence that GS-FH1 is stable when suitable binding sites are available. We cannot exclude that there is more GS-FH1 because there was more endogenous FHOD3L with which to heterodimerize. If the GS-FH1 heterodimer were simply poisoning the wild type protein, we do not expect that it would be bound correctly to sarcomeres. If, instead, heterodimers have some activity, it seems far from sufficient to rescue sarcomere formation, suggesting that two functional FH1 domains are critical.

Furthermore, we do not see evidence of correlation between protein levels and rescue at the level present in these cells (addressed below). Unfortunately, the proposed IP to test whether FHOD3L binds actin in vivo would only potentially report on filament side binding (both direct and indirect). It would not address whether the GS-FH1 mutant functions as a nucleator, elongator, bundler and/or capping protein in vivo.

The critical question that we can address is whether the phenotype is due to low protein levels, assuming the protein present is functional, or due to loss of elongation activity by FHOD3L. To address this question, we returned to our data.

First, we plotted the distributions of the intensities of the cells we analyzed further, in addition to the automated readout of all of the cells in the dish (Fig. 4 supplement 1). These cells were selected randomly and, as should be the case, the distributions of their intensities agree well with the original distributions for the three different rescue constructs: FHOD3L, K1193L, and GS-FH1 (Fig. 6 supplement 1). We then asked whether there was any correlation in HA intensities with the sarcomere metrics. As seen in our pilot data, no correlation is evident in any of the three cases across the range of intensities we collected (400 – 2700 a.u.) (old Fig. 6 supplement C,D,E). We now replace the data from pilot experiments with analysis of HA intensities and sarcomere metrics from the data sets included in the paper (new Fig 6. Supplement 1). Again, little to no correlation was observed (the single highest r-squared value is 0.2 and the remaining eight values are less than or equal to 0.08).

To more specifically address the question of whether low HA fluorescence intensity is likely to reflect sufficient protein levels to build sarcomeres we re-examined two data sets from the FHOD3L WT rescue data. We found that, by chance, the first replicate of data from the wild type rescue has a comparable intensity distribution to that of the GSFH1 rescues (580 +/- 261 / cell vs. 548 +/- 105 / cell). In addition, we collected all of the data from cells with intensity levels <720, designed to mimic the distribution of the GS-FH1 cells (Fig. 6 supplement 3). We then compared the sarcomere metrics (sarcomere number, sarcomere length, sarcomere width) between the full data set and the two low intensity subsets:

• Sarcomere number is the only non-normal metric. We therefore used the Mann Whitney U test, which shows no difference between all 3 WT distributions.

• We compared Z-line lengths by one-way ANOVA and Tukey's post hoc tests, again finding no significant difference for all distributions.

• Sarcomere length shows a weakly significant difference (p=0.038) between the whole WT data set and bio rep 1, but no difference between the whole WT data set and the HA<720 group.

Thus, cells expressing wild type FHOD3L at levels comparable to levels detected in GS-FH1 mutant rescues, are fully rescued. Based on these findings we conclude that the expression levels in the GS-FH1 are high enough to rescue the FHOD3 knock down, supporting our conclusion that the defect is due to loss of elongation activity. We have added this analysis and discussion to the revised manuscript.

Recommendations for the authors:

Reviewing Editor Comments:

You will see that the 3 reviewers are very positive about your work and appreciate the elegant combination of biochemical assays and functional tests in cardiomyocytes. We've had a long discussion with them and we all agree that two experiments deserve further effort to make the conclusions of your paper more convincing.

Thank you.

The first experiment is the TIRF elongation assay, where the two biochemist Reviewers remain doubtful that these short events are really due to the presence of a formin at the end of the filament. One of them suggests that two-color imaging with a labeled formin should clearly prove this point.

We agree that the elongation assays can be improved. Given the similarity of processivity of Fhod3L, Fhod3S and Drosophila FhodA (measured by a distinct method), we are inclined to believe them. However, the reviewer raises an excellent point about the accuracy of the measurements given the resolution (and noise) of the data. We are interested in the two-color imaging assay but do not believe it will necessarily simplify the analysis. We suspect that Fhod spends more time at/near the barbed end than is apparent based on elongation rates. The fact that we see repeated events on individual filaments at such low concentrations of FHOD3L (0.1 nM) supports this idea. Otherwise, the likelihood of FHOD3L finding barbed ends so often is really quite low.

We will return to these experiments, using alternate methods, curious to see what else we learn. In the meantime, we conducted more thorough analysis, including controls, and improved visualization of example traces. Data for elongation analysis and kymographs were acquired with Jfilament. We stretched the x-axis (time) in kymographs for FHOD3L-CT (Fig. 2F), FHOD3S-CT (Fig. 2, supplement 2C), FHOD3L-CT K1193L (Fig. 3, supplement 1A), and actin alone (Fig 2G), and highlighted regions of analysis. The slopes for these regions, separated based on intensity, were fit to the data in KaleidaGraph. The fits are offset from the data such that they do not obscure the filaments and corresponding rates are given. The fact that we never see fast dim regions when FHOD3 is not present, as shown in Fig. 2H and that the frequency of dim events is markedly increased (Fig. 2-supplements 1G and 2E) give us confidence that the events are real. We acknowledge in the text that the precise values of the short events may be inaccurate due to the resolution of our experiments. We hope the reviewers are convinced by the improved analysis.

The second experiment is the sarcomere assembly defect phenotype in the GS-FH1 rescue condition. This requires further investigation, as the extremely low level of GS-FH1 signal in transfected cells in Figure 6A may reflect a failure of actin-binding/nucleation in vivo, rather than its inability to elongate F-actin. Although you show that GS-FH1 can bind to sarcomeres when they are present, this may be due to a lack of siRNA activity in these cells, such that endogenous FHOD3L is still present. In this possible scenario, GS-FH1 could dimerize with endogenous FHOD3L.

We agree that the sarcomeres we see are likely to be residual and could reflect some remaining endogenous FHOD3. The reviewers are concerned about the low protein levels in the GSFH1 rescues. First, we do not agree that the levels are “extremely” low. Through careful analysis, we established that 3xHA-FHOD3L intensities between 300 and 3000 a.u./um² were sufficient for full rescue. The mean for the GSFH1 experiments is 533 +/- 93, which is well within this range. Furthermore, we did not observe correlation between sarcomere number, length, or width and HA intensity over the full range collected for wild type FHOD3L or within the GS-FH1 data. We previously showed pilot data but now show correlation analysis for every analyzed cell (Fig. 4 – figure supplement 1 D-F). We conducted this analysis on all of the mutant rescue experiments (Fig. 6-supplement 1). Finally, we identified two subpopulations of the wildtype rescue data. One is all of the cells with HA intensity < 720, which gives a distribution of mean 545 +/- 85. The second set is the first biological replicate of wild type rescue, which has a distribution of mean 560 +/- 160. Again correlation shows little relationship between HA levels and sarcomere metrics. Nevertheless, we show intensity level matched images in Fig 6, as opposed to images reflecting average intensities.

The critical question remains whether the phenotype is due to low protein levels or due to loss of elongation by FHOD3L. Notably, we now show a cell that is full of sarcomeres and has relatively high FHOD3L levels as well, consistent with available binding sites stabilizing mutant protein but not ruling out heterodimerization (Fig. 6 – figure supplement 2C). Others have expressed mutant FHOD3L in a wild type background in mice. They observed poisoning, consistent with heterodimerization. Thus, it is possible that, as suggested, the FHOD3L-GSFH1 detected in sarcomeres is in fact heterodimerized with residual endogenous FHOD3L. In this case, we would still conclude that the protein is not functional enough to rescue, supporting a role for the FH1 domain.

In the future, we plan to perform experiments with compromised, but not inactive, FH1 domains, as we discuss in the paper.

We hope that you will find these comments useful.

Yes, the comments were thoughtful and helped us write a better paper. Thank you.

Reviewer #1 (Recommendations for the authors):

Some experiments should be described and analyzed more carefully. This lack of clarity calls into question the interpretation of some experiments. Overall, this study is not yet as convincing as it should be.

Main recommendations:

(1) Formin elongation phases in the TIRF experiment are not convincing. They are rare and it is difficult to see any significant difference between the control movie without FHOD3L-CT and the movie with FHOD3L-CT. Filaments assembled in the absence of FHOD3L-CT also show some fluorescence inhomogeneity (which is normal), and measurements of formin elongation rates and capping times are not convincing (for example, the kymograph of the control profilin-actin situation in Figure 2F also shows a fast elongation phase on the right).

Please see response above. We conducted more thorough analysis and created improved visualizations. We hope the data are more convincing now.

It is also difficult to understand how an accurate measurement can be made from these noisy kymographs, and the method section should explain that precisely.

This is a valid point. We added details of analysis to the methods section and we discuss the fact that the measurements are at the limit of our resolution in the paper. We rely on the large (~3-fold) difference in elongation, more than specific elongation rates for our interpretation.

One of the problems is that these events are too transient to quantify well with noisy data. I noticed that the formin concentration used in these movies is quite low (0.1 nM FHOD3L-CT). Is there a reason for this? Is it possible to increase the formin concentration to increase the number of formin capping/elongation events and provide more convincing movies?

We acknowledge that the data are noisy. We felt that it was necessary to perform experiments with filaments only tethered at one end, leaving the growing end free. We did so, in part, because when we did experiments with biotinylated actin to anchor the filaments down, we observed pauses in the absence of formin. Ultimately, we compromised, using anchored seeds and a relatively low concentration of NEM-myosin to decrease motion of the actin filaments.

The experiments were performed with such low FHOD3L-CT because it was a potent nucleator in TIRF assays, making data analysis nearly impossible with more formin present. FHOD3S-CT and FHOD3L-CT K1193L behaved somewhat differently between these experiments and we were able to perform them with 1 nM formin.

Not seeing formin at the tip of the filaments is an additional difficulty because we do not know if these pauses occur because formin is stuck to the coverslips (which could very well happen with these sticky proteins) or freely bound at the end of a filament as the text suggests. Is there any argument in favor of one scenario over the other?

This will be an important experiment. As described above, we suspect that Fhod spends more time at/near the barbed end than is apparent based on elongation data. The fact that we see repeated events on individual filaments at such low concentrations of FHOD3L (0.1 nM) supports this idea. Otherwise, the likelihood of FHOD3L finding barbed ends so often is really quite low. In order to address the question about the cause of pauses, we reviewed our data, finding that 38 of 40 bursts were preceded by pauses. We do, however, discuss that we cannot rule out non-specific interactions with the surface.

(2) Pyrene elongation assays in the presence of profilin are actually more convincing to test the elongation ability of formins. However, such an assay is not presented for all mutants. It should be.

While we agree to some extent with this comment, we did not include the pyrene data for all of the mutants because the shapes of the curves were even more complicated than those seen with wild type FHOD3L-CT rendering them uninterpretable.

(3) Some experiments (e.g. in Figure 2E) are performed with yeast profilin, while others (e.g. in Figure 2F) are performed with human profilin. Obviously, both profilins could modulate formin activity differently and the side-by-side interpretation of both experiments is difficult. Could the authors stick to human profilin for all experiments?

We used to always perform pyrene assays with yeast profilin because it was known to be insensitive to pyrene. These data were collected before we realized that the affinity of human profilin for actin is so high that we could probably do everything with this profilin. We have compared the two profilins for other formins, e.g. Delphilin, Capu, and did not observe detectable differences.

Minor recommendations:

(1) The pyrene assays with the light blue colored curve choice are not ideal. I have difficulties seeing some of the curves.

Thank you. We added symbols to a subset of the traces to make them more visible.

(2) In the same curves, I can't understand what the +3.75 and 0.078 numbers mean. Could these results be plotted in a clearer way?

These values are the lowest concentrations in the range tests. They were matching light blue with black outline for visibility. We added symbols and changed the color of the numbering for improved visibility/understanding.

(3) In Figure 2D, is the Kd of I1163A really determined only from 2 experimental data points?

Of course not. We now show the figure with extended axes in Fig. 2 - figure supplement 1C.

(4) In Figure 2C, the shape of the curves suggests that this is not a pure capping assay, but a mix of capping and nucleation. It's not dramatic but could lead to an under-estimation of the capping efficiency.

We agree with the reviewer that the complicated shapes confound interpretation. Our analysis is based on the earliest slopes, in part, for this reason. We added discussion of this complication to the text.

Reviewer #3 (Recommendations for the authors):

Suggestions for additional experiments:

(1) To evaluate whether GS-FH1 alone can indeed interact with existing actin filaments in vivo, the authors may consider performing immunoprecipitation assays with GS-FH1 extracted from rescued NRVMs.

An IP of GS-FH1 from cells could show actin filament side binding but, unfortunately, will not provide any information about filament end binding, which is of much greater interest.

It will be helpful to show phalloidin staining in GS-FH1 rescues in a similar manner as in Figure 6-supplement 1, panel B, and compare that with mock rescue in Figure 4 panel D. It will be essential to prove this prior to concluding that actin elongation activity is essential for sarcomere assembly.

This is an excellent suggestion. We now include images of phalloidin stained cells from both K1193L and GS-FH1 rescues (Fig. 6A’ – supplement 2A,B). We were intrigued to see small actin punctae that were sometimes aligned. We speculate that these could be pre-premyofibrils and suggest that this is further evidence that the GS-FH1 protein is not completely unstable.

(2) Prior to sarcomere assembly, a-actinin is known to form short bundles with actin filaments (I-Z-I complex) without clearly defined periodicity. This semi-ordered state then transforms into the more ordered sarcomeres with periodic spacing. It will be valuable to show the phalloidin staining in addition to the a-actinin IF consistently across all conditions. This may lead to further insights into the defects of sarcomere assembly. Along the same vein, higher magnification images showcasing several sarcomeres will help the readers evaluate these defects.

We agree that there are additional valuable measurements to be made. In order to favor synchronized contraction, we plated the cells at too high a density to reliably identify IZI complexes. We have included some zoomed in images of the phalloidin staining.

Recommendations for improving the writing:

The authors mentioned the interaction between cardiac MyBP-C and FHOD3L as essential for the localization of FHOD3L to the C-line of the sarcomere. Can they discuss whether this interaction is important for the role of FHOD3L in sarcomere assembly? If so, how?

This is a very interesting question that we cannot answer at this time.

Minor corrections to the text and figures:

In the legend of Figure 2-Figure Supplement 1, the labels of (F) and (E) are swapped.

Thank you for catching this.

Author response:

eLife Assessment

This useful study presents Altair-LSFM, a solid and well-documented implementation of a light-sheet fluorescence microscope (LSFM) designed for accessibility and cost reduction. While the approach offers strengths such as the use of custom-machined baseplates and detailed assembly instructions, its overall impact is limited by the lack of live-cell imaging capabilities and the absence of a clear, quantitative comparison to existing LSFM platforms. As such, although technically competent, the broader utility and uptake of this system by the community may be limited.

We thank the reviewers and editors for their thoughtful evaluation of our work and for recognizing the technical strengths of the Altair-LSFM platform, including the custom-machined baseplates and detailed documentation provided to support accessibility and reproducibility. We respectfully disagree, however, with the assessment that the system lacks live-cell imaging capabilities. We are fully confident in the system’s suitability for live-cell applications and will demonstrate this by including representative live-cell imaging data in the revised manuscript, along with detailed instructions for implementing environment control. Moreover, we will expand our discussion to include a broader, more quantitative comparison to existing LSFM platforms—highlighting trade-offs in cost, performance, and accessibility—to better contextualize Altair’s utility and adaptability across diverse research settings.

Public Reviews:

Reviewer #1 (Public review):

Summary:

The article presents the details of the high-resolution light-sheet microscopy system developed by the group. In addition to presenting the technical details of the system, its resolution has been characterized and its functionality demonstrated by visualizing subcellular structures in a biological sample.

Strengths:

(1) The article includes extensive supplementary material that complements the information in the main article.

(2) However, in some sections, the information provided is somewhat superficial.

Our goal was to make the supplemental content as comprehensive and useful as possible. In addition to the materials provided with the manuscript, our intention is for the online documentation (available at thedeanlab.github.io/altair) to serve as a living resource that evolves in response to user feedback. For this reason, we are especially interested in identifying and expanding any sections that are perceived as superficial, and we would greatly appreciate the reviewer’s guidance on which areas would benefit from further elaboration.

Weaknesses:

(1) Although a comparison is made with other light-sheet microscopy systems, the presented system does not represent a significant advance over existing systems. It uses high numerical aperture objectives and Gaussian beams, achieving resolution close to theoretical after deconvolution. The main advantage of the presented system is its ease of construction, thanks to the design of a perforated base plate.

We appreciate the reviewer’s assessment and the opportunity to clarify our intent. Our primary goal was not to introduce new optical functionality beyond that of existing high-performance light-sheet systems, but rather to reduce the barrier to entry for non-specialist labs.

(2) Using similar objectives (Nikon 25x and Thorlabs 20x), the results obtained are similar to those of the LLSM system (using a Gaussian beam without laser modulation). However, the article does not mention the difficulties of mounting the sample in the implemented configuration.

We agree that there are practical challenges associated with handling 5 mm diameter coverslips. However, the Nikon 25x can readily be replaced by a Zeiss W Plan-Apochromat 20x/1.0 objective, which eliminates the need for the 5 mm coverslip[1]. In the revised manuscript, we will more explicitly detail the practical challenges in handling a 5 mm coverslip and mention the alternative detection objective.

(3) The authors present a low-cost, open-source system. Although they provide open source code for the software (navigate), the use of proprietary electronics (ASI, NI, etc.) makes the system relatively expensive. Its low cost is not justified.

We understand the reviewer’s concern regarding the use of proprietary control hardware such as the ASI Tiger Controller and NI data acquisition cards. While lower-cost alternatives for analog and digital control (e.g., microcontroller-based systems) do exist, our choice was intentional. By relying on a unified and professionally supported platform, we minimize the complexity of sourcing, configuring, and integrating components from disparate vendors—each of which would otherwise demand specialized technical expertise. Moreover, in future releases, we aim to further streamline the system by eliminating the need for the NI card, consolidating all optoelectronic control through the ASI Tiger Controller. This approach allows users to purchase a fully assembled and pre-configured system that can be operational with minimal effort.

It is worth noting that the ASI components are not the primary cost driver. The full set—including XYZ and focusing stages, a filter wheel, a tube lens, the Tiger Controller, and basic optomechanical adapters—costs approximately $27,000, or ~18% of the total system cost. Additional cost reductions are possible. For example, replacing the motorized sample positioning and focusing stages with manual alternatives could reduce the cost by ~$12,000. However, this would eliminate key functionality such as autofocusing, 3D tiling, and multi-position acquisition. Open-source mechanical platforms such as OpenFlexure could in principle be adapted, but they would require custom assembly and would need to be integrated into our control software. Similarly, the filter wheel could be omitted in favor of a multi-band emission filter, reducing the cost by ~$5,000. However, this comes at the expense of increased spectral crosstalk, often necessitating spectral unmixing. An industrial CMOS camera—such as the Ximea MU196CR-ON, recently demonstrated in a Direct View Oblique Plane Microscopy configuration[2]—could substitute for the sCMOS cameras typically used in high-end imaging. However, these industrial sensors often exhibit higher noise floors and lower dynamic range, limiting sensitivity for low-signal imaging applications.

While a $150,000 system represents a significant investment, we consider it relatively cost-effective in the context of advanced light-sheet microscopy. For comparison, commercially available systems with similar optical performance—such as LLSM systems from 3i or Zeiss—are several-fold more expensive.

(4) The fibroblast images provided are of exceptional quality. However, these are fixed samples. The system lacks the necessary elements for monitoring cells in vivo, such as temperature or pH control.

We thank the reviewer for their positive comment regarding the quality of our fibroblast images. As noted, the current manuscript focuses on the optical design and performance characterization of the system, using fixed specimens to validate resolution and imaging stability. We acknowledge the importance of environmental control for live-cell imaging. Temperature regulation is routinely implemented in our lab using flexible adhesive heating elements paired with a power supply and PID controller. For pH stabilization in systems that lack a 5% CO₂ atmosphere, we typically supplement the imaging medium with 10–25 mM HEPES buffer. In the revised manuscript, we will introduce a modified sample chamber capable of maintaining user-specified temperatures, along with detailed assembly instructions. We will also include representative live-cell imaging data to demonstrate the feasibility of in vitro imaging using this system.

Reviewer #2 (Public review):

Summary:

The authors present Altair-LSFM (Light Sheet Fluorescence Microscope), a high-resolution, open-source microscope, that is relatively easy to align and construct and achieves sub-cellular resolution. The authors developed this microscope to fill a perceived need that current open-source systems are primarily designed for large specimens and lack sub-cellular resolution or are difficult to construct and align, and are not stable. While commercial alternatives exist that offer sub-cellular resolution, they are expensive. The authors' manuscript centers around comparisons to the highly successful lattice light-sheet microscope, including the choice of detection and excitation objectives. The authors thus claim that there remains a critical need for high-resolution, economical, and easy-to-implement LSFM systems.

Strengths:

The authors succeed in their goals of implementing a relatively low-cost (~ USD 150K) open-source microscope that is easy to align. The ease of alignment rests on using custom-designed baseplates with dowel pins for precise positioning of optics based on computer analysis of opto-mechanical tolerances, as well as the optical path design. They simplify the excitation optics over Lattice light-sheet microscopes by using a Gaussian beam for illumination while maintaining lateral and axial resolutions of 235 and 350 nm across a 260-um field of view after deconvolution. In doing so they rest on foundational principles of optical microscopy that what matters for lateral resolution is the numerical aperture of the detection objective and proper sampling of the image field on to the detection, and the axial resolution depends on the thickness of the light-sheet when it is thinner than the depth of field of the detection objective. This concept has unfortunately not been completely clear to users of high-resolution light-sheet microscopes and is thus a valuable demonstration. The microscope is controlled by an open-source software, Navigate, developed by the authors, and it is thus foreseeable that different versions of this system could be implemented depending on experimental needs while maintaining easy alignment and low cost. They demonstrate system performance successfully by characterizing their sheet, point-spread function, and visualization of sub-cellular structures in mammalian cells, including microtubules, actin filaments, nuclei, and the Golgi apparatus.

We thank the reviewer for their thoughtful summary of our work. We are pleased that the foundational optical principles, design rationale, and emphasis on accessibility came through clearly. We agree that the approach used to construct the microscope is highly modular, and we anticipate that these design principles will serve as the basis for additional system variants tailored to specific biological samples and experimental contexts. To support this, we provide all Zemax simulations and CAD files openly on our GitHub repository, enabling advanced users to build upon our design and create new functional variants of the Altair system.

Weaknesses:

There is a fixation on comparison to the first-generation lattice light-sheet microscope, which has evolved significantly since then:

(1) The authors claim that commercial lattice light-sheet microscopes (LLSM) are "complex, expensive, and alignment intensive", I believe this sentence applies to the open-source version of LLSM, which was made available for wide dissemination. Since then, a commercial solution has been provided by 3i, which is now being used in multiple cores and labs but does require routine alignments. However, Zeiss has also released a commercial turn-key system, which, while expensive, is stable, and the complexity does not interfere with the experience of the user. Though in general, statements on ease of use and stability might be considered anecdotal and may not belong in a scientific article, unreferenced or without data.

The referee is correct that our comparisons reference the original LLSM design, which was simultaneously disseminated as an open-source platform and commercialized by 3i. While we acknowledge that newer variants of LLSM have been developed—including systems incorporating adaptive optics[3] and the MOSAIC platform (which remains unpublished)—the original implementation remains the most widely described and cited in the literature. It is therefore the most appropriate point of comparison for contextualizing Altair’s performance, complexity, and accessibility. Importantly, this version of LLSM is far from obsolete; it continues to be one of the most commonly used imaging systems at Janelia Research Campus’s Advanced Imaging Center.

We acknowledge that more recent commercial implementation by Zeiss has addressed several of the practical limitations associated with the original design. In particular, we agree that the Zeiss Lattice Lightsheet 7 system, which integrates a meniscus lens to facilitate oblique imaging through a coverslip, offers a user-friendly experience—albeit with a modest tradeoff in resolution (reported deskewed resolution: 330 nm × 330 nm × 500–1000 nm).

While we recognize that statements on usability and stability can be subjective, one objective proxy for system complexity is the number of optical elements that require precise alignment during assembly. The original LLSM setup includes approximately 29 optical components that must each be carefully positioned laterally, angularly, and coaxially along the optical path. In contrast, the first-generation Altair system contains only 9 such elements. By this metric, Altair is considerably simpler to assemble and align, supporting our overarching goal of making high-resolution light-sheet imaging more accessible to non-specialist laboratories. In the revised manuscript, we will clarify the scope of our comparison and provide more precise language about what we mean by complexity (e.g., number of optical elements needed to align).

(2) One of the major limitations of the first generation LLSM was the use of a 5 mm coverslip, which was a hinderance for many users. However, the Zeiss system elegantly solves this problem, and so does Oblique Plane Microscopy (OPM), while the Altair-LSFM retains this feature, which may dissuade widespread adoption. This limitation and how it may be overcome in future iterations is not discussed.

We agree that the use of 5 mm diameter coverslips, while enabling high-NA imaging in the current Altair-LSFM configuration, may serve as an inconvenience for many users. We will discuss this more explicitly in the revised manuscript. Specifically, we note that changing the detection objective is sufficient to eliminate the need for a 5 mm coverslip. For example, as demonstrated in Moore et al., Lab Chip 2021, pairing the Zeiss W Plan-Apochromat 20x/1.0 objective with the Thorlabs TL20X-MPL allows imaging beyond the physical surfaces of both objectives, removing the constraint imposed by small-format coverslips[1]. In the revised manuscript, we will propose this modification as a straightforward path for increasing compatibility with more conventional sample mounting formats.

(3) Further, on the point of sample flexibility, all generations of the LLSM, and by the nature of its design, the OPM, can accommodate live-cell imaging with temperature, gas, and humidity control. It is unclear how this would be implemented with the current sample chamber. This limitation would severely limit use cases for cell biologists, for which this microscope is designed. There is no discussion on this limitation or how it may be overcome in future iterations.

We appreciate the reviewer’s emphasis on the importance of environmental control for live-cell imaging applications. It is worth noting that the original LLSM design, including the system commercialized by 3i, provided temperature control only, without integrated gas or humidity regulation. Despite this, it has been successfully used by a wide range of scientists to generate important biological insights.

We agree that both OPM and the Zeiss implementation of LLSM offer clear advantages in terms of environmental control, as we previously discussed in detail in Sapoznik et al., eLife, 2020[4]. However, assembly of high numerical aperture OPM systems is highly technical, and no open-source variant of OPM delivers sub-cellular scale resolution yet.

(4) The authors' comparison to LLSM is constrained to the "square" lattice, which, as they point out, is the most used optical lattice (though this also might be considered anecdotal). The LLSM original design, however, goes far beyond the square lattice, including hexagonal lattices, the ability to do structured illumination, and greater flexibility in general in terms of light-sheet tuning for different experimental needs, as well as not being limited to just sample scanning. Thus, the Alstair-LSFM cannot compare to the original LLSM in terms of versatility, even if comparisons to the resolution provided by the square lattice are fair.

We thank the reviewer for this comment. It is true that our discussion focused primarily on the square lattice implementation of LLSM. While this could be viewed as a subset of the system’s broader capabilities, we chose this focus intentionally, as the square lattice remains by far the most commonly used variant in practice. Even in the original LLSM publication, 16 out of 20 figure subpanels utilized the square lattice, with only one panel each representing the hexagonal lattice in SIM mode, a standard Bessel beam in incoherent SIM mode, a hex lattice in dithered mode, and a single Bessel in dithered mode. This usage pattern largely reflects the operational simplicity of the square lattice: it minimizes sidelobe growth and enables more straightforward alignment and data processing compared to hexagonal or structured illumination modes.

In 2019, we performed an exhaustive accounting of published illumination modes in LLSM and found that the SIM mode had only been used in two additional peer-reviewed publications at that time. We will consider updating this table in the revised manuscript and will expand our discussion to acknowledge the broader flexibility of the LLSM platform—including its capacity for structured illumination and alternative light-sheet geometries. However, we will also emphasize that, despite these advanced capabilities, the square lattice remains the dominant mode used by the community and therefore serves as a fair and practical benchmark for comparison.

(5) There is no demonstration of the system's live-imaging capabilities or temporal resolution, which is the main advantage of existing light-sheet systems.

In the revised manuscript, we will include a demonstration of live-cell imaging to directly validate the system’s suitability for dynamic biological applications. We will also characterize the temporal resolution of the system. As a sample-scanning microscope, the imaging speed is primarily limited by the performance of the Z-piezo stage. For simplicity and reduced optoelectronic complexity, we currently power the piezo through the ASI Tiger Controller. We will expand the supplementary material to describe the design criteria behind this choice, including potential trade-offs, and provide data quantifying the achievable volume rates under typical operating conditions.

While the microscope is well designed and completely open source, it will require experience with optics, electronics, and microscopy to implement and align properly. Experience with custom machining or soliciting a machine shop is also necessary. Thus, in my opinion, it is unlikely to be implemented by a lab that has zero prior experience with custom optics or can hire someone who does. Altair-LSFM may not be as easily adaptable or implementable as the authors describe or perceive in any lab that is interested, even if they can afford it. The authors indicate they will offer "workshops," but this does not necessarily remove the barrier to entry or lower it, perhaps as significantly as the authors describe.

We appreciate the reviewer’s perspective and agree that building any high-performance custom microscope—Altair-LSFM included—requires a baseline familiarity with optics and instrumentation. Our goal is not to eliminate this requirement entirely, but to significantly reduce the technical and logistical barriers that typically accompany custom light-sheet microscope construction.

Importantly, no machining experience or in-house fabrication capabilities are required—users can simply submit provided design files and specifications directly to the vendor. We will make this process as straightforward as possible by supplying detailed instructions, recommended materials, and vendor-ready files. Additionally, we draw encouragement from the success of related efforts such as mesoSPIM, which has seen over 30 successful implementations worldwide using a similar model of exhaustive online documentation, open-source control software, and community support through user meetings and workshops.

We recognize that documentation alone is not always sufficient, and we are committed to further lowering barriers to adoption. To this end, we are actively working with commercial vendors to streamline procurement and reduce the logistical burden on end users. Additionally, Altair-LSFM is supported by a Biomedical Technology Development and Dissemination (BTDD) grant, which provides dedicated resources for hosting workshops, offering real-time community support, and generating supplementary materials such as narrated video tutorials. We will expand our discussion in the revised manuscript to better acknowledge these implementation challenges and outline our ongoing strategies for supporting a broad and diverse user base.

There is a claim that this design is easily adaptable. However, the requirement of custom-machined baseplates and in silico optimization of the optical path basically means that each new instrument is a new design, even if the Navigate software can be used. It is unclear how Altair-LSFM demonstrates a modular design that reduces times from conception to optimization compared to previous implementations.

We appreciate the reviewer’s comment and agree that our language regarding adaptability may have been too strong. It was not our intention to suggest that the system can be easily modified without prior experience. Meaningful adaptations of the optical or mechanical design would require users to have expertise in optical layout, optomechanical design, and alignment.

That said, for labs with sufficient expertise, we aim to facilitate such modifications by providing comprehensive resources—including detailed Zemax simulations, CAD models, and alignment documentation. These materials are intended to reduce the development burden for those seeking to customize the platform for specific experimental needs.

In the revised manuscript, we will clarify this point and explicitly state in the discussion what technical expertise is required to modify the system. We will also revise our language around adaptability to better reflect the intended audience and realistic scope of customization.

Reviewer #3 (Public review):

Summary:

This manuscript introduces a high-resolution, open-source light-sheet fluorescence microscope optimized for sub-cellular imaging.

The system is designed for ease of assembly and use, incorporating a custom-machined baseplate and in silico optimized optical paths to ensure robust alignment and performance. The authors demonstrate lateral and axial resolutions of ~235 nm and ~350 nm after deconvolution, enabling imaging of sub-diffraction structures in mammalian cells.

The important feature of the microscope is the clever and elegant adaptation of simple gaussian beams, smart beam shaping, galvo pivoting and high NA objectives to ensure a uniform thin light-sheet of around 400 nm in thickness, over a 266 micron wide Field of view, pushing the axial resolution of the system beyond the regular diffraction limited-based tradeoffs of light-sheet fluorescence microscopy.

Compelling validation using fluorescent beads and multicolor cellular imaging highlights the system's performance and accessibility. Moreover, a very extensive and comprehensive manual of operation is provided in the form of supplementary materials. This provides a DIY blueprint for researchers who want to implement such a system.

Strengths:

(1) Strong and accessible technical innovation: With an elegant combination of beam shaping and optical modelling, the authors provide a high-resolution light-sheet system that overcomes the classical light-sheet tradeoff limit of a thin light-sheet and a small field of view. In addition, the integration of in silico modelling with a custom-machined baseplate is very practical and allows for ease of alignment procedures. Combining these features with the solid and super-extensive guide provided in the supplementary information, this provides a protocol for replicating the microscope in any other lab.

(2) Impeccable optical performance and ease of mounting of samples: The system takes advantage of the same sample-holding method seen already in other implementations, but reduces the optical complexity. At the same time, the authors claim to achieve similar lateral and axial resolution to Lattice-light-sheet microscopy (although without a direct comparison (see below in the "weaknesses" section). The optical characterization of the system is comprehensive and well-detailed. Additionally, the authors validate the system imaging sub-cellular structures in mammalian cells.

(3) Transparency and comprehensiveness of documentation and resources: A very detailed protocol provides detailed documentation about the setup, the optical modeling, and the total cost.

Weaknesses:

(1) Limited quantitative comparisons: Although some qualitative comparison with previously published systems (diSPIM, lattice light-sheet) is provided throughout the manuscript, some side-by-side comparison would be of great benefit for the manuscript, even in the form of a theoretical simulation. While having a direct imaging comparison would be ideal, it's understandable that this goes beyond the interest of the paper; however, a table referencing image quality parameters (taken from the literature), such as signal-to-noise ratio, light-sheet thickness, and resolutions, would really enhance the features of the setup presented. Moreover, based also on the necessity for optical simplification, an additional comment on the importance/difference of dual objective/single objective light-sheet systems could really benefit the discussion.

In the revised manuscript, we will expand our discussion to include a broader range of light-sheet microscope designs and imaging modes, including both single- and dual-objective configurations. We agree that highlighting the trade-offs between these approaches—such as working distance, sample geometry constraints, and alignment complexity—will enhance the overall context and utility of the manuscript.

To further aid comparison, we will include a summary table referencing key image quality parameters such as lateral and axial resolution, and illumination beam NA for Altair-LSFM. Where available, we will reference values from published work—such as the axial resolution reported in Valm et al. (Nature, 2017)—to provide a clearer benchmark. Because such comparisons can be technically nuanced, especially when comparing across systems with different geometries and sample mounting constraints, we will also include a supplementary note outlining the assumptions and limitations of these comparisons.

(2) Limitation to a fixed sample: In the manuscript, there is no mention of incubation temperature, CO₂ regulation, Humidity control, or possible integration of commercial environmental control systems. This is a major limitation for an imaging technique that owes its popularity to fast, volumetric, live-cell imaging of biological samples.

We thank the reviewer for highlighting this important consideration. In the revised manuscript, we will provide a detailed description of how temperature control can be implemented using flexible adhesive heating elements, a power supply, and a PID controller. Step-by-step assembly instructions and recommended components will be included to facilitate adoption by users interested in live-cell imaging. We also note that most light-sheet microscopy systems capable of sub-cellular resolution—including the original LLSM design, diSPIM, and ASLM—typically do not incorporate integrated CO₂ or humidity control. These systems often rely on HEPES-buffered media to maintain pH stability, which is generally sufficient for short- to intermediate-term imaging. While full environmental control may be necessary for extended time-lapse studies, it is not a prerequisite for high-resolution volumetric imaging in many applications. Nonetheless, we will include a discussion of the challenges associated with adding CO₂ and humidity control to open or semi-enclosed architectures like Altair-LSFM, and outline potential future paths for integration with commercial incubation systems.

(3) System cost and data storage cost: While the system presented has the advantage of being open-source, it remains relatively expensive (considering the 150k without laser source and optical table, for example). The manuscript could benefit from a more direct comparison of the performance/cost ratio of existing systems, considering academic settings with budgets that most of the time would not allow for expensive architectures. Moreover, it would also be beneficial to discuss the adaptability of the system, in case a 30k objective could not be feasible. Will this system work with different optics (with the obvious limitations coming with the lower NA objective)? This could be an interesting point of discussion. Adaptability of the system in case of lower budgets or more cost-effective choices, depending on the needs.

We thank the reviewer for raising this important point. First, we would like to clarify that the quoted $150k cost estimate includes the optical table and laser source. We apologize for any confusion and will communicate this more effectively in the revised manuscript.

We agree that adaptability is a key concern, especially in academic settings with limited budgets. The detection path can be readily altered depending on experimental needs and cost constraints. For example, in our discussion of alternatives to the 5 mm coverslip geometry, we will describe how switching to a Zeiss W Plan-Apochromat 20x/1.0 in combination with a compatible excitation objective allows high-resolution imaging while accommodating more conventional sample formats. We will expand this to include cost-effective alternatives as well.

We will also expand our discussion on cost-reduction strategies and the associated trade-offs. These include replacing motorized stages with manual ones, omitting the filter wheel in favor of a multi-band emission filter, or using industrial-grade cameras in place of scientific CMOS detectors. While each change entails some loss in functionality or sensitivity, such modifications allow users to tailor the system to their specific budget and application.

Finally, we recognize the challenge in communicating exact costs of commercial systems due to variability in configuration and pricing. Nonetheless, we will include approximate figures where possible and note that comparable commercial systems—such as LLSM platforms from 3i and Zeiss—are several-fold more expensive than the system presented here.

Last, not much is said about the need for data storage. Light-sheet microscopy's bottleneck is the creation of increasingly large datasets, and it could be beneficial to discuss more about the storage needs and the quantity of data generated.

Data storage is indeed a critical consideration in light-sheet microscopy. In the revised manuscript, we will provide a note outlining typical volume dimensions for live-cell imaging experiments along with the associated data overhead. This will include estimates for voxel counts, bit depth, time-lapse acquisitions, and multi-channel datasets to help users anticipate storage needs. We will also briefly discuss strategies for managing large datasets, file types and compression formats.

Conclusion:

Altair-LSFM represents a well-engineered and accessible light-sheet system that addresses a longstanding need for high-resolution, reproducible, and affordable sub-cellular light-sheet imaging. While some aspects-comparative benchmarking and validation, limitation for fixed samples-would benefit from further development, the manuscript makes a compelling case for Altair-LSFM as a valuable contribution to the open microscopy scientific community.

References

(1) Moore, R. P. et al. A multi-functional microfluidic device compatible with widefield and light sheet microscopy. Lab Chip 22, 136-147 (2021). https://doi.org/10.1039/d1lc00600b

(2) Lamb, J. R., Mestre, M. C., Lancaster, M. & Manton, J. D. Direct-view oblique plane microscopy. Optica 12, 469-472 (2025). https://doi.org/10.1364/OPTICA.558420

(3) Liu, T. L. et al. Observing the cell in its native state: Imaging subcellular dynamics in multicellular organisms. Science 360 (2018). https://doi.org/10.1126/science.aaq1392

(4) Sapoznik, E. et al. A versatile oblique plane microscope for large-scale and high-resolution imaging of subcellular dynamics. eLife 9 (2020). https://doi.org/10.7554/eLife.57681

(5) Huisken, J. & Stainier, D. Y. Even fluorescence excitation by multidirectional selective plane illumination microscopy (mSPIM). Opt Lett 32, 2608-2610 (2007). https://doi.org/10.1364/ol.32.002608

(6) Ricci, P. et al. Removing striping artifacts in light-sheet fluorescence microscopy: a review. Prog Biophys Mol Biol 168, 52-65 (2022). https://doi.org/10.1016/j.pbiomolbio.2021.07.003

eLife Assessment

This useful study presents Altair-LSFM, a solid and well-documented implementation of a light-sheet fluorescence microscope (LSFM) designed for accessibility and cost reduction. While the approach offers strengths such as the use of custom-machined baseplates and detailed assembly instructions, its overall impact is limited by the lack of live-cell imaging capabilities and the absence of a clear, quantitative comparison to existing LSFM platforms. As such, although technically competent, the broader utility and uptake of this system by the community may be limited.

Reviewer #1 (Public review):

Summary:

The article presents the details of the high-resolution light-sheet microscopy system developed by the group. In addition to presenting the technical details of the system, its resolution has been characterized and its functionality demonstrated by visualizing subcellular structures in a biological sample.

Strengths:

(1) The article includes extensive supplementary material that complements the information in the main article.

(2) However, in some sections, the information provided is somewhat superficial.

Weaknesses:

(1) Although a comparison is made with other light-sheet microscopy systems, the presented system does not represent a significant advance over existing systems. It uses high numerical aperture objectives and Gaussian beams, achieving resolution close to theoretical after deconvolution. The main advantage of the presented system is its ease of construction, thanks to the design of a perforated base plate.

(2) Using similar objectives (Nikon 25x and Thorlabs 20x), the results obtained are similar to those of the LLSM system (using a Gaussian beam without laser modulation). However, the article does not mention the difficulties of mounting the sample in the implemented configuration.

(3) The authors present a low-cost, open-source system. Although they provide open source code for the software (navigate), the use of proprietary electronics (ASI, NI, etc.) makes the system relatively expensive. Its low cost is not justified.

(4) The fibroblast images provided are of exceptional quality. However, these are fixed samples. The system lacks the necessary elements for monitoring cells in vivo, such as temperature or pH control.

Reviewer #2 (Public review):

Summary:

The authors present Altair-LSFM (Light Sheet Fluorescence Microscope), a high-resolution, open-source microscope, that is relatively easy to align and construct and achieves sub-cellular resolution. The authors developed this microscope to fill a perceived need that current open-source systems are primarily designed for large specimens and lack sub-cellular resolution or are difficult to construct and align, and are not stable. While commercial alternatives exist that offer sub-cellular resolution, they are expensive. The authors' manuscript centers around comparisons to the highly successful lattice light-sheet microscope, including the choice of detection and excitation objectives. The authors thus claim that there remains a critical need for high-resolution, economical, and easy-to-implement LSFM systems.

Strengths:

The authors succeed in their goals of implementing a relatively low-cost (~ USD 150K) open-source microscope that is easy to align. The ease of alignment rests on using custom-designed baseplates with dowel pins for precise positioning of optics based on computer analysis of opto-mechanical tolerances, as well as the optical path design. They simplify the excitation optics over Lattice light-sheet microscopes by using a Gaussian beam for illumination while maintaining lateral and axial resolutions of 235 and 350 nm across a 260-um field of view after deconvolution. In doing so they rest on foundational principles of optical microscopy that what matters for lateral resolution is the numerical aperture of the detection objective and proper sampling of the image field on to the detection, and the axial resolution depends on the thickness of the light-sheet when it is thinner than the depth of field of the detection objective. This concept has unfortunately not been completely clear to users of high-resolution light-sheet microscopes and is thus a valuable demonstration. The microscope is controlled by an open-source software, Navigate, developed by the authors, and it is thus foreseeable that different versions of this system could be implemented depending on experimental needs while maintaining easy alignment and low cost. They demonstrate system performance successfully by characterizing their sheet, point-spread function, and visualization of sub-cellular structures in mammalian cells, including microtubules, actin filaments, nuclei, and the Golgi apparatus.

Weaknesses:

There is a fixation on comparison to the first-generation lattice light-sheet microscope, which has evolved significantly since then:

(1) The authors claim that commercial lattice light-sheet microscopes (LLSM) are "complex, expensive, and alignment intensive", I believe this sentence applies to the open-source version of LLSM, which was made available for wide dissemination. Since then, a commercial solution has been provided by 3i, which is now being used in multiple cores and labs but does require routine alignments. However, Zeiss has also released a commercial turn-key system, which, while expensive, is stable, and the complexity does not interfere with the experience of the user. Though in general, statements on ease of use and stability might be considered anecdotal and may not belong in a scientific article, unreferenced or without data.

(2) One of the major limitations of the first generation LLSM was the use of a 5 mm coverslip, which was a hinderance for many users. However, the Zeiss system elegantly solves this problem, and so does Oblique Plane Microscopy (OPM), while the Altair-LSFM retains this feature, which may dissuade widespread adoption. This limitation and how it may be overcome in future iterations is not discussed.

(3) Further, on the point of sample flexibility, all generations of the LLSM, and by the nature of its design, the OPM, can accommodate live-cell imaging with temperature, gas, and humidity control. It is unclear how this would be implemented with the current sample chamber. This limitation would severely limit use cases for cell biologists, for which this microscope is designed. There is no discussion on this limitation or how it may be overcome in future iterations.

(4) The authors' comparison to LLSM is constrained to the "square" lattice, which, as they point out, is the most used optical lattice (though this also might be considered anecdotal). The LLSM original design, however, goes far beyond the square lattice, including hexagonal lattices, the ability to do structured illumination, and greater flexibility in general in terms of light-sheet tuning for different experimental needs, as well as not being limited to just sample scanning. Thus, the Alstair-LSFM cannot compare to the original LLSM in terms of versatility, even if comparisons to the resolution provided by the square lattice are fair.

(5) There is no demonstration of the system's live-imaging capabilities or temporal resolution, which is the main advantage of existing light-sheet systems.

While the microscope is well designed and completely open source, it will require experience with optics, electronics, and microscopy to implement and align properly. Experience with custom machining or soliciting a machine shop is also necessary. Thus, in my opinion, it is unlikely to be implemented by a lab that has zero prior experience with custom optics or can hire someone who does. Altair-LSFM may not be as easily adaptable or implementable as the authors describe or perceive in any lab that is interested, even if they can afford it. The authors indicate they will offer "workshops," but this does not necessarily remove the barrier to entry or lower it, perhaps as significantly as the authors describe.

There is a claim that this design is easily adaptable. However, the requirement of custom-machined baseplates and in silico optimization of the optical path basically means that each new instrument is a new design, even if the Navigate software can be used. It is unclear how Altair-LSFM demonstrates a modular design that reduces times from conception to optimization compared to previous implementations.

Reviewer #3 (Public review):

Summary:

This manuscript introduces a high-resolution, open-source light-sheet fluorescence microscope optimized for sub-cellular imaging.

The system is designed for ease of assembly and use, incorporating a custom-machined baseplate and in silico optimized optical paths to ensure robust alignment and performance. The authors demonstrate lateral and axial resolutions of ~235 nm and ~350 nm after deconvolution, enabling imaging of sub-diffraction structures in mammalian cells.

The important feature of the microscope is the clever and elegant adaptation of simple gaussian beams, smart beam shaping, galvo pivoting and high NA objectives to ensure a uniform thin light-sheet of around 400 nm in thickness, over a 266 micron wide Field of view, pushing the axial resolution of the system beyond the regular diffraction limited-based tradeoffs of light-sheet fluorescence microscopy.

Compelling validation using fluorescent beads and multicolor cellular imaging highlights the system's performance and accessibility. Moreover, a very extensive and comprehensive manual of operation is provided in the form of supplementary materials. This provides a DIY blueprint for researchers who want to implement such a system.

Strengths:

(1) Strong and accessible technical innovation:

With an elegant combination of beam shaping and optical modelling, the authors provide a high-resolution light-sheet system that overcomes the classical light-sheet tradeoff limit of a thin light-sheet and a small field of view. In addition, the integration of in silico modelling with a custom-machined baseplate is very practical and allows for ease of alignment procedures. Combining these features with the solid and super-extensive guide provided in the supplementary information, this provides a protocol for replicating the microscope in any other lab.

(2) Impeccable optical performance and ease of mounting of samples:

The system takes advantage of the same sample-holding method seen already in other implementations, but reduces the optical complexity. At the same time, the authors claim to achieve similar lateral and axial resolution to Lattice-light-sheet microscopy (although without a direct comparison (see below in the "weaknesses" section). The optical characterization of the system is comprehensive and well-detailed. Additionally, the authors validate the system imaging sub-cellular structures in mammalian cells.

(3) Transparency and comprehensiveness of documentation and resources:

A very detailed protocol provides detailed documentation about the setup, the optical modeling, and the total cost.

Weaknesses:

(1) Limited quantitative comparisons:

Although some qualitative comparison with previously published systems (diSPIM, lattice light-sheet) is provided throughout the manuscript, some side-by-side comparison would be of great benefit for the manuscript, even in the form of a theoretical simulation. While having a direct imaging comparison would be ideal, it's understandable that this goes beyond the interest of the paper; however, a table referencing image quality parameters (taken from the literature), such as signal-to-noise ratio, light-sheet thickness, and resolutions, would really enhance the features of the setup presented. Moreover, based also on the necessity for optical simplification, an additional comment on the importance/difference of dual objective/single objective light-sheet systems could really benefit the discussion.

(2) Limitation to a fixed sample:

In the manuscript, there is no mention of incubation temperature, CO₂ regulation, Humidity control, or possible integration of commercial environmental control systems. This is a major limitation for an imaging technique that owes its popularity to fast, volumetric, live-cell imaging of biological samples.

(3) System cost and data storage cost:

While the system presented has the advantage of being open-source, it remains relatively expensive (considering the 150k without laser source and optical table, for example). The manuscript could benefit from a more direct comparison of the performance/cost ratio of existing systems, considering academic settings with budgets that most of the time would not allow for expensive architectures. Moreover, it would also be beneficial to discuss the adaptability of the system, in case a 30k objective could not be feasible. Will this system work with different optics (with the obvious limitations coming with the lower NA objective)? This could be an interesting point of discussion. Adaptability of the system in case of lower budgets or more cost-effective choices, depending on the needs.

Last, not much is said about the need for data storage. Light-sheet microscopy's bottleneck is the creation of increasingly large datasets, and it could be beneficial to discuss more about the storage needs and the quantity of data generated.

Conclusion:

Altair-LSFM represents a well-engineered and accessible light-sheet system that addresses a longstanding need for high-resolution, reproducible, and affordable sub-cellular light-sheet imaging. While some aspects-comparative benchmarking and validation, limitation for fixed samples-would benefit from further development, the manuscript makes a compelling case for Altair-LSFM as a valuable contribution to the open microscopy scientific community.

eLife Assessment

This important study identifies a novel Legionella effector, Lfat1, which binds F-actin via a coiled-coil domain and structurally resembles the RID toxin, with cryo-EM revealing key interactions mediated by a hydrophobic helical hairpin. While the study is mostly complete and has compelling data, a few minor changes are recommended.

Reviewer #1 (Public review):

The manuscript by Zeng et al. describes the discovery of an F-actin-binding Legionella pneumophila effector, which they term Lfat1. Lfat1 contains a putative fatty acyltransferase domain that structurally resembles the Rho-GTPase Inactivation (RID) domain toxin from Vibrio vulnificus, which targets small G-proteins. Additionally, Lfat1 contains a coiled-coil (CC) domain.

The authors identified Lfat1 as an actin-associated protein by screening more than 300 Legionella effectors, expressed as GFP-fusion proteins, for their co-localization with actin in HeLa cells. Actin binding is mediated by the CC domain, which specifically binds to F-actin in a 1:1 stoichiometry. Using cryo-EM, the authors determined a high-quality structure of F-actin filaments bound to the actin-binding domain (ABD) of Lfat1. The structure reveals that actin binding is mediated through a hydrophobic helical hairpin within the ABD (residues 213-279). A Y240A mutation within this region increases the apparent dissociation constant by two orders of magnitude, indicating a critical role for this residue in actin interaction.

The ABD alone was also shown to strongly associate with F-actin upon overexpression in cells. The authors used a truncated version of the Lfat1 ABD to engineer an F-actin-binding probe, which can be used in a split form. Finally, they demonstrate that full-length Lfat1, when overexpressed in cells, fatty acylates host small G-proteins, likely on lysine residues.

While this is a solid study, the authors should consider the following points when preparing a revised manuscript:

Major points:

(1) Legionella effectors are often activated by binding to eukaryote-specific host factors, including actin. The authors should test the following: a) whether Lfat1 can fatty acylate small G-proteins in vitro; b) whether this activity is dependent on actin binding; and c) whether expression of the Y240A mutant in mammalian cells affects the fatty acylation of Rac3 (Figure 6B), or other small G-proteins.

(2) It should be demonstrated that lysine residues on small G-proteins are indeed targeted by Lfat1. Ideally, the functional consequences of these modifications should also be investigated. For example, does fatty acylation of G-proteins affect GTPase activity or binding to downstream effectors?

(3) Line 138: Can the authors clarify whether the Lfat1 ABD induces bundling of F-actin filaments or promotes actin oligomerization? Does the Lfat1 ABD form multimers that bring multiple filaments together? If Lfat1 induces actin oligomerization, this effect should be experimentally tested and reported. Additionally, the impact of Lfat1 binding on actin filament stability should be assessed. This is particularly important given the proposed use of the ABD as an actin probe.

(4) Line 180: I think it's too premature to refer to the interaction as having "high specificity and affinity." We really don't know what else it's binding to.

(5) The authors should reconsider the color scheme used in the structural figures, particularly in Figures 2D and S4.

(6) In Figure 3E, the WT curve fits the data poorly, possibly because the actin concentration exceeds the Kd of the interaction. It might fit better to a quadratic.

(7) The authors propose that the individual helices of the Lfat1 ABD could be expressed on separate proteins and used to target multi-component biological complexes to F-actin by genetically fusing each component to a split alpha-helix. This is an intriguing idea, but it should be tested as a proof of concept to support its feasibility and potential utility.

Reviewer #2 (Public review):

Summary:

The manuscript by Zheng et al reports the structural and biochemical study of novel effectors from the bacterial pathogen Legionella pneumophila. The authors continued from results from their earlier screening for L. pneumophila proteins that affect host F-actin dynamics to show that Llfat1 (Lpg1387) interacts with actin via a novel actin-binding domain (ABD). The authors also determined the structure of the Lfat1 ABD-F-actin complex, which allowed them to develop this ABD as a probe for F-actin. Finally, the authors demonstrated that Llfat1 is a lysine fatty acyltransferase that targets several small GTPases in host cells.

Strengths:

This is a very complete work that shows the structure of a novel bacterial actin-binding protein in complex with F-actin, and the biochemical activity of the protein was also revealed. Overall, this is a very exciting study and should be of great interest to scientists in both bacterial pathogenesis and the actin cytoskeleton of eukaryotic cells.

Weaknesses:

(1) The authors should use biochemical reactions to analyze the KFAT of Llfat1 on one or two small GTPases shown to be modified by this effector in cellulo. Such reactions may allow them to determine the role of actin binding in its biochemical activity. This notion is particularly relevant in light of recent studies that actin is a co-factor for the activity of LnaB and Ceg14 (PMID: 39009586; PMID: 38776962; PMID: 40394005). In addition, the study should be discussed in the context of these recent findings on the role of actin in the activity of L. pneumophila effectors.

(2) The development of the ABD domain of Llfat1 as an F-actin domain is a nice extension of the biochemical and structural experiments. The authors need to compare the new probe to those currently commonly used ones, such as Lifeact, in labeling of the actin cytoskeleton structure.

eLife Assessment

The work provides important insights into how this lncRNA regulates gene expression via complex mechanisms, however, the biological relevance awaits validation in other models. This paper provides extensive and carefully analysed data that is of value in efforts to understand the role of the lncRNA EPB41L4A-AS1 in a human cell line. The data is generally convincing and supported by clever integrative analysis; however, the known extensive artefacts from individual Gapmer oligonucleotides cast some doubt over the interpretation of those experiments where only one targeting and one control Gapmer are used.

Reviewer #1 (Public review):

Monziani and Ulitsky present a large and exhaustive study on the lncRNA EPB41L4A-AS1 using a variety of genomic methods. They uncover a rather complex picture of an RNA transcript that appears to act via diverse pathways to regulate the expression of large numbers of genes, including many snoRNAs. The activity of EPB41L4A-AS1 seems to be intimately linked with the protein SUB1, via both direct physical interactions and direct/indirect of SUB1 mRNA expression.

The study is characterised by thoughtful, innovative, integrative genomic analysis. It is shown that EPB41L4A-AS1 interacts with SUB1 protein and that this may lead to extensive changes in SUB1's other RNA partners. Disruption of EPB41L4A-AS1 leads to widespread changes in non-polyA RNA expression, as well as local cis changes. At the clinical level, it is possible that EPB41L4A-AS1 plays disease-relevant roles, although these seem to be somewhat contradictory with evidence supporting both oncogenic and tumour suppressive activities.

A couple of issues could be better addressed here. Firstly, the copy number of EPB41L4A-AS1 is an important missing piece of the puzzle. It is apparently highly expressed in the FISH experiments. To get an understanding of how EPB41L4A-AS1 regulates SUB1, an abundant protein, we need to know the relative stoichiometry of these two factors. Secondly, while many of the experiments use two independent Gapmers for EPB41L4A-AS1 knockdown, the RNA-sequencing experiments apparently use just one, with one negative control (?). Evidence is emerging that Gapmers produce extensive off-target gene expression effects in cells, potentially exceeding the amount of on-target changes arising through the intended target gene. Therefore, it is important to estimate this through the use of multiple targeting and non-targeting ASOs, if one is to get a true picture of EPB41L4A-AS1 target genes. In this Reviewer's opinion, this casts some doubt over the interpretation of RNA-seq experiments until that work is done. Nonetheless, the Authors have designed thorough experiments, including overexpression rescue constructs, to quite confidently assess the role of EPB41L4A-AS1 in snoRNA expression.

It is possible that EPB41L4A-AS1 plays roles in cancer, either as an oncogene or a tumour suppressor. However, it will in the future be important to extend these observations to a greater variety of cell contexts.

This work is valuable in providing an extensive and thorough analysis of the global mechanisms of an important regulatory lncRNA and highlights the complexity of such mechanisms via cis and trans regulation and extensive protein interactions.

Reviewer #2 (Public review):

Summary:

In this manuscript, Monziani et al. identified long noncoding RNAs (lncRNAs) that act in cis and are coregulated with their target genes located in close genomic proximity. The authors mined the GeneHancer database, and this analysis led to the identification of four lncRNA-target pairs. The authors decided to focus on lncRNA EPB41L4A-AS1.

They thoroughly characterised this lncRNA, demonstrating that it is located in the cytoplasm and the nuclei, and that its expression is altered in response to different stimuli. Furthermore, the authors showed that EPB41L4A-AS1 regulates EPB41L4A transcription, leading to a mild reduction in EPB41L4A protein levels. This was not recapitulated with siRNA-mediated depletion of EPB41L4AAS1. RNA-seq in EPB41L4A-AS1-depleted cells with single LNA revealed 2364 DEGs linked to pathways including the cell cycle, cell adhesion, and inflammatory response. To understand the mechanism of action of EPB41L4A-AS1, the authors mined the ENCODE eCLIP data and identified SUB1 as an lncRNA interactor. The authors also found that the loss of EPB41L4A-AS1 and SUB1 leads to the accumulation of snoRNAs, and that SUB1 localisation changes upon the loss of EPB41L4A-AS1. Finally, the authors showed that EPB41L4A-AS1 deficiency did not change the steady-state levels of SNORA13 nor RNA modification driven by this RNA. The phenotype associated with the loss of EPB41L4A-AS1 is linked to increased invasion and EMT gene signature.

Overall, this is an interesting and nicely done study on the versatile role of EPB41L4A-AS1 and the multifaceted interplay between SUB1 and this lncRNA, but some conclusions and claims need to be supported with additional experiments. My primary concerns are using a single LNA gapmer for critical experiments, increased invasion, and nucleolar distribution of SUB1- in EPB41L4A-AS1-depleted cells. These experiments need to be validated with orthogonal methods.

Strengths:

The authors used complementary tools to dissect the complex role of lncRNA EPB41L4A-AS1 in regulating EPB41L4A, which is highly commendable. There are few papers in the literature on lncRNAs at this standard. They employed LNA gapmers, siRNAs, CRISPRi/a, and exogenous overexpression of EPB41L4A-AS1 to demonstrate that the transcription of EPB41L4A-AS1 acts in cis to promote the expression of EPB41L4A by ensuring spatial proximity between the TAD boundary and the EPB41L4A promoter. At the same time, this lncRNA binds to SUB1 and regulates snoRNA expression and nucleolar biology. Overall, the manuscript is easy to read, and the figures are well presented. The methods are sound, and the expected standards are met.

Weaknesses:

The authors should clarify how many lncRNA-target pairs were included in the initial computational screen for cis-acting lncRNAs and why MCF7 was chosen as the cell line of choice. Most of the data uses a single LNA gapmer targeting EPB41L4A-AS1 lncRNA (eg, Fig. 2c, 3B, and RNA-seq), and the critical experiments should be using at least 2 LNA gapmers. The specificity of SUB1 CUT&RUN is lacking, as well as direct binding of SUB1 to lncRNA EPB41L4A-AS1, which should be confirmed by CLIP qPCR in MCF7 cells. Finally, the role of EPB41L4A-AS1 in SUB1 distribution (Figure 5) and cell invasion (Figure 8) needs to be complemented with additional experiments, which should finally demonstrate the role of this lncRNA in nucleolus and cancer-associated pathways. The use of MCF7 as a single cancer cell line is not ideal.

Reviewer #3 (Public review):

Summary:

In this paper, the authors made some interesting observations that EPB41L4A-AS1 lncRNA can regulate the transcription of both the nearby coding gene and genes on other chromosomes. They started by computationally examining lncRNA-gene pairs by analyzing co-expression, chromatin features of enhancers, TF binding, HiC connectome, and eQTLs. They then zoomed in on four pairs of lncRNA-gene pairs and used LNA antisense oligonucleotides to knock down these lncRNAs. This revealed EPB41L4A-AS1 as the only one that can regulate the expression of its cis-gene target EPB41L4A. By RNA-FISH, the authors found this lncRNA to be located in all three parts of a cell: chromatin, nucleoplasm, and cytoplasm. RNA-seq after LNA knockdown of EPB41L4A-AS1 showed that this increased >1100 genes and decreased >1250 genes, including both nearby genes and genes on other chromosomes. They later found that EPB41L4A-AS1 may interact with SUB1 protein (an RNA-binding protein) to impact the target genes of SUB1. EPB41L4A-AS1 knockdown reduced the mRNA level of SUB1 and altered the nuclear location of SUB1. Later, the authors observed that EPB41L4A-AS1 knockdown caused an increase of snRNAs and snoRNAs, likely via disrupted SUB1 function. In the last part of the paper, the authors conducted rescue experiments that suggested that the full-length, intron- and SNORA13-containing EPB41L4A-AS1 is required to partially rescue snoRNA expression. They also conducted SLAM-Seq and showed that the increased abundance of snoRNAs is primarily due to their hosts' increased transcription and stability. They end with data showing that EPB41L4A-AS1 knockdown reduced MCF7 cell proliferation but increased its migration, suggesting a link to breast cancer progression and/or metastasis.

Strengths:

Overall, the paper is well-written, and the results are presented with good technical rigor and appropriate interpretation. The observation that a complex lncRNA EPB41L4A-AS1 regulates both cis and trans target genes, if fully proven, is interesting and important.

Weaknesses:

The paper is a bit disjointed as it started from cis and trans gene regulation, but later it switched to a partially relevant topic of snoRNA metabolism via SUB1. The paper did not follow up on the interesting observation that there are many potential trans target genes affected by EPB41L4A-AS1 knockdown and there was limited study of the mechanisms as to how these trans genes (including SUB1 or NPM1 genes themselves) are affected by EPB41L4A-AS1 knockdown. There are discrepancies in the results upon EPB41L4A-AS1 knockdown by LNA versus by CRISPR activation, or by plasmid overexpression of this lncRNA.

Author response:

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

Public Reviews:

Reviewer #1 (Public review):

Summary:

Mollá-Albaladejo et al. investigate the neurons downstream of GR64f and Gr66a, called G2Ns. They identify downstream neurons using trans-Tango labeling with RFP and then perform bulk RNA-seq on the RFP-sorted cells. Gene expression is up- or downregulated between the cell populations and between fed and starved states. They specifically identify Leukocinin as a neuropeptide that is upregulated in starved Gr66a cells. Leucokinin cells, identified by a GAL4 line indeed show higher expression when starved, especially in the SEZ. Furthermore, Leucokinin cells colocalize with the transTango signal from downstream neurons of both GRs. This connection is confirmed with GRASP. According to EM data, Leucokinin cells in the SEZ receive a lot of input and connect to many downstream neurons. In behavior experiments performed with flies lacking Leucokinin neurons, flies show reduced responsiveness to sugar and bitter mixtures when starved. The authors suggest that Leucokinin neurons integrate bitter and sugar tastes and that their output is modified by a hunger state.

Strengths:

The authors use a multitude of tools to identify SELK neurons downstream of taste sensory neurons and as starvation-sensitive cells. This study provides an example of how combining genetic labeling, RNA-seq, and EM analysis can be combined to investigate neural circuits.

Weaknesses:

The authors do not show a functional connection between sensory neurons and SELK neurons. Additionally, data from RNA seq, anatomical studies, and EM analysis are sometimes contradictory in terms of connectivity. GRASP signal is not foolproof that cells are synaptically connected.

We appreciate the reviewer’s comments. Unfortunately, we have not successfully demonstrated a functional response of SELK neurons using in vivo calcium imaging with UAS-GCaMP7 (we tried f, m, and s versions), primarily due to challenges in obtaining stable signals. We stimulated GRNs using sucrose, caffeine, or a mixture of both, and maybe even if the concentrations were high, they were not enough to induce a response.

Regarding GRASP, we acknowledge its limitations as a standalone technique for establishing genuine synaptic connections between neurons, as some signals may reflect false positives resulting from the mere proximity of the candidate neurons. To strengthen our findings, we complemented these results by demonstrating the positive colocalization of the Leucokinin antibody signal over the Gr66aGal4>trans-TANGO and Gr64f-Gal4>trans-TANGO (Figure 4), confirming that Leucokinin neurons are indeed postsynaptic to both sweet and bitter GRNs. Moreover, we incorporated BacTrace data to highlight the direct connectivity between sweet and bitter GRNs (now Figure 5E).

In the revised manuscript, we have introduced the active-GRASP technique (Macpherson et al., 2015). In this version of GRASP, the presynaptic half of GFP (GFP 1-10) is fused to synaptobrevin, which becomes accessible in the membrane of the presynaptic neuron within the synaptic cleft upon presynaptic stimulation (in our case, by stimulating with sucrose sweet Gr64f^GRNs and with caffeine the bitter Gr66a^GRNs). Utilizing this technique, we successfully demonstrated (see new Figure 5B and 5D) that when presented with water, no signal was detected in the Gr66a-LexA, Lk-Gal4 > active-GRASP, or Gr64f-LexA, Lk-Gal4 > active-GRASP transgene flies. However, in the presence of caffeine, Gr66aLexA, Lk-Gal4 > active-GRASP transgene flies exhibited a clear signal in the SEZ, and similarly, sucrose presentation to Gr64f-LexA, Lk-Gal4 > active-GRASP transgene flies yielded a detectable signal. The results obtained from active-GRASP provide additional evidence supporting the connectivity between SELK neurons and both Gr64f^GRNs and Gr66a^GRNs, further indicating the functional connectivity of the GRNs and SELK neurons.

The authors describe a behavioral phenotype when flies are starved, however, they do not use a specific driver for the described cell type, thus they should also tone down their claims.

We agree with the reviewer that the Lk-Gal4 driver line used labels SELK, LHLK, and ABLK neurons. The behavior examined in this paper, the Proboscis Extension Response (PER), measures the initiation of feeding. Although the neural circuit involved in this behavior is primarily confined to the SEZ where SELK neurons are located, we cannot rule out the possibility that other Lk neurons may also play a role in the process. To restrict expression of the Tetanus Toxin, we have utilized the tsh-Gal80 (Clyne et al., 2008) transgene in combination with the Lk-Gal4>UAS-TNT and Lk-Gal4>UAS-TNT^imp constructs to prevent the expression of the Tetanus Toxin in ABLK neurons, thereby restricting its expression to the SELK and LHLK neurons in the central brain. The new results (Sup Figure 7A) indicate that ABLK neurons do not play a role in integrating sweet and bitter information. However, we acknowledge the reviewer's point that we are still silencing LHLK neurons, so we have adjusted our claims to align more closely with our data

Generally, the authors do not provide a big advancement to the field and some of the results are contradictory with previous publications.

We believe our work does not contradict previous findings, nor does it invalidate the role of ABLK neurons in water homeostasis or the role of LHLK neurons in regulating sleep via starvation. We provide additional information on the possible role of SELK neurons in integrating gustatory information. The location of SELK neurons in the SEZ suggests that they may play a role in feeding behavior, and we have demonstrated that these neurons are indeed involved in integrating gustatory information to influence feeding decisions. We consider we have contributed by highlighting a new role for the Leucokinin neuropeptide in feeding behavior.

Reviewer #2 (Public review):

Summary:

A core task of the brain is processing sensory cues from the environment. The neural mechanisms of how sensory information is transmitted from peripheral sense organs to subsequent being processing in defined brain centers remain an important topic in neuroscience. The taste system hereby assesses the palatability of food by evaluating the chemical composition and nutrient content while integrating the current need for energy by assessing the satiation level of the organism. The current manuscript provides insights into the early circuits of gustatory coding using the fruit fly as a model. By combining trans-tango and FACS- based bulk RNAseq to assess the target neurons of sweet sensing (using Gr64fGal4) and bitter sensing (using Gr66a-Gal4) in a first set of experiments the authors investigate genes that are differentially expressed or co-expressed in normal and starved conditions. With a focus on neuropeptides and neurotransmitters, different expressions in the different conditions were assessed resulting in the identification of Leucokinin as a potentially interesting gene. The notion is further supported by RNAseq of Lk- Gal4>mCD8:GFP sorted cells and immunostainings. GRASP and BacTrace experiments further support that the two Lk- expressing cells in the SEZ should indeed be postsynaptic to both types of sensories. Using EM-based connectomics data (based on a previous publication by Engert et al.), the authors also look for downstream targets of the bitter versus sweet gustatory neurons to identify the Lk-neurons. Based on the morphology they identify candidates and further depict the potential downstream neurons in the connectome, which appears largely in agreement with GRASP experiments. Finally silencing the Lk- neurons shows an increased PER response in starved flies (when combined with bitter compounds) as well as increased feeding neurons shows an increased PER response in starved flies (when combined with bitter compounds) as well as increased feeding in a FlyPad assay. Strengths:

Overall this is an intriguing manuscript, which provides insight into the organization of 2nd order gustatory neurons. It specifically provides strong evidence for the Lk-neurons as a target of sweet and bitter GRNs and provides evidence for their role in regulating sweet vs bitter-based behavioral responses. Particularly the integration of different techniques and datasets in an elegant fashion is a strong side of the manuscript. Moreover to put the known LK-neurons into the context of 2nd order gustatory signalling is strengthening the knowledge about this pathway.

Weaknesses:

I do not see any major weakness in the current manuscript. Novelty is to some degree lessened by the fact, that the RNAseq approach did not identify new neurons but rather put the known LK-neurons as major findings. Similarly, the final behavioral section is not very deep and to some degree corroborates the previous publication by the Keene and Nässel labs - that said, the model they propose is indeed novel (but lacks depth in analyses; e.g. there is no physiology that would support the modulation of Lk neurons by either type of GRN). The connectomic section appears a bit out of place and after reading it it's not really clear what one should make of the potential downstream neurons (particularly since the Lk-receptor expression has been previously analyzed); here it might have been interesting to address if/how Lk-neurons may signal directly via a classical neurotransmitter (an information that might be found easily in the adult brain single-cell data).

We thank the reviewer for the comment. Indeed, we attempted in vivo Ca imaging but were unsuccessful. We have rewritten the connectomic section to better integrate it with the rest of the text and have reanalyzed the data obtained. We considered gathering data from the single-cell adult dataset, but this dataset includes the entire adult fly brain, encompassing SELK and LHLK neurons, making it impossible to differentiate between the two types of Lk neurons. Any further analysis will require transcriptomic analysis of SELK via scRNAseq under the different metabolic conditions tested in this study work.

Reviewer #3 (Public review):

Summary:

To make feeding decisions, animals need to process three types of information: positive cues like sweetness, negative cues like bitterness, and internal states such as hunger or satiety. This study aims to identify where the information is integrated into the fruit fly brain. The authors applied RNA sequencing on second-order gustatory neurons responsible for sweet and bitter processing, under fed and starved conditions. The sequencing data reveal significant changes in gene expression across sweet vs. bitter pathways and fed vs. starved states. The authors focus on the neuropeptide Leucokinin (Lk), whose expression is dependent on the starvation state. They identify a pair of neurons, named SELK neurons, which express Lk and receive direct input from both sweet and bitter gustatory neurons. These SELK neurons are ideal candidates to integrate gustatory and internal state information. Behavioral experiments show that blocking these neurons in starved flies alters their tolerance to bitter substances during feeding.

Strengths:

(1) The study employs a well-designed approach, targeting specific neuronal populations, which is more efficient and precise compared to traditional large-scale genetic screening methods.

(2) The RNAseq results provide valuable data that can be utilized in future studies to explore other molecules beyond Lk.

(3) The identification of SELK neurons offers a promising avenue for future research into how these neurons integrate conflicting gustatory signals and internal state information.

Weaknesses:

(1) Unfortunately, due to technical challenges, the authors were unable to directly image the functional activity of SELK neurons.

(2) In the behavioral experiments, tetanus toxin was used to block SELK neurons. Since these neurons may release multiple neurotransmitters or neuropeptides, the results do not specifically demonstrate that Leucokinin (Lk) is the critical factor, as suggested in Figure 8. To address this, I recommend using RNAi to inhibit Lk expression in SELK neurons and comparing the outcomes to wild-type controls via the PER assay.

We appreciate the author's comments and suggestions. As noted, Tetanus Toxin silences the neuron’s activity, affecting the functioning of various neurotransmitters and neuropeptides released by the targeted neuron. In response to the reviewer's recommendation, we employed an RNAi line specifically designed to silence Leucokinin production in Lk-expressing neurons.

The results presented in Supplementary Figure 7B demonstrate that knocking down Leucokinin in Lk neurons significantly reduces the flies' tolerance to caffeine in sweet food.

It is crucial to highlight that the sucrose concentration used in Figure 7C was 50mM, whereas in Supplementary Figure 7B, it was increased to 100mM. This adjustment was necessary because the Lk-Gal4, UAS-RNAi, and Lk-Gal4>UAS-RNAi transgenic lines exhibited reduced sensitivity to sucrose compared to the Lk-Gal4>UAS-TNT or Lk-Gal4>UAS-TNT^imp lines. We aimed to establish a sucrose concentration that would elicit a 50% Proboscis Extension Response (PER) without adding any other compound, thereby allowing us to evaluate the additional effect of caffeine in the food.

However, according to the data derived from the connectome, SELK neurons might be cholinergic, and this neurotransmitter might be involved in controlling also the behavior of the flies.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

To get more evidence for connections between sensory cells and SELK neurons, could the authors also analyze a second available EM data set? Would setting a different threshold (>5 synapses) reveal connections to both sensories? Comparisons between SELK in- and outputs from EM data and Tango labeling also seem to differ quite a lot based on provided images - can the authors count cell bodies in the stainings? Further proof would be to provide functional imaging data that shows that SELK neurons respond to sugar and bitter compounds.

In this study, we utilized the recently published EM dataset for the Drosophila central brain connectome (Dorkenwald et al., 2024; Flywire.ai). Changing the number of synapses affects the counts of pre- and postsynaptic neurons. We set a threshold of more than five synapses, as recommended by Flywire, to avoid false positives (Dorkenwald et al., 2024). This threshold has been widely used in recent papers (Engert et al., 2022; Shiu et al., 2022; Walker et al., 2025).

The neuron counts in the connectomic data differ from those in the trans- and retro-TANGO experiments. In our initial trans-TANGO experiment, which labeled postsynaptic neurons in the Gr64fGal4 and Gr66a-Gal4 transgenic lines, we counted the labeled neurons (see Supplementary Figure 1C) and observed considerable variability between different brains. Due to anticipated variability, we did not count the labeled neurons from trans-TANGO and retro-TANGO techniques in the Leucokinin neurons. Furthermore, neither technique labels all postsynaptic or presynaptic neurons, respectively. A recent study on the retro-TANGO technique (Sorkac et al., 2023) found a minimum threshold: the presynaptic neuron must form a certain number of synapses with the neuron of interest to be adequately labeled. According to this paper, the established threshold is 17 synapses. It is likely that the trans-TANGO technique also has a threshold relating to the number of labeled neurons, contingent on the synapse count. This would explain the discrepancy between the two results.

Unfortunately, we have not been able to provide functional data pointing to the activation of SELK neurons by sucrose or caffeine. However, our active-GRASP data indicates that the connectivity between Gr64f^GRNs and Gr66a^GRNs with SELK neurons is present and functional.

How many Leucokinin-positive cells are in the SEZ? Does the RNA-seq data provide further information about the SELK neurons? Potential receptor candidates for how they integrate hunger signals? AMPKa was described to be required in LHLK neurons.

There are two SELK neurons in the SEZ. Due to the nature of our bulk RNA sequencing (RNAseq), we cannot link any additional gene expressions detected in our transcriptomic analysis specifically to the SELK neurons regarding the integration of various signaling processes. Furthermore, the single-cell RNA sequencing (scRNAseq) data available from the Drosophila brain, as reported by Li et al. (2022), does not allow accurate differentiation between SELK and LHLK neurons. To understand how these neurons integrate both metabolic and sensory information, it is crucial to conduct a focused RNAseq study specifically on the SELK neurons to understand how these neurons integrate both metabolic and sensory information. This targeted analysis would provide the necessary insights to elucidate their functional roles better. However, according to the data derived from the connectome, SELK neurons might be cholinergic, and this neurotransmitter might be involved in controlling also the behavior of the flies.

According to previous studies (Yurgel et al., 2019), the Lk-GAL4 line is also expressed in the VNC, thus the authors could make use of the tsh-GAL80 tool to clean up the line. This study also performed GCaMP imaging in fed and 24h starved animals in SELK and couldn't find a difference, can the authors explain this discrepancy?

We thank the reviewer for this suggestion. We have now added a new piece of data using the tsh-Gal80 transgene in our PER experiments (Supplementary Figure 7A). Blocking the expression of TNT in the ABLK neurons does not affect the main conclusion of the behavioral results. As stated previously, we were unable to obtain in vivo Ca imaging responses in SELK neurons upon exposure to sucrose, caffeine, or mixtures of sucrose and caffeine. We do not believe this is a discrepancy with previous works like Yurgel et al., 2019. It is likely that we faced technical issues regarding expression stability and that the stimulation was possibly too weak to detect changes in GFP levels

Reviewer #2 (Recommendations for the authors):

As mentioned above I do not have any major comments on the manuscript, but there are a few points that I feel should be considered:

(1) The identification of the Lk-candidate neurons in the connectome remains a bit mysterious. In the method sections, this reads as follows "manual and visual criteria were applied to identify the neurons of interest ". a) What precisely was done to get to the candidates?b) Are there alternative candidates that may be Lk-neurons? c) How would another neuron affect the conclusion of the downstream analysis?

We thank the reviewer for this comment. We have now modified and added new information in the connectomic section, reinforcing our conclusions and correcting the results obtained.

Our GRASP, BacTRace, and immunohistochemistry experiments pointed to SELK neurons as postsynaptic to both Gr64f^GRNs (sweet) and Gr66a^GRNs (bitter). To identify which neurons in the connectome could be the SELK neurons, we utilized a previously described set of GRNs already identified in the connectome (Shiu et al., 2022). We extracted all postsynaptic neurons to the sweet and bitter GRNs identified and intersected both datasets, retaining only those candidate hits receiving simultaneous input from sweet and bitter GRNs. This process yielded a total of 333 hits. Through visual inspection, we discarded all hits that were merely neuronal fragments or neurons that clearly were not our candidates. We narrowed the list down to a final set of 17 candidate neurons whose arborization was located in the SEZ. We reduced the candidates to two final entries from this list: ID 720575940623529610 (GNG.276) and ID 720575940630808827 (GNG.685). The GNG.276 neuron had a counterpart in the SEZ identified as GNG.246. Both of these neurons were annotated as DNg70 in the Flywire database. GNG.685 had a counterpart identified as GNG.595, and these two neurons were classified as DNg68. In both cases, the neuronal candidates, DNg70 and DNg68, were classified as descending neurons, a characteristic of previously described SELK neurons (Nässel et al., 2021). In our initial analysis published in bioRxiv and sent for revision, we identified DNg70 as potentially the SELK neurons based solely on the morphology of the neurons via visual inspection. However, we employed a better method to determine which candidate is more likely to be the SELK neurons, concluding that DNg68, rather than DNg70, represents the SELK neurons. Briefly, we performed an immunohistochemistry for GFP in the Lk-Gal4>UAS-CD8:GFP flies. We aligned the resulting image in a Drosophila reference brain (JRC2018 U) using the CMTK Registration plugin in ImageJ. The resulting image was skeletonized using the Single Neurite Tracer plugin in ImageJ and later uploaded to the Flywire Gateway platform to compare the structure of the aligned and skeletonized SELK neurons to our candidates. This comparison clearly indicated that the DNg68 neurons are the best candidates for representing the SELK neurons, rather than DNg70. We have updated the text and Figures 6 and Supplementary Figure 6 to reflect the new results. These new results do not alter the conclusions of the paper.

(2) In the transcriptomic experiments It seems that the raw transcripts are reporters, rather than normalised data. Why?

All transcriptomic data is normalized. In Figure 1 the differential expression was calculated using Deseq2 normalized counts. In Figure 2, Transcripts Per Million (TPM) were calculated using the Salmon package and normalized for the gene length.

(3) The expression of nAChRbeta1 in the transcriptomic data is rather striking. However, this remains currently not addressed: is this expression real?

We have not confirmed the upregulation or downregulation in gene expression for other but for Leucokinin, which is our main interest. We found the presence of nAChRbeta1 interesting, as GRNs are cholinergic (Jaeger et al., 2018), suggesting that it would make sense to find cholinergic receptors in G2Ns. However, it is possible that these receptors are expressed in all G2Ns and serve as a common means of communication.

(4) The description of the behavioural experiments in the results section is rather brief. I had a hard time following it since the genotypes are not repeated nor is it stated what is different in the experimental group vs control (but instead simply what changes in the experimental group, in a rather discussion-like fashion).

We thank the reviewer for the comment, we have rewritten this section to improve its clarity.

(5) If I understand the genetics for the behavioural experiments correctly it addresses the entire Lk-Gal4 expressing population, thus it is not possible to describe the role of the two SEZ neurons, but rather LkGal4 neurons. This should be clarified.

We thank the reviewer for this comment. Indeed, the Lk-Gal4 driver we used drives expression in all Leucokinin neurons, making it impossible to distinguish between the SELK, LHLK, or ABLK neurons. We have added a new piece of behavioral data by using the tsh-Gal80 transgene to prevent the expression of TNT in the ABLK neurons (Supplementary Figure 7A), but still we cannot distinguish between SELK and LHLK. We have rewritten the text to clarify this fact.

Reviewer #3 (Recommendations for the authors):

Overall, the manuscript is well-written, I only have one minor suggestion for improvement. In Figure 8C, please clarify the use of TNT to block Lk release.

We thank the reviewer for the comment, we have clarified the use of TNT in the text.

References Clyne, J. D. & Miesenböck, G. Sex-Specific Control and Tuning of the Pattern Generator for Courtship Song in Drosophila. Cell 133, 354–363 (2008).

Dorkenwald, S. et al. Neuronal wiring diagram of an adult brain. Nature 634, 124–138 (2024).

Engert, S., Sterne, G. R., Bock, D. D. & Scott, K. Drosophila gustatory projections are segregated by taste modality and connectivity. Elife 11, e78110 (2022).

Jaeger, A. H. et al. A complex peripheral code for salt taste in Drosophila. Elife 7, e37167 (2018).

Macpherson, L. J. et al. Dynamic labelling of neural connections in multiple colours by trans-synaptic fluorescence complementation. Nat Commun 6, 10024 (2015).

Nässel, D. R. Leucokinin and Associated Neuropeptides Regulate Multiple Aspects of Physiology and Behavior in Drosophila. Int J Mol Sci 22, 1940 (2021).

Shiu, P. K., Sterne, G. R., Engert, S., Dickson, B. J. & Scott, K. Taste quality and hunger interactions in a feeding sensorimotor circuit. eLife 11, e79887 (2022).

Walker, S. R., Peña-Garcia, M. & Devineni, A. V. Connectomic analysis of taste circuits in Drosophila. Sci. Rep. 15, 5278 (2025).

eLife Assessment

This study provides valuable insights into the organization of second-order circuits of gustatory neurons, particularly in how these circuits integrate opposing taste inputs and are modulated by metabolic state to regulate feeding behavior. Through an elegant combination of complementary techniques, the authors identify the target neurons involved in gustatory integration. The evidence supporting their conclusions is convincing.

Reviewer #1 (Public review):

Summary:

Mollá-Albaladejo et al. investigate the neurons downstream of GR64f and Gr66a, called G2Ns. They identify downstream neurons using trans-Tango labeling with RFP and then perform bulk RNA-seq on the RFP-sorted cells. Gene expression is up- or downregulated between the cell populations and between fed and starved states. They specifically identify Leukocinin as a neuropeptide that is upregulated in starved Gr66a cells. Leucokinin cells, identified by a GAL4 line, indeed show higher expression when starved, especially in the SEZ. Furthermore, Leucokinin cells colocalize with the trans-Tango signal from downstream neurons of both GRs. This connection is confirmed with GRASP and active GRASP. According to EM data, Leucokinin cells in the SEZ receive a lot of input and connect to many downstream neurons. In behavior experiments performed with flies lacking Leucokinin neurons, flies show reduced responsiveness to sugar and bitter mixtures when starved. The authors suggest that Leucokinin neurons integrate bitter and sugar tastes and that their output is modified by a hunger state.

Strengths:

The authors use a multitude of tools to identify SELK neurons downstream of taste sensory neurons and as starvation-sensitive cells. This study provides an example of how combining genetic labeling, RNA-seq, and EM analysis can be used to investigate the function of specific neural circuits.

Weaknesses:

The authors now provide more evidence to show a functional connection between sensory neurons and SELK neurons, for example, by using active GRASP, however, different staining methods reveal different connectivity patterns. The authors describe a behavioral phenotype when flies are starved, however, the phenotype can still not clearly be assigned to the SELK neurons.

Reviewer #2 (Public review):

Summary:

A core task of the brain is processing sensory cues from the environment. The neural mechanisms of how sensory information is transmitted from peripheral sense organs to subsequent being processing in defined brain centers remains an important topic in neuroscience. The taste system hereby assesses the palatability of food by evaluating the chemical composition and nutrient content while integrating the current need of energy by assessing the satiation level of the organism. The current manuscript provides insights into the early circuits gustatory coding using the fruit fly as model. By combining trans-tango and FACS-based bulk RNAseq to assess the target neurons of sweet sensing (using by Gr64f-Gal4) and bitter sensing (using Gr66a-Gal4) in a first set of experiments the authors investigate genes that are differentially expressed or co-expressed in normal and starved conditions. With a focus on neuropeptides and neurotransmitters differential expression in the different conditions were assessed resulting in the identification of Leucokinin as potentially interesting gene. The notion is further supported by RNAseq of Lk-Gal4>mCD8:GFP sorted cells and immunostainings. GRASP and BacTrace experiments further supports that the two Lk expressing cells in the SEZ should indeed be postsynaptic to both type of sensors. Using EM-based connectomics data (based on a previous publication by Engert et al.), the authors also look for downstream targets of the bitter versus sweet gustatory neurons to identify the Lk-neurons. Based on morphology they identify candidates and further depict the potential downstream neurons in the connectome, which appears largely in agreement with GRASP experiments. Finally silencing the Lk-neurons shows an increased PER response in starved flies (when combined with bitter compounds) as well as increased feeding in a FlyPad assay.

Strengths:

Overall this is an intriguing manuscript, which provides insight into the organization of 2nd order gustatory neurons. It specifically provides strong evidence for the Lk-neurons as target of sweet and bitter GRNs and provides evidence for their role in regulating sweet vs bitter based behavioral responses. Particularly the integration of different techniques and datasets in an elegant fashion is a strong side of the manuscript. Moreover to put the known LK-neurons into the context of 2nd order gustatory signalling is strengthening the knowledge about this pathway.

Weaknesses:

I do not see any major weakness in the current manuscript. Novelty is to some degree lessened by the fact, that the RNAseq approach did not identify new neurons but rather put the known LK-neurons as major finding. Similarly the final behavioral section is not very deep and to some degree corroborates the previous publication by the Keene and Nässel labs- that said, the model they propose is indeed novel (but lacks depth in analyses, e.g. there is no physiology that would support the modulation of Lk neurons by either type of GRN). The connectomic section appears a bit out of place and after reading it it's not really clear what one should make of the potential downstream neurons (particularly since the Lk-receptor expression has been previously analyzed); here it might have been interesting to address if/how Lk-neurons may signal directly via a classical neurotransmitter (an information that might be found easily in the adult brain single-cell data).

Comments on the latest version:

I feel all points have been included to a satisfactory degree.

Reviewer #3 (Public review):

Summary:

To make feeding decisions, animals need to process three types of information: positive cues like sweetness, negative cues like bitterness, and internal states such as hunger or satiety. This study aims to identify where the information is integrated in the fruit fly brain. The authors applied RNA sequencing on second-order gustatory neurons responsible for sweet and bitter processing, under fed and starved conditions. The sequencing data reveal significant changes in gene expression across sweet vs. bitter pathways and fed vs. starved states. The authors focus on the neuropeptide Leucokinin (Lk), whose expression is dependent on the starvation state. They identify a pair of neurons, named SELK neurons, which express Lk and receive direct input from both sweet and bitter gustatory neurons. These SELK neurons are ideal candidates to integrate gustatory and internal state information. Behavioral experiments show that blocking these neurons in starved flies alters their tolerance to bitter substances during feeding.

Strengths:

(1) The study employs a well-designed approach, targeting specific neuronal populations, which is more efficient and precise compared to traditional large-scale genetic screening methods.

(2) The RNAseq results provide valuable data that can be utilized in future studies to explore other molecules beyond Lk.

(3) The identification of SELK neurons offers a promising avenue for future research into how these neurons integrate conflicting gustatory signals and internal state information.

Weaknesses:

Unfortunately, due to technical challenges, the authors were unable to directly image the functional activity of SELK neurons.

Author response:

Reviewer #1:

As this code was developed for use with a 4096 electrode array, it is important to be aware of double-counting neurons across the many electrodes. I understand that there are ways within the code to ensure that this does not happen, but care must be taken in two key areas. Firstly, action potentials traveling down axons will exhibit a triphasic waveform that is different from the biphasic waveform that appears near the cell body, but these two signals will still be from the same neuron (for example, see Litke et al., 2004 "What does the eye tell the brain: Development of a System for the Large-Scale Recording of Retinal Output Activity"; figure 14). I did not see anything that would directly address this situation, so it might be something for you to consider in updated versions of the code.

We thank the reviewer for this insightful comment. We agree that signals from the same neuron may be collected by adjacent channels. To address this concern in our software, we plan to add a routine to SpikeMAP that allows users to discard nearby channels where spike count correlations exceed a pre-determined threshold. Because there is no ground truth to map individual cells to specific channels on the hd-MEA, a statistical approach is warranted.

Secondly, spike shapes are known to change when firing rates are high, like in bursting neurons (Harris, K.D., Hirase, H., Leinekugel, X., Henze, D.A. & Buzsáki, G. Temporal interaction between single spikes and complex spike bursts in hippocampal pyramidal cells. Neuron 32, 141-149 (2001)). I did not see this addressed in the present version of the manuscript.

This is a valid concern. To ensure that firing rates are relatively constant over the duration of a recording, we will plot average spike rates using rolling windows of a fixed duration. We expect that population firing rates will remain relatively stable across the duration of recordings.

Another area for possible improvement would be to build on the excellent validation experiments you have already conducted with parvalbumin interneurons. Although it would take more work, similar experiments could be conducted for somatostatin and vasoactive intestinal peptide neurons against a background of excitatory neurons. These may have different spike profiles, but your success in distinguishing them can only be known if you validate against ground truth, like you did for the PV interneurons.

We agree that further cycles of experiments could be performed with SOM, VIP, and other neuronal subtypes, and we hope that researchers will take advantage of SpikeMAP too. We will clarify this possibility in the Discussion section of the manuscript.

Reviewer #2:

Summary:

While I find that the paper is nicely written and easy to follow, I find that the algorithmic part of the paper is not really new and should have been more carefully compared to existing solutions. While the GT recordings to assess the possibilities of a spike sorting tool to distinguish properly between excitatory and inhibitory neurons are interesting, spikeMAP does not seem to bring anything new to state-of-the-art solutions, and/or, at least, it would deserve to be properly benchmarked. I would suggest that the authors perform a more intensive comparison with existing spike sorters.

We thank the reviewer for this comment. As detailed in Table 1, SpikeMAP is the only method that performs E/I sorting on large-scale multielectrodes, hence a comparison to competing methods is not currently possible. That being said, many of the pre-processing steps of SpikeMAP (Figure 1) involve methods that are already well-established in the literature and available under different packages. To highlight the contribution of our work and facilitate the adoption of SpikeMAP, we plan to provide a “modular” portion of SpikeMAP that is specialized in performing E/I sorting and can be added to the pipeline of other packages such as KiloSort more clearly.  This modularized version of the code will be shared freely along with the more complete version already available.

Weaknesses:

(1) The global workflow of spikeMAP, described in Figure 1, seems to be very similar to that of Hilgen et al. 2020 (10.1016/j.celrep.2017.02.038). Therefore, the first question is what is the rationale of reinventing the wheel, and not using tools that are doing something very similar (as mentioned by the authors themselves). I have a hard time, in general, believing that spikeMAP has something particularly special, given its Methods, compared to state-of-the-art spike sorters.

We agree with the reviewers that there are indeed similarities between our work and the Hilgen et al. paper. However, while the latter employs optogenetics to stimulate neurons on a large-scale array, their technique does not specifically target inhibitory (e.g., PV) neurons as described in our work. We will clarify our paper accordingly.

This is why, at the very least, the title of the paper is misleading, because it lets the reader think that the core of the paper will be about a new spike sorting pipeline. If this is the main message the authors want to convey, then I think that numerous validations/benchmarks are missing to assess first how good spikeMAP is, with reference to spike sorting in general, before deciding if this is indeed the right tool to discriminate excitatory vs inhibitory cells. The GT validation, while interesting, is not enough to entirely validate the paper. The details are a bit too scarce for me, or would deserve to be better explained (see other comments after).

The title of our work will be edited to make it clear that while elements of the pipeline are well-established and available from other packages, we are the first to extend this pipeline to E/I sorting on large-scale arrays.

(2) Regarding the putative location of the spikes, it has been shown that the center of mass, while easy to compute, is not the most accurate solution [Scopin et al, 2024, 10.1016/j.jneumeth.2024.110297]. For example, it has an intrinsic bias for finding positions within the boundaries of the electrodes, while some other methods, such as monopolar triangulation or grid-based convolution, might have better performances. Can the authors comment on the choice of the Center of Mass as a unique way to triangulate the sources?

We agree with the reviewer and will point out limits of the center-of-mass algorithm based on the article of Scopin et al (2024). Further, we will augment the existing code library to include monopolar triangulation or grid-based convolution as options available to end-users.

(3) Still in Figure 1, I am not sure I really see the point of Spline Interpolation. I see the point of such a smoothing, but the authors should demonstrate that it has a key impact on the distinction of Excitatory vs. Inhibitory cells. What is special about the value of 90kHz for a signal recorded at 18kHz? What is the gain with spline enhancement compared to without? Does such a value depend on the sampling rate, or is it a global optimum found by the authors?

We will clarify these points. Specifically, the value of 90kHz was chosen because it provided a reasonable temporal characterization of spikes; this value, however, can be adjusted within the software based on user preference.

(4) Figure 2 is not really clear, especially panel B. The choice of the time scale for the B panel might not be the most appropriate, and the legend filtered/unfiltered with a dot is not clear to me in Bii.

We will re-check Fig.2B which seems to have error in rendering, likely due to conversion from its original format.

In panel E, the authors are making two clusters with PCA projections on single waveforms. Does this mean that the PCA is only applied to the main waveforms, i.e. the ones obtained where the amplitudes are peaking the most? This is not really clear from the methods, but if this is the case, then this approach is a bit simplistic and does not really match state-of-the-art solutions. Spike waveforms are quite often, especially with such high-density arrays, covering multiple channels at once, and thus the extracellular patterns triggered by the single units on the MEA are spatio-temporal motifs occurring on several channels. This is why, in modern spike sorters, the information in a local neighbourhood is often kept to be projected, via PCA, on the lower-dimensional space before clustering. Information on a single channel only might not be informative enough to disambiguate sources. Can the authors comment on that, and what is the exact spatial resolution of the 3Brain device? The way the authors are performing the SVD should be clarified in the methods section. Is it on a single channel, and/or on multiple channels in a local neighbourhood?

Here, the reviewer is suggesting that it may be better to perform PCA on several channels at once, since spikes can occur at several channels at the same time. To address this concern, small routine will be written allowing users to choose how many nearby channels to be selected for PCA.

(5) About the isolation of the single units, here again, I think the manuscript lacks some technical details. The authors are saying that they are using a k-means cluster analysis with k=2. This means that the authors are explicitly looking for 2 clusters per electrode? If so, this is a really strong assumption that should not be held in the context of spike sorting, because, since it is a blind source separation technique, one cannot pre-determine in advance how many sources are present in the vicinity of a given electrode. While the illustration in Figure 2E is ok, there is no guarantee that one cannot find more clusters, so why this choice of k=2? Again, this is why most modern spike sorting pipelines do not rely on k-means, to avoid any hard-coded number of clusters. Can the authors comment on that?

It is true that k=2 is a pre-determined choice in our software. In practice, we found that k>2 leads to poorly defined clusters. However, we will ensure that this parameter can be adjusted in the software. Furthermore, if the user chooses not to pre-define this value, we will provide the option to use a Calinski-Harabasz criterion to select k.

(6) I'm surprised by the linear decay of the maximal amplitude as a function of the distance from the soma, as shown in Figure 2H. Is it really what should be expected? Based on the properties of the extracellular media, shouldn't we expect a power law for the decay of the amplitude? This is strange that up to 100um away from the soma, the max amplitude only dropped from 260 to 240 uV. Can the authors comment on that? It would be interesting to plot that for all neurons recorded, in a normed manner V/max(V) as function of distances, to see what the curve looks like.

We share the reviewer’s concern and will add results that include a population of neurons to assess the robustness of this phenomenon.

(7) In Figure 3A, it seems that the total number of cells is rather low for such a large number of electrodes. What are the quality criteria that are used to keep these cells? Did the authors exclude some cells from the analysis, and if yes, what are the quality criteria that are used to keep cells? If no criteria are used (because none are mentioned in the Methods), then how come so few cells are detected, and can the authors convince us that these neurons are indeed "clean" units (RPVs, SNRs, ...)?

We applied stringent criteria to exclude cells, and we will revise the main text to be clear about these criteria, which include a minimum spike rate and the use of LDA to separate out PCA clusters. For the cells that were retained, we will include SNR estimates.

(8) Still in Figure 3A, it looks like there is a bias to find inhibitory cells at the borders, since they do not appear to be uniformly distributed over the MEA. Can the authors comment on that? What would be the explanation for such a behaviour? It would be interesting to see some macroscopic quantities on Excitatory/Inhibitory cells, such as mean firing rates, averaged SNRs... Because again, in Figure 3C, it is not clear to me that the firing rates of inhibitory cells are higher than Excitatory ones, whilst they should be in theory.       

We will include a comparison of firing rates for E and I neurons. It is possible that I cells are located at the border of the MEA due to the site of injections of the viral vector, and not because of an anatomical clustering of I cells per se. We will clarify the text accordingly.

(9) For Figure 3 in general, I would have performed an exhaustive comparison of putative cells found by spikeMAP and other sorters. More precisely, I think that to prove the point that spikeMAP is indeed bringing something new to the field of spike sorting, the authors should have compared the performances of various spike sorters to discriminate Exc vs Inh cells based on their ground truth recordings. For example, either using Kilosort [Pachitariu et al, 2024, 10.1038/s41592-024-02232-7], or some other sorters that might be working with such large high-density data [Yger et al, 2018, 10.7554/eLife.34518].

As mentioned previously, Kilosort and related approaches do not address the problem of E/I identification (see Table 1). However, they do have pre-processing steps in common with SpikeMAP. We will add some specific comparison points – for instance, the use of k-means and PCA (which is more common across packages) and the use of cubic spline interpolation (which is less common). Further, we will provide a stand-alone E/I sorting module that can be added to the pipeline of other packages, so that users can use this functionality without having to migrate their entire analysis.

(10) Figure 4 has a big issue, and I guess the panels A and B should be redrawn. I don't understand what the red rectangle is displaying.

We apologize for this issue. It seems there was a rendering problem when converting the figure from its original format. We will address this issue in the revised version of the manuscript.

(11) I understand that Figure 4 is only one example, but I have a hard time understanding from the manuscript how many slices/mice were used to obtain the GT data? I guess the manuscript could be enhanced by turning the data into an open-access dataset, but then some clarification is needed. How many flashes/animals/slices are we talking about? Maybe this should be illustrated in Figure 4, if this figure is devoted to the introduction of the GT data.

We will mention how many flashes/animals/slices were employed in the GT data and provide open access to these data.

(12) While there is no doubt that GT data as the ones recorded here by the authors are the most interesting data from a validation point of view, the pretty low yield of such experiments should not discourage the use of artificially generated recordings such as the ones made in [Buccino et al, 2020, 10.1007/s12021-020-09467-7] or even recently in [Laquitaine et al, 2024, 10.1101/2024.12.04.626805v1]. In these papers, the authors have putative waveforms/firing rate patterns for excitatory and inhibitory cells, and thus, the authors could test how good they are in discriminating the two subtypes.

We thank the reviewer for the suggestion that SpikeMAP could be tested on artificially generated spike trains and will add the citation of the two papers mentioned. We hope future efforts will employ SpikeMAP on both synthetic and experimental data to explore the neural dynamics of E and I neurons in healthy and pathological circuits of the brain.

eLife Assessment

In this manuscript, the authors describe a software package for automatic differentiation of action potentials generated by excitatory and inhibitory neurons, acquired using high-density microelectrode arrays. The work is valuable as it offers a tool with the potential to automatically identify these neuron types in vitro. However, it is incomplete due to limited comparison with ground truth data from optogenetically identified interneuron subtypes and with existing spike sorting pipelines available to users.

Reviewer #1 (Public review):

Summary:

The authors note that while many software packages exist for spike sorting, these do not automatically differentiate with known accuracy between excitatory and inhibitory neurons. Moreover, most existing spike sorting packages are for in vivo use, where the majority of electrodes are separated from each other by several hundred microns or more. There is a need for spike sorting packages that can take advantage of high-density electrode arrays where all electrodes are within a few tens of microns of other electrodes. Here, the authors offer such a software package with SpikeMAP, and they validate its performance in identifying parvalbumin interneurons that were optogenetically stimulated.

Strengths:

The main strength of this work is that the authors use ground truth measures to show that SpikeMAP can take features of spike shapes to correctly identify known parvalbumin interneurons against a background of other neuron types. They use spike width and peak to peak distance as the key features for distinguishing between neuron types, a method that has been around for many years (Barthó, Peter, et al. "Characterization of neocortical principal cells and interneurons by network interactions and extracellular features." Journal of neurophysiology 92.1 (2004): 600-608.), but whose performance has not been validated in the context of high density electrode arrays.

Another strength of this approach is that it is automated - a necessity if your electrode array has 4096 electrodes. Hand-sorting or even checking such a large number of channels is something even the cruelest advisor would not wish upon a graduate student. With such large channel counts, it is essential to have automated methods that are known to work accurately. Hence, the combination of validation and automation is an important advance.

A nice feature of this work is that with high-density electrode arrays, the spike waveforms appear on multiple nearby electrodes simultaneously. And since spike amplitudes fall off with distance, this allows triangulation of neuron locations within the regular electrode array. Thus, spike correlations between neuron types, or within neuron types, can be plotted as a function of distance. While SpikeMAP is not the first to do this (Peyrache, Adrien, et al. "Spatiotemporal dynamics of neocortical excitation and inhibition during human sleep." Proceedings of the National Academy of Sciences 109.5 (2012): 1731-1736.), it is a welcome capability of this package.

It is also good that the code for this package is open-source, allowing a community of people (I expect in vitro labs will especially want to use this) to use the code and further improve it.

Weaknesses:

As this code was developed for use with a 4096 electrode array, it is important to be aware of double-counting neurons across the many electrodes. I understand that there are ways within the code to ensure that this does not happen, but care must be taken in two key areas. Firstly, action potentials traveling down axons will exhibit a triphasic waveform that is different from the biphasic waveform that appears near the cell body, but these two signals will still be from the same neuron (for example, see Litke et al., 2004 "What does the eye tell the brain: Development of a System for the Large-Scale Recording of Retinal Output Activity"; figure 14). I did not see anything that would directly address this situation, so it might be something for you to consider in updated versions of the code. Secondly, spike shapes are known to change when firing rates are high, like in bursting neurons (Harris, K.D., Hirase, H., Leinekugel, X., Henze, D.A. & Buzsáki, G. Temporal interaction between single spikes and complex spike bursts in hippocampal pyramidal cells. Neuron 32, 141-149 (2001)). I did not see this addressed in the present version of the manuscript.

Another area for possible improvement would be to build on the excellent validation experiments you have already conducted with parvalbumin interneurons. Although it would take more work, similar experiments could be conducted for somatostatin and vasoactive intestinal peptide neurons against a background of excitatory neurons. These may have different spike profiles, but your success in distinguishing them can only be known if you validate against ground truth, like you did for the PV interneurons.

Appraisal:

This work addresses the need for an automated spike sorting software package for high-density electrode arrays. Although no spike sorting software is flawless, the package presented here, SpikeMAP, has been validated on PV interneurons, inspiring a degree of confidence. This is a good start, and further validation on other neuron types could increase that confidence. Groups doing in vitro experiments, where 4096 electrode arrays are more common, could find this system particularly helpful.

Reviewer #2 (Public review):

Summary:

In this paper, entitled "SpikeMAP: An unsupervised spike sorting pipeline for cortical excitatory and inhibitory 2 neurons in high-density multielectrode arrays with ground-truth validation", the authors present spikeMAP, a pipeline for the analysis of large-scale recordings of in vitro cortical activity. According to the authors, spikeMAP not only allows for the detection of spikes produced by single neurons (spike sorting), but also allows for the reliable distinction between genetically determined cell types by utilizing viral and optogenetic strategies as ground-truth validation. While I find that the paper is nicely written and easy to follow, I find that the algorithmic part of the paper is not really new and should have been more carefully compared to existing solutions. While the GT recordings to assess the possibilities of a spike sorting tool to distinguish properly between excitatory and inhibitory neurons are interesting, spikeMAP does not seem to bring anything new to state-of-the-art solutions, and/or, at least, it would deserve to be properly benchmarked. I would suggest that the authors perform a more intensive comparison with existing spike sorters.

Strengths:

The GT recordings with optogenetic activation of the cells, based on the opsins, is interesting and might provide useful data to quantify how good spike sorting pipelines are, in vitro, to discriminate between excitatory and inhibitory neurons. Such an approach can be quite complementary to artificially generated ground truth.

Weaknesses:

(1) The global workflow of spikeMAP, described in Figure 1, seems to be very similar to that of Hilgen et al. 2020 (10.1016/j.celrep.2017.02.038). Therefore, the first question is what is the rationale of reinventing the wheel, and not using tools that are doing something very similar (as mentioned by the authors themselves). I have a hard time, in general, believing that spikeMAP has something particularly special, given its Methods, compared to state-of-the-art spike sorters. This is why, at the very least, the title of the paper is misleading, because it lets the reader think that the core of the paper will be about a new spike sorting pipeline. If this is the main message the authors want to convey, then I think that numerous validations/benchmarks are missing to assess first how good spikeMAP is, with reference to spike sorting in general, before deciding if this is indeed the right tool to discriminate excitatory vs inhibitory cells. The GT validation, while interesting, is not enough to entirely validate the paper. The details are a bit too scarce for me, or would deserve to be better explained (see other comments after).

(2) Regarding the putative location of the spikes, it has been shown that the center of mass, while easy to compute, is not the most accurate solution [Scopin et al, 2024, 10.1016/j.jneumeth.2024.110297]. For example, it has an intrinsic bias for finding positions within the boundaries of the electrodes, while some other methods, such as monopolar triangulation or grid-based convolution,n might have better performances. Can the authors comment on the choice of the Center of Mass as a unique way to triangulate the sources?

(3) Still in Figure 1, I am not sure I really see the point of Spline Interpolation. I see the point of such a smoothing, but the authors should demonstrate that it has a key impact on the distinction of Excitatory vs. Inhibitory cells. What is special about the value of 90kHz for a signal recorded at 18kHz? What is the gain with spline enhancement compared to without? Does such a value depend on the sampling rate, or is it a global optimum found by the authors?

(4) Figure 2 is not really clear, especially panel B. The choice of the time scale for the B panel might not be the most appropriate, and the legend filtered/unfiltered with a dot is not clear to me in Bii. In panel E, the authors are making two clusters with PCA projections on single waveforms. Does this mean that the PCA is only applied to the main waveforms, i.e. the ones obtained where the amplitudes are peaking the most? This is not really clear from the methods, but if this is the case, then this approach is a bit simplistic and does not really match state-of-the-art solutions. Spike waveforms are quite often, especially with such high-density arrays, covering multiple channels at once, and thus the extracellular patterns triggered by the single units on the MEA are spatio-temporal motifs occurring on several channels. This is why, in modern spike sorters, the information in a local neighbourhood is often kept to be projected, via PCA, on the lower-dimensional space before clustering. Information on a single channel only might not be informative enough to disambiguate sources. Can the authors comment on that, and what is the exact spatial resolution of the 3Brain device? The way the authors are performing the SVD should be clarified in the methods section. Is it on a single channel, and/or on multiple channels in a local neighbourhood?

(5) About the isolation of the single units, here again, I think the manuscript lacks some technical details. The authors are saying that they are using a k-means cluster analysis with k=2. This means that the authors are explicitly looking for 2 clusters per electrode? If so, this is a really strong assumption that should not be held in the context of spike sorting, because, since it is a blind source separation technique, one can not pre-determine in advance how many sources are present in the vicinity of a given electrode. While the illustration in Figure 2E is ok, there is no guarantee that one can not find more clusters, so why this choice of k=2? Again, this is why most modern spike sorting pipelines do not rely on k-means, to avoid any hard-coded number of clusters. Can the authors comment on that?

(6) I'm surprised by the linear decay of the maximal amplitude as a function of the distance from the soma, as shown in Figure 2H. Is it really what should be expected? Based on the properties of the extracellular media, shouldn't we expect a power law for the decay of the amplitude? This is strange that up to 100um away from the soma, the max amplitude only dropped from 260 to 240 uV. Can the authors comment on that? It would be interesting to plot that for all neurons recorded, in a normed manner V/max(V) as function of distances, to see what the curve looks like.

(7) In Figure 3A, it seems that the total number of cells is rather low for such a large number of electrodes. What are the quality criteria that are used to keep these cells? Did the authors exclude some cells from the analysis, and if yes, what are the quality criteria that are used to keep cells? If no criteria are used (because none are mentioned in the Methods), then how come so few cells are detected, and can the authors convince us that these neurons are indeed "clean" units (RPVs, SNRs, ...)?

(8) Still in Figure 3A, it looks like there is a bias to find inhibitory cells at the borders, since they do not appear to be uniformly distributed over the MEA. Can the authors comment on that? What would be the explanation for such a behaviour? It would be interesting to see some macroscopic quantities on Excitatory/Inhibitory cells, such as mean firing rates, averaged SNRs... Because again, in Figure 3C, it is not clear to me that the firing rates of inhibitory cells are higher than Excitatory ones, whilst they should be in theory.

(9) For Figure 3 in general, I would have performed an exhaustive comparison of putative cells found by spikeMAP and other sorters. More precisely, I think that to prove the point that spikeMAP is indeed bringing something new to the field of spike sorting, the authors should have compared the performances of various spike sorters to discriminate Exc vs Inh cells based on their ground truth recordings. For example, either using Kilosort [Pachitariu et al, 2024, 10.1038/s41592-024-02232-7], or some other sorters that might be working with such large high-density data [Yger et al, 2018, 10.7554/eLife.34518].

(10) Figure 4 has a big issue, and I guess the panels A and B should be redrawn. I don't understand what the red rectangle is displaying.

(11) I understand that Figure 4 is only one example, but I have a hard time understanding from the manuscript how many slices/mices were used to obtain the GT data? I guess the manuscript could be enhanced by turning the data into an open-access dataset, but then some clarification is needed. How many flashes/animals/slices are we talking about? Maybe this should be illustrated in Figure 4, if this figure is devoted to the introduction of the GT data.

(12) While there is no doubt that GT data as the ones recorded here by the authors are the most interesting data from a validation point of view, the pretty low yield of such experiments should not discourage the use of artificially generated recordings such as the ones made in [Buccino et al, 2020, 10.1007/s12021-020-09467-7] or even recently in [Laquitaine et al, 2024, 10.1101/2024.12.04.626805v1]. In these papers, the authors have putative waveforms/firing rate patterns for excitatory and inhibitory cells, and thus, the authors could test how good they are in discriminating the two subtypes.

eLife Assessment

This study provides important information on the ultrastructural organization of layer 1 of the human neocortex. The quantitative assessment of various synaptic parameters, astrocytic coverage and mitochondrial morphology is based on convincing experimental approaches. These data provide new information on the detailed morphology of human neocortical tissue that will be of interest to neuroscientists working on different network functions.

Reviewer #1:

Summary:

The Authors investigated the anatomical features of the excitatory synaptic boutons in layer 1 of the human temporal neocortex. They examined the size of the synapse, the macular or the perforated appearance and the size of the synaptic active zone, the number and volume of the mitochondria, the number of the synaptic and the dense core vesicles, also differentiating between the readily releasable, the recycling and the resting pool of synaptic vesicles. The coverage of the synapse by astrocytic processes was also assessed, and all the above parameters were compared to other layers of the human temporal neocortex. The Authors conclude that the subcellular morphology of the layer 1 synapses is suitable for the functions of the neocortical layer, i.e. the synaptic integration within the cortical column. The low glial coverage of the synapses might allow the glutamate spillover from the synapses enhancing synaptic crosstalk within this cortical layer.

Strengths:

The strengths of this paper are the abundant and very precious data about the fine structure of the human neocortical layer 1. Quantitative electron microscopy data (especially that derived from the human brain) are very valuable, since this is a highly time- and energy consuming work. The techniques used to obtain the data, as well as the analyses and the statistics performed by the Authors are all solid, strengthen this manuscript, and support the conclusions drawn in the discussion.

Reviewer #2:

The study of Rollenhagen et al examines the ultrastructural features of Layer 1 of human temporal cortex. The tissue was derived from drug-resistant epileptic patients undergoing surgery, and was selected as further from the epilepsy focus, and as such considered to be non-epileptic. The analyses has included 4 patients with different age, sex, medication and onset of epilepsy. The manuscript is a follow-on study with 3 previous publications from the same authors on different layers of the temporal cortex:

Layer 4 - Yakoubi et al 2019 eLife

Layer 5 - Yakoubi et al 2019 Cerebral Cortex,

Layer 6 - Schmuhl-Giesen et al 2022 Cerebral Cortex

They find, the L1 synaptic boutons mainly have single active zone a very large pool of synaptic vesicles and are mostly devoid of astrocytic coverage.

Strengths:

The MS is well written easy to read. Result section gives a detailed set of figures showing many morphological parameters of synaptic boutons and surrounding glial elements. The authors provide comparative data of all the layers examined by them so far in the Discussion. Given that anatomical data in human brain are still very limited, the current MS has substantial relevance. The work appears to be generally well done, the EM and EM tomography images are of very good quality. The analyses is clear and precise.

Weaknesses:

The authors made all the corrections required and answered all of my concerns, included additional data sets, and clarified statements where needed.

Author response:

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

Public Reviews:

Reviewer #1 (Public review):

Summary:

The Authors investigated the anatomical features of the excitatory synaptic boutons in layer 1 of the human temporal neocortex. They examined the size of the synapse, the macular or the perforated appearance and the size of the synaptic active zone, the number and volume of the mitochondria, the number of the synaptic and the dense core vesicles, also differentiating between the readily releasable, the recycling and the resting pool of synaptic vesicles. The coverage of the synapse by astrocytic processes was also assessed, and all the above parameters were compared to other layers of the human temporal neocortex. The Authors conclude that the subcellular morphology of the layer 1 synapses is suitable for the functions of the neocortical layer, i.e. the synaptic integration within the cortical column. The low glial coverage of the synapses might allow the glutamate spillover from the synapses enhancing synaptic crosstalk within this cortical layer.

Strengths:

The strengths of this paper are the abundant and very precious data about the fine structure of the human neocortical layer 1. Quantitative electron microscopy data (especially that derived from the human brain) are very valuable, since this is a highly time- and energy consuming work. The techniques used to obtain the data, as well as the analyses and the statistics performed by the Authors are all solid, strengthen this manuscript, and support the conclusions drawn in the discussion.

Comments on latest version:

The third version of this paper has been substantially improved. The English is significantly better, there are only few paragraphs and sentences which are hard to understand (see my comments and suggestions below). Almost all of my suggestions were incorporated.

We would like to thank the reviewer for the comments and incorporated the suggestions within the latest version of the manuscript.

Remaining minor concerns:

About epileptic and non-epileptic (non-affected) tissue. I am aware that temporal lobe neocortical tissue derived from epileptic patients is regarded as non-affected by many groups, and they are quite similar to the cortex of non-epileptic (tumour) patients in their electrophysiological properties and synaptic physiology. But please, note, that one paper you cited did not use samples from epileptic patients, but only tissue from non-epileptic tumor patients (Molnár et al. PLOS 2008).

When you look deeper, and make thorough comparison of tissues derived from epileptic and non-epileptic patients, there are differences in the fine structure, as well as in several electrophysiological features. See for example Tóth et al., J Physiol, 2018, where higher density of excitatory synapses were found in L2 of neocortical samples derived from epileptic patients compared to non-epileptic (tumor) patients. Furthermore, the appearance of population bursts is similar, but their occurrence is more frequent and their amplitude is higher in tissue from epileptic compared to non-epileptic patients. So, I still cannot agree, that temporal neocortex of epileptic patients with the seizure focus in the hippocampus would be non-affected. Therefore I suggested to use the term biopsy tissue.

We are thankful for this comment on using non-epileptic tissue also by others. We are also aware that Molnár et al. 2008 worked with tumor tissue.

It is still not emphasized in the first paragraph of the Discussion, that only excitatory axon terminals were investigated.

We now mentioned in the first paragraph of the discussion that only excitatory synaptic boutons were investigated.

The text in the Results and the Discussion are somewhat inconsistent.

The last two paragraphs of the Results section ends with several sentences which should be part of the discussion, such as line 328: This finding strongly supports multivesicular release... or line 344: --- pointing towards a layer-specific regulation of the putative RRP. Moreover, the results suggest that... and line 370: ... it is most likely... Please, correct this.

We disagree with the reviewer on these points because these sentences summarizes the findings.

The first paragraph of the Discussion summarizes the work of the quantitative EM work and gives one conclusion about the astrocytic coverage. This last sentence is inconsistent with the other parts of the paragraph. I would either write that "astrocytic coverage was also investigated" (or something similar), or move this sentence to the paragraph which discusses the astrocytic coverage.

Results line 180-183. "Special connections" between astrocytic processes and synaptic boutons are mentioned, but not shown. Either show these (but then prove with staining!), or leave out this paragraph.

We deleted this paragraph as suggested.

Reviewer #2 (Public review):

Summary:

The study of Rollenhagen et al examines the ultrastructural features of Layer 1 of human temporal cortex. The tissue was derived from drug-resistant epileptic patients undergoing surgery, and was selected as further from the epilepsy focus, and as such considered to be non-epileptic. The analyses has included 4 patients with different age, sex, medication and onset of epilepsy. The manuscript is a follow-on study with 3 previous publications from the same authors on different layers of the temporal cortex:

Layer 4 - Yakoubi et al 2019 eLife

Layer 5 - Yakoubi et al 2019 Cerebral Cortex,

Layer 6 - Schmuhl-Giesen et al 2022 Cerebral Cortex

They find, the L1 synaptic boutons mainly have single active zone a very large pool of synaptic vesicles and are mostly devoid of astrocytic coverage.

Strengths:

The MS is well written easy to read. Result section gives a detailed set of figures showing many morphological parameters of synaptic boutons and surrounding glial elements. The authors provide comparative data of all the layers examined by them so far in the Discussion. Given that anatomical data in human brain are still very limited, the current MS has substantial relevance. The work appears to be generally well done, the EM and EM tomography images are of very good quality. The analyses is clear and precise.

Weaknesses:

The authors made all the corrections required and answered all of my concerns, included additional data sets, and clarified statements where needed.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

Minor suggestions:

Synaptic density, lines 189-193. If you say "comparatively" high, then compare to something (cite your own work for the other layers, and tell the approximative values for the other layers). Same in line 194 comparably high to what? Other option: say "relatively high".

We corrected the sentences as suggested by the reviewer.

Line 206: When present, mitochondria (comma missing)

Corrected as suggested by the reviewer.

Line 265: Dot is missing at the end of the sentence (after Shapira et al. 2003)

Corrected as suggested by the reviewer.

Lines 300-301: Check the English for this sentence: significant difference BETWEEN TWO sublaminae and not significant difference for both sublaminae.

Corrected as suggested by the reviewer.

Lines 304-305: Check the sentence, please, it is not understandable without the text in parenthesis.

Corrected as suggested by the reviewer.

Line 354 Dot missing at the end of the sentence (after Figure 6A, B)

Corrected as suggested by the reviewer.

Line 354-358: Please rephrase this sentence (too complicated, not understandable). I do not understand why results of the L4, L5, L6 are described here. What does it mean "Astrocytes and their fine processes formed a relatively dense, but a comparably loose network within the neuropil in L1"? Dense or loose?

In the experiment measuring the volume fraction of astrocytic processes (Figure 6C), all six cortical layers were analyzed, thus we compared the values obtained for L1 with the results for L4, L5 and L6. For more clarity, we rephrased the sentence: “Astrocytes and their fine processes formed a relatively dense network in L4 and L5, but a comparably loose one within the neuropil in L1…” We also rephrased other sentences in this paragraph (as also suggested below).

Lines 359-369: Please rephrase this paragraph. The sentences are too complicated, have too many parentheses, and are not understandable. I suggest to write first how many synapses were examined in L1 and L4, then how many of them were on spine and on dendrites (either n or %). Then give the values how many (n or %) of them were "tripartite synapses", out of spine synapses and of dendritic synapses in both layers. How many of them were partially covered in both layers. Please, write the data in a systematic way. The best would be to give the values in a table as well. This way it will be more understandable (now, it is chaotic, hard to follow).

We rephrased the paragraph and added a new table (3).

Line 383: Dot missing from the end of the sentence.

Corrected as suggested by the reviewer.

Line 436: Reconsider "comparably low compared to". The comparably means what in this case? The whole paragraph is hard to understand, please, check and review for improvements to the use of English or use chatGPT to check it.

We corrected the sentence according to the reviewer’s suggestion.

Line 487: Same thing again: "The comparably largest size of the RP in L1 when compared..." What would you like to say with "comparably"? Check the meaning of this word in a dictionary, please. I have the feeling that you are using this word instead of "relatively".

Corrected as suggested by the reviewer.

Line 488 "and TO that found fot L4 and L5 in rodents..."

Corrected as suggested by the reviewer.

Line 493-495: Same again, comparably when compared, correct, please.

Corrected as suggested by the reviewer.

Supplemental figures: Now I do understand why Hu-01 and Hu-02 are twice, and I think, 3 patients were examined for L1a and three for L1b. But which side is which on the subfigures? Left side (Hu-01, 02 03) was used for L1a, or L1b? Could you write this in the legend, or mark on the figure (at least at one subfigure), please?

We implemented a comment for clarity.

eLife Assessment

This theoretical study makes a useful contribution to our understanding of a subtype of type 2 diabetes - ketosis-prone diabetes mellitus (KPD) - with a potential impact on our broader understanding of diabetes and glucose regulation. The article presents an ordinary differential equation-based model for KPD that incorporates a number of distinct timescales - fast, slow, as well as intermediate, incorporating a key hypothesis of reversible beta cell deactivation. The presented evidence is solid and shows that observed clinical disease trajectories may be explained by a simple mathematical model in a particular parameter regime.

Reviewer #1 (Public review):

The goal of this work is to understand the clinical observation of a subgroup of diabetics who experience extremely high levels of blood glucose levels after a period of high carbohydrate intake. These symptoms are similar to the onset of Type 1 diabetes but, crucially, have been observed to be fully reversible in some cases.

The authors interpret these observations by analyzing a simple yet insightful mathematical model in which β-cells temporarily stop producing insulin when exposed to high levels of glucose. For a specific model realization of such dynamics (and for specific parameter values) they show that such dynamics lead to two distinct stable states. One is the relatively normal/healthy state in which β-cells respond appropriately to glucose by releasing insulin. In contrast, when enough β-cells "refuse" to produce insulin in a high-glucose environment, there is not enough insulin to reduce glucose levels, and the high-glucose state remains locked in because the high-glucose levels keep β-cells in their inactive state. The presented mathematical analysis shows that in their model the high-glucose state can be entered through an episode of high glucose levels and that subsequently the low-glucose state can be re-entered through prolonged insulin intake.

The strength of this work is twofold. First, the intellectual sharpness of translating clinical observations of ketosis-prone type 2 diabetes (KPD) into the need for β-cell responses on intermediate timescales. Second, the analysis of a specific model clearly establishes that the clinical observations can be reproduced with a model in which β-cells dynamics reversibly enter a non-insulin-producing state in a glucose-dependent fashion.

The likely impact of this work is a shift in attention in the field from a focus on the short and long-term dynamics in glucose regulation and diabetes progression to the intermediate timescales of β-cell dynamics. I expect this to lead to much interest in probing the assumptions behind the model to establish what exactly the process is by which patients enter a 'KPD state'. Furthermore, I expect this work to trigger much research on how KPD relates to "regular" type 2 diabetes and to lead to experimental efforts to find/characterize previously overlooked β-cell phenotypes.

In summary, the authors claim that observed clinical dynamics and possible remission of KPD can be explained through introducing a temporarily inactive β-cell state into a "standard model" of diabetes. The evidence for this claim comes from analyzing a mathematical model and clearly presented.

Reviewer #2 (Public review):

In this manuscript, Ridout et al. present an intriguing extension of beta cell mass-focused models for diabetes. Their model incorporates reversible glucose-dependent inactivation of beta cell mass, which can trigger sudden-onset hyperglycemia due to bistability in beta cell mass dynamics. Notably, this hyperglycemia can be reversed with insulin treatment. The model is simple, elegant, and thought-provoking.

Author response:

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

Public reviews:

Concerning the grounding in experimental phenomenology, it would be beneficial to identify specific experiments to strengthen the model. In particular, what evidence supports reversible beta cell inactivation? This could potentially be tested in mice, for instance, by using an inducible beta cell reporter, treating the animals with high glucose levels, and then measuring the phenotype of the marked cells. Such experiments, if they exist, would make the motivation for the model more compelling.

There is some direct evidence of reversible beta cell inactivation in rodent / in vitro models. We had already mentioned this in the discussion, but we have added some text emphasizing / clarifying the role of this evidence (lines 359–362).

Others have also argued that some analyses of insulin treatment in conventional T2D, which has a stronger effect in patients with higher glucose before treatment, provides indirect evidence of reversal of glucotoxicity. We have also mentioned this in the revised paper (lines 284–285).

For quantitative experiments, the authors should be more specific about the features of beta cell dysfunction in KPD. Does the dysfunction manifest in fasting glucose, glycemic responses, or both? Is there a ”pre-KPD” condition? What is known about the disease’s timescale?

The answers to some of these questions are not entirely clear—patients present with very high glucose, and thus must be treated immediately. Due to a lack of antecedent data it is not entirely clear what the pre-KPD condition is, but there is some evidence that KPD is at least not preceded by diabetes symptoms. This point is already noted in the introduction of the paper and Table 1. However, we have added a small note clarifying that this does not rule out mild hyperglycemia, as in prediabetes (and indeed, as our model might predict) (lines 76–77). Similarly, due to the necessity of immediate insulin treatment, it is not clear from existing data whether the disorder manifests more strongly in fasting glucose or glucose response, although it is likely in both. (We might infer this since continuous insulin treatment does not produce fasting hypoglycemia, and the complete lack of insulin response to glucose shortly after presentation should produce a strong effect in glycemic response.) We believe our existing description of KPD lists all of the relevant timescales, however we have also slightly clarified this description in response to the first referee’s comments (lines 66–73, 83)

The authors should also consider whether their model could apply to other conditions besides KPD. For example, the phenomenology seems similar to the ”honeymoon” phase of T1D. Making a strong case for the model in this scenario would be fascinating.

This is an excellent idea, which had not occurred to us. We have briefly discussed this possibility in the remission (lines 281–291), but plan to analyze it in more detail in a future manuscript.

Reviewer #1 (Recommendations for the author):

Whenever simulation results are presented, parameter values should be specified right there in the figure captions.

We have added the values of glucotoxicity parameters to the caption of Figure 2. In other figures, we have explicitly mentioned which panel of Figure 2 the parameters are taken from. Description of the non-glucotoxicity parameters is a bit cumbersome (there are a lot of them, but our model of fast dynamics is slightly different from Topp et al. so it does not suffice to simply say we took their parameters) so we have referred the reader to the Materials and Methods for those.

I was confused by the language in Figure 4. Could the authors clarify whether they argue that: (1) the observed KPD behaviour is the result of the system switching from one stable state to another when perturbed with high glucose intake? (2) the observed KPD behaviour is the result of one of the steady states disappearing with high glucose intake?

What we mean to say is that during a period of high sugar intake or exogeneous insulin treatment, one of the fixed points is temporarily removed—it is still a fixed point of the “normal” dynamics, but not a fixed point of the dynamics with the external condition added. Since when glucose (insulin) intake is high enough, only the low (high)-β fixed point is present, under one of these conditions the dynamics flow toward that fixed point. When the external influx of glucose/insulin is turned off, both fixed points are present again—but if the dynamics have moved sufficiently far during the external forcing, the fixed point they end up in will have switched from one fixed point to the other. We have edited the text to make this clearer (lines 153–185). Do note, however, that in response to both referee’s comments (see below), Figures 3 and 4 have been replaced with more illuminating ones. This specific point is now addressed by the new Figure 3.

The adaptation of the prefactor ’c’ was confusing to me. I think I understood it in the end, but it sounded like, ”here’s a complication, but we don’t explain it because it doesn’t really matter”. I think the authors can explain this better (or potentially leave out the complication with ’c’ altogether?).

Indeed, the existence of an adaptation mechanism is important for our overall picture of diabetes pathogenesis, but not for many of our analyses, which assume prediabetes. Nonetheless, we agree that the current explanation of it’s role is confusing because of its vagueness. We have elaborated the explanation of the type of dynamics we assume for c, adding an equation for its dynamics to the “Model” section of the Materials and methods, explained in lines 456–465. We have also amended Figure 1 to note this compensation.

I expect the main impact of this work will be to get clinical practitioners and biomedical researchers interested in the intermediate timescale dynamics of β-cells and take seriously the possibility that reversible inactive states might exist. But this impact will only be achieved when the results are clearly and easily understandable by an audience that is not familiar with mathematical modelling. I personally found it difficult to understand what I was supposed to see in the figures at first glance. Yes, the subtle points are indeed explained in the figure captions, but it might be advantageous to make the points visually so clear that a caption is barely needed. For example, when claiming that a change in parameters leads to bistability, why not plot the steady state values as a function of that parameter instead of showing curves from which one has to infer a steady state?

I would advise the authors to reconsider their visual presentation by, e.g., presenting the figures to clinical practitioners or biomedical researchers with just a caption title to test whether such an audience can decipher the point of the figure! This is of course merely a personal suggestion that the authors may decide to ignore. I am making this suggestion only because I believe in the quality of this work and that improving the clarity of the figures and the ease with which one can understand the main points would potentially lead to a much larger impact on the presented results.

This is a very good point. We have made several changes. Firstly, we have added smaller panels showing the dynamics of β to Figure 2; previously, the reader had to infer what was happening to β from G(t). Secondly, we have completely replaced the two figures showing dβ/dt, and requiring the reader to infer the fixed points of β, with bifurcation diagrams that simply show the fixed points of G and β. The new figures show through bifurcation diagrams how there are multiple fixed points in KPD, how glucose or insulin treatment force the switching of fixed points, and how the presence of bistability depends on the rate of glucotoxicity. (These new figures are Fig. 3–5 in the revised manuscript.)

Could the authors explicitly point out what could be learned from their work for the clinic? At the moment treatment consists of giving insulin to patients. If I understand correctly, nothing about the current treatment would change if the model is correct. Is there maybe something more subtle that could be relevant to devising an optimal treatment for KPD patients?

This is another very good point. We have added a new figure (Fig. 7) in our results section showing how this model, or one like it, can be analyzed to suggest an insulin treatment schedule (once parameters for an individual patient can be measured), and added some discussion of this point (lines 224–240) as well as lifestyle changes our model might suggest for KPD patients to the discussion (lines 413–425).

Similarly, could the authors explicitly point out how their model could be experimentally tested? For example, are the functions f(G) and g(G) experimentally accessible? Related to that, presumably the shape of those functions matters to reproduce the observed behaviour. Could the authors comment on that / analyze how reproducing the observed behaviour puts constraints on the shape of the used functions and chosen parameter values?

g(G) has not been carefully measured in cellular data, however it could be in more quantative versions of existing experiments. Further, our model indeed requires some general features for the forms of f(G) and g(G) to produce KPD-like phenomena. We have added some comment on this to the discussion section of the revised manuscript (lines 367–372).

Could the authors explicitly spell out which parameters they think differ between individual KPD patients, and which parameters differ between KPD patients and ’regular’ type 2 diabetics?

In general we expect all parameters should vary both among KPD patients and between KPD / “conventional” T2D. The primary parameter determining whether KPD and conventional T2D, is seen, however, is the ratio kIN/kRE. We have elaborated on both these points in the revised mansuscript. (Lines 186–192, 250–257.)

I was confused about the timescale of remission. At one point the authors write “KPD patients can often achieve partial remission: after a few weeks or months of treatment with insulin” but later the authors state that “the duration of the remission varies from 6 months to 10 years”.

The former timescale is the typical timescale achieve remission. After remission is reached, however, it may or may not last—patients may experience a relapse, where their condition worsens and they again require insulin. We have edited the text to clarify this distinction (lines 66–73).

When the authors talk about intermediate timescales in the main text could they specify an actual unit of time, such as days, weeks, or months as it would relate to the rate constants in their model for those transitions?

We have done so (lines 86–87, figure 1 caption, figure 2 caption). Getting KPD-like behavior requires (at high glucose) the deactivation process to be somewhat faster than the reactivation process, so the relevant scales are between weeks (reactivation) and days (deactivation at high G).

The authors state ”Our simple model of β-cell adaptation also neglects the known hyperglycemiainduced leftward shift in the insulin secretion curve f(G) in Eq. (2)) ”. This seems an important consideration. Could the authors comment on why they did not model this shift, and/or explicitly discuss how including it is expected to change the model dynamics?

We agree that this process seems potentially relevant, as it seems to happen on a relatively fast timescale compared to glucose-induced β-cell death. It is, however, not so well characterized quantitatively that including it is a simple matter of putting in known values—we would be making assumptions that would complicate the interpretation of our results.

It is clear that this effect will need to be considered when quanitatively modelling real patient data. However, it is also straightforward to argue that this effect by itself cannot produce KPD-like symptoms, and will only tend to reduce the rate of glucotoxocity necessary to produce bibstability. We have added a discussion of this in the revisions (lines 307–315). We have also, in general, expanded the discussion of the effects that each neglected detail we have mentioned is expected to have (lines 292–315).

The authors end with a statement that their results may “contribute to explanation of other observations that involve rapid onset or remission of diabetes-like phenomena, such as during pregnancy or for patients on very low calorie diets.” Could the authors spell out exactly how their model potentially relates to these phenomena?

Our thinking is that, even when another direct cause, such as loss of insulin resistance, is implicated in reversal of diabetes, some portion of the effect may be explained by reversal of glucotoxicity. This is indeed at this point just a hypothesis, but we have expanded on it briefly in the revision. (Lines 281–291.)

Minor typos:

In Figure 2.D the last zero of 200 on the axis was cut off.

Line 359 - there is a missing word ”in the analysis”.

We have fixed these typos, thanks.

Reviewer #2 (Recommendations for the author):

The manuscript could be significantly improved in two key areas: the presentation of the analysis, and the relation with experimental phenomenology.

Regarding the analysis presentation, the figures could be substantially enhanced with minimal effort from the authors. At present, they are sparse, lack legends, and offer only basic analysis. The authors should consider presenting, for example, a bifurcation diagram for beta cell mass and fasting glucose levels as a function of kIN, and how insulin sensitivity and average meal intake modulate this relationship. The goal should be to present clear, testable predictions in an intuitive manner. Currently, the specific testable predictions of the model are unclear.

The response to this question is copied from the reponses to related questions from the first referee.

This is a very good point. We have made several changes. Firstly, we have added smaller panels showing the dynamics of β to Figure 2; previously, the reader thad to infer what was happening to β from G(t). Secondly, we have completely replaced the two figures showing dβ/dt, and requiring the reader to infer the fixed points of β, with bifurcation diagrams that simply show the fixed points of G and β. The new figures show through bifurcation diagrams how there are multiple fixed points in KPD, how glucose or insulin treatment force the switching of fixed points, and how the presence of bistability depends on the rate of glucotoxicity. We have also supplemented our phase diagram that shows the effects of SI and the total beta cell population with bifurcation diagrams showing β as SI and βTOT are varied. (These new figures are Fig. 3–5 in the present manuscript.) Finally, we have added another figure analyzing the model’s predictions for the optimal insulin treatment and the resulting time needed to achieve remission (Fig. 7)

eLife Assessment

The findings are important and intriguing, with theoretical or practical implications beyond a single subfield. The computational methods employed are clever and sophisticated and the strength of evidence is convincing. Both the hypotheses and the exploratory nature of additional analyses are clearly stated.

Reviewer #1 (Public review):

Summary:

The authors use a sophisticated and novel task design and Bayesian computational modeling to test their hypothesis that information generalization (operationalized as a combination of self-insertion and social contagion) in social situations is disrupted in Borderline Personality Disorder. Their main finding relates to the observation that two different models best fit the two tested groups: While the model assuming both self-insertion and social contagion to be present when estimating others' social value preferences fit the control group best, a model assuming neither of these processes provided the best fit to BPD participants.

Strengths:

The two revisions have substantially strengthened the paper and the manuscript is much clearer and easier to follow now. The introduction now precisely states the author's hypotheses, and the connections to the theoretical framework are presented with much greater clarity. I appreciate that the authors now clearly label exploratory analyses where applicable.

The strengths of the presented work lie in the sophisticated task design and the thorough investigation of their theory by use of mechanistic computational models to elucidate social decision-making and learning processes in BPD. Although at present it is not clear whether the differing strategies in impression formation observed in BPD are in any way causal to negative outcomes in the condition, the study represents an important step towards better understanding cognitive processes in BPD. The paradigm and models are also potentially relevant for the investigation of other psychiatric conditions.

eLife Assessment

Argunşah et al. investigate the mechanisms underlying the differential response dynamics of barrel vs septa domains in shaping the responses to single vs multiple whiskers. Based on the observation of a higher density of SST+ interneurons in the septa, the authors investigate the hypothesis that Elfn1-dependent short-term plasticity shapes these responses. This important study is, however, supported by incomplete evidence; factors restricting the strength of evidence are the limited spatial resolution of the multi-unit activity, as well as the lack of a mechanistic explanation. This provocative and intellectually stimulating hypothesis provides a contribution to work on how different cell types shape cortical representation.

Reviewer #1 (Public review):

Summary:

Argunşah et al. describe and investigate the mechanisms underlying the differential response dynamics of barrel vs septa domains of the whisker-related primary somatosensory cortex (S1). Upon repeated stimulation, the authors report that the response ratio between multi- and single-whisker stimulation increases in layer (L) 4 neurons of the septal domain, while remaining constant in barrel L4 neurons. This difference is attributed to the short-term plasticity properties of interneurons, particularly somatostatin-expressing (SST+) neurons. This claim is supported by the increased density of SST+ neurons found in L4 of the septa compared to barrels, along with a stronger response of (L2/3) SST+ neurons to repeated multi- vs single-whisker stimulation. The role of the synaptic protein Elfn1 is then examined. Elfn1 KO mice exhibited little to no functional domain separation between barrel and septa, with no significant difference in single- versus multi-whisker response ratios across barrel and septal domains. Consistently, a decoder trained on WT data fails to generalize to Elfn1 KO responses. Finally, the authors report a relative enrichment of S2- and M1-projecting cell densities in L4 of the septal domain compared to the barrel domain.

Strengths:

This paper describes and aims to study a circuit underlying differential response between barrel columns and septal domains of the primary somatosensory cortex. This work supports the view that barrel and septal domains contribute differently to processing single versus multi-whisker inputs, suggesting that the barrel cortex multiplexes sensory information coming from the whiskers in different domains.

Weaknesses:

While the observed divergence in responses to repeated SWS vs MWS between the barrel and septal domains is intriguing, the presented evidence falls short of demonstrating that short-term plasticity in SST+ neurons critically underpins this difference. The absence of a mechanistic explanation for this observation limits the work's significance. The measurement of SST neurons' response is not specific to a particular domain, and the Elfn1 manipulation does not seem to be specific to either stimulus type or a particular domain. The study's reach is further constrained by the fact that results were obtained in anesthetized animals, which may not generalize to awake states. The statistical analysis appears inappropriate, with the use of repeated independent tests, dramatically boosting the false positive error rate. Furthermore, the manuscript suffers from imprecision; its conclusions are occasionally vague or overstated.

The authors suggest a role for SST+ neurons in the observed divergence in SWS/MWS responses between barrel and septal domains. However, this remains speculative, and some findings appear inconsistent. For instance, the increased response of SST+ neurons to MWS versus SWS is not confined to a specific domain. Why, then, would preferential recruitment of SST+ neurons lead to divergent dynamics between barrel and septal regions? The higher density of SST+ neurons in septal versus barrel L4 is not a sufficient explanation, particularly since the SWS/MWS response divergence is also observed in layers 2/3, where no difference in SST+ neuron density is found. Moreover, SST+ neuron-mediated inhibition is not necessarily restricted to the layer in which the cell body resides. It remains unclear through which differential microcircuits (barrel vs septum) the enhanced recruitment of SST+ neurons could account for the divergent responses to repeated SWS versus MWS stimulation.

The Elfn1 KO mouse model seems too unspecific to suggest the role of the short-term plasticity in SST+ neurons in the differential response to repeated SWS vs MWS stimulation across domains. Why would Elfn1-dependent short-term plasticity in SST+ neurons be specific to a pathway, or a stimulation type (SWS vs MWS)? Moreover, the authors report that Elfn1 knockout alters synapses onto VIP+ as well as SST+ neurons (Stachniak et al., 2021; previous version of this paper)-so why attribute the phenotype solely to SST+ circuitry? In fact, the functional distinctions between barrel and septal domains appear largely abolished in the Elfn1 KO.

Reviewer #2 (Public review):

Summary:

Argunsah and colleagues demonstrate that SST-expressing interneurons are concentrated in the mouse septa and differentially respond to repetitive multi-whisker inputs. Identifying how a specific neuronal phenotype impacts responses is an advance.

Strengths:

(1) Careful physiological and imaging studies.

(2) Novel result showing the role of SST+ neurons in shaping responses.

(3) Good use of a knockout animal to further the main hypothesis.

(4) Clear analytical techniques.

Weaknesses:

No major weaknesses were identified by this reviewer. Overall I appreciated the paper but feel it overlooked a few issues and had some recommendations on how additional clarifications could strengthen the paper. These include:

(1) Significant work from Jerry Chen on how S1 neurons that project to M1 versus S2 respond in a variety of behavioral tasks should be included (e.g. PMID: 26098757). Similarly, work from Barry Connor's lab on intracortical versus thalamocortical inputs to SST neurons, as well as excitatory inputs onto these neurons (e.g. PMID: 12815025) should be included.

(2) Using Layer 2/3 as a proxy to what is happening in layer 4 (~line 234). Given that layer 2/3 cells integrate information from multiple barrels, as well as receiving direct VPm thalamocortical input, and given the time window that is being looked at can receive input from other cortical locations, it is not clear that layer 2/3 is a proxy for what is happening in layer 4.

(3) Line 267, when discussing distinct temporal response, it is not well defined what this is referring to. Are the neurons no longer showing peaks to whisker stimulation, or are the responses lasting a longer time? It is unclear why PV+ interneurons which may not be impacted by the Elfn1 KO and receive strong thalamocortical inputs, are not constraining activity.

(4) Line 585 "the earliest CSD sink was identified as layer 4..." were post-hoc measurements made to determine where the different shank leads were based on the post-hoc histology?

(5) For the retrograde tracing studies, how were the M1 and S2 injections targeted (stereotaxically or physiologically)? How was it determined that the injections were in the whisker region (or not)?

(6) Were there any baseline differences in spontaneous activity in the speta versus barrel regions, and did this change in the KO animals?

Reviewer #3 (Public review):

Summary:

This study investigates the functional differences between barrel and septal columns in the mouse somatosensory cortex, focusing on how local inhibitory dynamics, particularly involving Elfn1-expressing SST⁺ interneurons, may mediate temporal integration of multi-whisker (MW) stimuli in septa. Using a combination of in vivo multi-unit recordings, calcium imaging, and anatomical tracing, the authors propose that septa integrate MW input in an Elfn1-dependent manner, enabling functional segregation from barrel columns.

Strengths:

The core hypothesis is interesting and potentially impactful. While barrels have been extensively characterized, septa remain less understood, especially in mice, and this study's focus on septal integration of MW stimuli offers valuable insights into this underexplored area. If septa indeed act as selective integrators of distributed sensory input, this would add a novel computational role to cortical microcircuits beyond what is currently attributed to barrels alone. The narrative of this paper is intellectually stimulating.

Weaknesses:

The methods used in the current study lack the spatial and cellular resolution needed to conclusively support the central claims. The main physiological findings are based on unsorted multi-unit activity (MUA) recorded via low-channel-count silicon probes. MUA inherently pools signals from multiple neurons across different distances and cell types, making it difficult to assign activity to specific columns (barrel vs. septa) or neuron classes (e.g., SST⁺ vs. excitatory). The recording radius (~50-100 µm or more) and the narrow width of septa (~50-100 µm or less) make it likely that MUA from "septal" electrodes includes spikes from adjacent barrel neurons. The authors do not provide spike sorting, unit isolation, or anatomical validation that would strengthen spatial attribution. Calcium imaging is restricted to SST⁺ and VIP⁺ interneurons in superficial layers (L2/3), while the main MUA recordings are from layer 4, creating a mismatch in laminar relevance.

Furthermore, while the role of Elfn1 in mediating short-term facilitation is supported by prior studies, no new evidence is presented in this paper to confirm that this synaptic mechanism is indeed disrupted in the knockout mice used here. Additionally, since Elfn1 is constitutively knocked out from development, the possibility of altered circuit formation-including changes in barrel structure and interneuron distribution, cannot be excluded and is not addressed.

eLife Assessment

This important study reports that the human posterior inferotemporal cortex (hPIT) functions as an attentional priority map, integrating both top-down and bottom-up attentional signals rather than serving solely as an object-processing region. The experiments and analyses are well conducted and provide convincing evidence that hPIT bridges dorsal and ventral attention networks and is robustly modulated by attention across diverse visual tasks. The study will be relevant for researchers investigating visual attention, high-level visual cortex, and the neural mechanisms that integrate endogenous and exogenous attentional control.

Reviewer #1 (Public review):

The manuscript titled "The distinct role of human PIT in attention control" by Huang et al. investigates the role of the human posterior inferotemporal cortex (hPIT) in spatial attention. Using fMRI experiments and resting-state connectivity analyses, the authors present compelling evidence that hPIT is not merely an object-processing area, but also functions as an attentional priority map, integrating both top-down and bottom-up attentional processes. This challenges the traditional view that attentional control is localized primarily in frontoparietal networks.

The manuscript is strong and of high potential interest to the cognitive neuroscience community. Below, I raise questions and suggestions to help with the reliability, methodology, and interpretation of the findings.

(1) The authors argue that hPIT satisfies the criteria for a priority map, but a clearer justification would strengthen this claim. For example, how does hPIT meet all four widely recognized criteria, such as spatial selectivity, attentional modulation, feature invariance, and input integration, when compared to classical regions such as LIP or FEF? A more systematic summary of how hPIT meets these benchmarks would be helpful. Additionally, to what extent are the observed attentional modulations in hPIT independent of general task difficulty or behavioral performance?

(2) The authors report that hPIT modulation is invariant to stimulus category, but there appear to be subtle category-related effects in the data. Were the face, scene, and scrambled images matched not only in terms of luminance and spatial frequency, but also in terms of factors such as semantic familiarity and emotional salience? This may influence attentional engagement and bias interpretation.

(3) The result that attentional load modulates hPIT is important and adds depth to the main conclusions. However, some clarifications would help with the interpretation. For example, were there observable individual differences in the strength of attentional modulation? How consistent were these effects across participants?

(4) The resting-state data reveal strong connections between hPIT and both dorsal and ventral attention networks. However, the analysis is correlational. Are there any complementary insights from task-based functional connectivity or latency analyses that support a directional flow of information involving hPIT? In addition, do the authors interpret hPIT primarily as a convergence hub receiving input from both DAN and VAN, or as a potential control node capable of influencing activity in these networks? Also, were there any notable differences between hemispheres in either the connectivity patterns or attentional modulation?

(5) A few additional questions arise regarding the anatomical characteristics of hPIT: How consistent were its location and size across participants? Were there any cases where hPIT could not be reliably defined? Given the proximity of hPIT to FFA and LOp, how was overlap avoided in ROI definition? Were the functional boundaries confirmed using independent contrasts?

Reviewer #2 (Public review):

Summary

This study investigates the role of the human posterior inferotemporal cortex (hPIT) in attentional control, proposing that hPIT serves as an attentional priority map that integrates both top-down (endogenous) and bottom-up (exogenous) attentional processes. The authors conducted three types of fMRI experiments and collected resting-state data from 15 participants. In Experiment 1, using three different spatial attention tasks, they identified the hPIT region and demonstrated that this area is modulated by attention across tasks. In Experiment 2, by manipulating the presence or absence of visual stimuli, they showed that hPIT exhibits strong attentional modulation in both conditions, suggesting its involvement in both bottom-up and top-down attention. Experiment 3 examined the sensitivity of hPIT to stimulus features and attentional load, revealing that hPIT is insensitive to stimulus category but responsive to task load - further supporting its role as an attentional priority map. Finally, resting-state functional connectivity analyses showed that hPIT is connected to both dorsal and ventral attention networks, suggesting its potential role as a bridge between the two systems. These findings extend prior work on monkey PITd and provide new insights into the integration of endogenous and exogenous attention.

Strengths

(1) The study is innovative in its use of specially designed spatial attention tasks to localize and validate hPIT, and in exploring the region's role in integrating both endogenous and exogenous attention, as prior works focus primarily on its involvement in endogenous attention.

(2) The authors provided very comprehensive experiment designs with clear figures and detailed descriptions.

(3) A broad range of analyses was conducted to support the hypothesis that hPIT functions as an attentional priority map -- including experiments of attentional modulation under both top-down and bottom-up conditions, sensitivity to stimulus features and task load, and resting-state functional connectivity. These analyses showed consistent results.

(4) Multiple appropriate statistical analyses - including t-tests, ANOVAs, and post-hoc tests - were conducted, and the results are clearly reported.

Weaknesses

(1) The sample size is relatively small (n = 15), and inter-subject variability is big in Figures 5 and 6, as seen in the spread of individual data points and error bars. The analysis of attention-modulated voxel map intersections appears to be influenced by multiple outliers.

(2) The authors acknowledge important limitations, including the lack of exploration of feature-based attention and the temporal constraints inherent to fMRI.

(3) Prior research has established that regions such as the prefrontal cortex (PFC) and posterior parietal cortex (PPC) are involved in both endogenous and exogenous attention and have been proposed as attentional priority maps. It remains unclear what is uniquely contributed by hPIT, how it functionally interacts with these classical attentional hubs, and whether its role is complementary or redundant. The study would benefit from more direct comparisons with these regions.

(4) The functional connectivity analysis is only performed on resting-state data, and this approach does not capture context-dependent interactions. Task-based data analysis can provide stronger evidence.

(5) The study does not report whether attentional modulation in hPIT is consistent across the two hemispheres. A comparison of hemispheric effects could provide important insight into lateralization and inter-individual variability, especially given the bilateral localization of hPIT.

Author response:

Reviewer #1 (Public review):

The manuscript titled "The distinct role of human PIT in attention control" by Huang et al. investigates the role of the human posterior inferotemporal cortex (hPIT) in spatial attention. Using fMRI experiments and resting-state connectivity analyses, the authors present compelling evidence that hPIT is not merely an object-processing area, but also functions as an attentional priority map, integrating both top-down and bottom-up attentional processes. This challenges the traditional view that attentional control is localized primarily in frontoparietal networks.

The manuscript is strong and of high potential interest to the cognitive neuroscience community. Below, I raise questions and suggestions to help with the reliability, methodology, and interpretation of the findings.

Thank you for a nice summary of the key points of our study. Below you will find our responses to your questions.

(1) The authors argue that hPIT satisfies the criteria for a priority map, but a clearer justification would strengthen this claim. For example, how does hPIT meet all four widely recognized criteria, such as spatial selectivity, attentional modulation, feature invariance, and input integration, when compared to classical regions such as LIP or FEF? A more systematic summary of how hPIT meets these benchmarks would be helpful. Additionally, to what extent are the observed attentional modulations in hPIT independent of general task difficulty or behavioral performance?

Great suggestions! For the first suggestion, we will include a clearer justification in the revised manuscript. For the second one, all participants received task practice prior to scanning, and task accuracy exceeded 90% (we will explicitly report the accuracy rate in revision), suggesting the tasks were not overly demanding. Although ceiling effects limit the interpretability of behavioral-performance correlations, we argue that higher task demands would likely require greater attentional effort, leading to stronger modulation in hPIT, which aligns with our findings when we manipulated the attentional load.

(2) The authors report that hPIT modulation is invariant to stimulus category, but there appear to be subtle category-related effects in the data. Were the face, scene, and scrambled images matched not only in terms of luminance and spatial frequency, but also in terms of factors such as semantic familiarity and emotional salience? This may influence attentional engagement and bias interpretation.

The response of hPIT is generally insensitive to stimulus category, however, the reviewer is correct in noticing that attentional modulation in hPIT is slightly stronger to faces than scenes and scrambled images. Although faces used in the task had neutral expressions and the scene pictures were also neutral, it is indeed possible that potential semantic familiarity or emotional salience may contribute to the subtle category-related effects in the results of experiment 3. This point will be noted in the revised manuscript.

(3) The result that attentional load modulates hPIT is important and adds depth to the main conclusions. However, some clarifications would help with the interpretation. For example, were there observable individual differences in the strength of attentional modulation? How consistent were these effects across participants?

Yes, individual differences exist. In the revised manuscript, we will include individual subject data points in the figure 6B.

(4) The resting-state data reveal strong connections between hPIT and both dorsal and ventral attention networks. However, the analysis is correlational. Are there any complementary insights from task-based functional connectivity or latency analyses that support a directional flow of information involving hPIT? In addition, do the authors interpret hPIT primarily as a convergence hub receiving input from both DAN and VAN, or as a potential control node capable of influencing activity in these networks? Also, were there any notable differences between hemispheres in either the connectivity patterns or attentional modulation?

We agree that besides resting-state connection, task-based functional connectivity analyses would have the potential to provide additional information about whether hPIT serves as a convergence node or a control hub. While fMRI data are not the best to generate directional flow of information due to the low temporal resolution, we will conduct task-based functional connectivity analyses.

We also observed modest hemispheric asymmetries in connectivity—for instance, both left and right hPIT showed stronger connectivity with right-hemisphere attention nodes. This will be described in the revised supplement.

(5) A few additional questions arise regarding the anatomical characteristics of hPIT: How consistent were its location and size across participants? Were there any cases where hPIT could not be reliably defined? Given the proximity of hPIT to FFA and LOp, how was overlap avoided in ROI definition? Were the functional boundaries confirmed using independent contrasts?

The size and location of hPIT are generally consistent across subjects, as shown in Supplementary Figure 1. The consistency is also supported by figure 4C. The hPIT is defined by conjunction maps across three tasks and then manually delineated avoiding overlapping voxels with FFA and LOp. The FFA was defined using an independent contrast (Exp3 contrast [face-scene]) and the Lop location was defined by anatomical parcellation (Glasser et al., 2016).

Reviewer #2 (Public review):

Summary

This study investigates the role of the human posterior inferotemporal cortex (hPIT) in attentional control, proposing that hPIT serves as an attentional priority map that integrates both top-down (endogenous) and bottom-up (exogenous) attentional processes. The authors conducted three types of fMRI experiments and collected resting-state data from 15 participants. In Experiment 1, using three different spatial attention tasks, they identified the hPIT region and demonstrated that this area is modulated by attention across tasks. In Experiment 2, by manipulating the presence or absence of visual stimuli, they showed that hPIT exhibits strong attentional modulation in both conditions, suggesting its involvement in both bottom-up and top-down attention. Experiment 3 examined the sensitivity of hPIT to stimulus features and attentional load, revealing that hPIT is insensitive to stimulus category but responsive to task load - further supporting its role as an attentional priority map. Finally, resting-state functional connectivity analyses showed that hPIT is connected to both dorsal and ventral attention networks, suggesting its potential role as a bridge between the two systems. These findings extend prior work on monkey PITd and provide new insights into the integration of endogenous and exogenous attention.

Strengths

(1) The study is innovative in its use of specially designed spatial attention tasks to localize and validate hPIT, and in exploring the region's role in integrating both endogenous and exogenous attention, as prior works focus primarily on its involvement in endogenous attention.

(2) The authors provided very comprehensive experiment designs with clear figures and detailed descriptions.

(3) A broad range of analyses was conducted to support the hypothesis that hPIT functions as an attentional priority map -- including experiments of attentional modulation under both top-down and bottom-up conditions, sensitivity to stimulus features and task load, and resting-state functional connectivity. These analyses showed consistent results.

(4) Multiple appropriate statistical analyses - including t-tests, ANOVAs, and post-hoc tests - were conducted, and the results are clearly reported.

Thank you for a nice summary of the key points and strengths of our study.

Weaknesses

(1) The sample size is relatively small (n = 15), and inter-subject variability is big in Figures 5 and 6, as seen in the spread of individual data points and error bars. The analysis of attention-modulated voxel map intersections appears to be influenced by multiple outliers.

We agree that the sample size (n = 15) is not ideal, and we acknowledge that some data points in Figures 5 and 6 appear to be potential outliers. However, according to conventional outlier detection criteria, all data points are within three standard deviations of the group mean and were therefore retained for analysis. Moreover, the attention-modulated voxel intersection map shown in Figure 4C is insensitive to outliers, because the intersection map plotted is based on the number of subjects.

(2) The authors acknowledge important limitations, including the lack of exploration of feature-based attention and the temporal constraints inherent to fMRI.

Yes, we hope to address these limitations in future studies.

(3) Prior research has established that regions such as the prefrontal cortex (PFC) and posterior parietal cortex (PPC) are involved in both endogenous and exogenous attention and have been proposed as attentional priority maps. It remains unclear what is uniquely contributed by hPIT, how it functionally interacts with these classical attentional hubs, and whether its role is complementary or redundant. The study would benefit from more direct comparisons with these regions.

In this study, we define the ROI base on intersection across three different types of spatial attention tasks, and the hPIT stands out in showing spatial attentional modulation across tasks. This could be due to the weak lateralized responses in PFC/PPC. To evaluate whether a region qualifies as a priority map, we applied four criteria (as mentioned in introduction). While dorsal and ventral attention network (DAN and VAN) regions can be considered important components of the priority map system, our findings suggest that among the regions tested, hPIT meets all four criteria. In Experiment 2, we included regions such as VFC (as part of PFC) and IPS (as part of PPC), and our findings suggest these areas are more involved in top-down attention. We agree with the reviewer’s suggestion and will perform additional analysis on PPC and PFC.

(4) The functional connectivity analysis is only performed on resting-state data, and this approach does not capture context-dependent interactions. Task-based data analysis can provide stronger evidence.

We acknowledge that resting-state FC is limited in assessing task-specific communication. To further investigate the role of hPIT, we plan to conduct task-based functional connectivity analyses.

(5) The study does not report whether attentional modulation in hPIT is consistent across the two hemispheres. A comparison of hemispheric effects could provide important insight into lateralization and inter-individual variability, especially given the bilateral localization of hPIT.

We thank the reviewer for this suggestion. hPIT was localized bilaterally using the same intersection-based method in Experiment 1. We have now performed additional analysis and found in Experiment 3, the difference in attentional modulation between high and low load conditions was significant in the right hPIT but not in the left. This result will be reported in the revised manuscript.

eLife Assessment

This study provides important findings on the neural circuits underlying dishabituation of the olfactory avoidance response in Drosophila. The data as presented provide solid evidence that the dishabituation involves distinct pathways from habituation. They show that reward-activated dopaminergic neurons provide input for within-modal dishabituation, while punishment-activated dopaminergic neurons provide input for cross-modal dishabituation. The work will interest neuroscientists, particularly behavioral neuroscientists working on habituation, neural circuits, and the dopaminergic system.

Reviewer #1 (Public review):

Summary:

Charonitakis and co-authors characterize dishabituation in adult flies, where they use olfactory habituation to octanol, then dishabituate the flies with disruptions of electric shock or yeast odors. They systematically investigate the neurotransmitters and neural circuits involved in dishabituation and figure out a lot about how this process works in the brain, as an independent circuit. I like the paper, and I like the very structured approach to figuring out the problem.

Strengths:

The introduction nicely sets the stage for the work presented, bringing in knowledge from other organisms and motivating the study.

The results section lays out a logical set of experiments, using a common set of behavioral assays in many flies exposed to thermogenetic or optogenetic manipulation. The paper systematically figures out the necessity and/or sufficiency of specific brain regions and neurotransmitters, culminating in a new understanding of how the important process of dishabituation works.

I like the bar graph representation for the data throughout, with the helpful icons - if a paper figures are going to be 90% bar graphs, it helps when they are super clear like this! And I like how all the parts build up to the conclusion in the last figure, nicely summarizing the thorough characterization of dishabituation.

Weaknesses:

There are no major concerns, but some material could be added for clarity and to make the work more accessible to a more general scientific audience. A figure clearly showing the habituation protocol and the use of the dishabituators would be a good addition, even if the procedure has been done before and is cited. There can always be readers who are seeing this for the first time.

It would also be nice to comment on other ways dishabituation can happen (for example, when the stimulus is removed for a short time and returns) and what their time scales are.

And more generally, the paper could perhaps improve by making a stronger case for why the results are important not just for flies but for neuroscience in general.

Reviewer #2 (Public review):

This is an interesting study in Drosophila comparing potentially differential requirements for subsets of Kenyon Cells (KCs) and Dopaminergic neurons (DANS) in olfactory dishabituation driven by either a novel odor ("homosensory") or footshock ("heterosensory). The authors measure olfactory aversion to Octanol (OCT) in a T-maze, induce olfactory habituation with a 4-minute prior exposure to OCT, and use either brief yeast odor (YO) or footshock (FS) to achieve dishabituation. The major observation that YO-mediated dishabituation is mediated by reward-activated DANs (PAM cluster), while FS-mediated dishabituation is mediated by punishment-activated PPL-DANs is generally solid and convincing. Also convincing are experiments showing the involvement of KCs in the pathway for YO and FS-induced dishabituation, and the argument that KCs drive DAN activation that causes dishabituation, though not experimentally shown, is more than reasonable. The work is significant because, as the authors take pains to point out, circuits and pathways for dishabituation have been very lightly studied, and clear identification of dopaminergic neuron subsets in dishabituation achieved by different means represents unique and interesting progress.

However, the claim that this represents a fundamental difference between homosensory and heterosensory pathways for dishabituation is overstated. The introductory section does not adequately present current broad models for habituation and dishabituation. There are many different time scales, even for Drosophila olfactory habituation. These, as well as potential underlying mechanistic differences, need to be acknowledged; any claim should be specifically qualified for the time scales being studied here. Additionally, there are several unclear, vague, and inaccurate sections and statements. A more careful, precise, and considered presentation of current views, as well as more measured claims of the impact of the findings, would substantially enhance my enthusiasm.

Reviewer #3 (Public review):

Summary:

In this manuscript, Charonitakis, Pasadaki et al. investigated the neural circuits underlying homosensory/within-modal and heterosensory/cross-modal dishabituation of the olfactory avoidance response in Drosophila. Taking advantage of the accessible and sophisticated gene expression manipulation tools in the flies, this study traced neural pathways underlying response facilitation caused by different types of sensory stimuli and revealed both distinct and convergent neural components underlying these different forms of behavioral plasticity. The study first demonstrated that olfactory habituation of the octanol avoidance response can be facilitated by either a different odor (homosensory stimulus) or a foot shock (heterosensory stimulus). Then, the flies' nervous system was manipulated with gene expression tools to identify key neural components involved in mediating the behavioral facilitation caused by different types of sensory stimuli. It was found that different sensory stimuli are input into different parts of the nervous system, and signals converge in the mushroom bodies to generate response facilitation. It was also found that these facilitatory pathways are different from the olfactory habituation pathway in the lateral horns.

Strengths:

The authors took full advantage of the advanced genetic tools in flies and performed a series of experiments to pinpoint neural components in each pathway.

Weaknesses:

The key issue is that the main concepts of this manuscript appear to be based on a misunderstanding/misinterpretation of the literature. As the authors set out to settle the debate "whether the novel dishabituating stimulus elicits sensitization of the habituated circuits, or it engages distinct neuronal routes to bypass habituation reinstating the naïve response", it seems that the authors based their investigation on the premise that "sensitization" is mediated by a facilitatory process within the S-R pathway, and "dishabituation" by a facilitatory process outside the S-R pathway. This is not the status quo in the field, particularly with the prevailing theory like the Dual-Process Theory.

The original version of Dual-Process Theory (Groves and Thompson 1970, but also see Thompson 2008, Neurobiol Learn Mem) already hypothesized that habituation happens within the specific S-R pathway, and sensitization occurs separately in an "organism-wide" state system that modulates the output of all S-R pathways. Dishabituation is recognized by the Dual-Process Theory as sensitization (organism-wide facilitation) manifested on top of existing habituation (depressed S-R pathway). This notion has been supported by a wide range of studies, including cat spinal cord reflex (e.g. Spencer et al. 1966) and work in Aplysia on heterosynaptic facilitation for both sensitization and dishabituation. Therefore, simply showing that the newly identified facilitatory pathways are outside the S-R habituation pathway is insufficient to demonstrate dishabituation.

As behavioral facilitation of a habituated response can be achieved by dishabituating (specific recovery of the S-R pathway) and/or superimposed sensitizing (organism-wide) processes, dishabituation and sensitization of this olfactory response must be first dissociated; however, the study provided no evidence for the dissociation. Without this piece of evidence, the claim of this paper that the newly identified pathways mediate dishabituation is not fully supported.

The literature review of this manuscript has some discrepancies. In the introduction, the authors wrote "initial studies in Aplysia were consistent with the "dual-process theory" (Groves and Thompson 1979), where response recovery due to dishabituation appeared to result from sensitization superimposed on habituation, thus driving reversal of the attenuated response (Carew, Castellucci et al. 1971, Hochner, Klein et al. 1986, Marcus, Nolen et al. 1988, Ghirardi, Braha et al. 1992, Cohen, Kaplan et al. 1997, Antonov, Kandel et al. 1999, Hawkins, Cohen et al. 2006)." Hochner 1986 and Marcus 1988 in fact indicated otherwise. Hochner 1986 suggests that dishabituation and sensitization involve different molecular processes, while Marcus 1988 showed that dishabituation and sensitization have different behavioral characteristics. Therefore, the authors' statement is not supported by the cited literature.

Author response:

Below, we will address point by point any and all concerns of the reviewers.

Reviewer #1:

There are no major concerns, but some material could be added for clarity and to make the work more accessible to a more general scientific audience.

We will add text for clarity and to make the work more accessible to a general audience per this comment and similar suggestions of the other reviewers.

(1.1) A figure clearly showing the habituation protocol and the use of the dishabituators would be a good addition, even if the procedure has been done before and is cited. There can always be readers who are seeing this for the first time.

We do think this is a good idea as the time scales of the experiment will be clearly marked as well and we plan to generate one in the revised manuscript.

(1.2) It would also be nice to comment on other ways dishabituation can happen (for example, when the stimulus is removed for a short time and returns) and what their time scales are.

If the stimulus is withheld, spontaneous recovery occurs, a process distinct from dishabituation and worth exploring on its own. In a previous publication (Semelidou et al. eLife 2018;7:e39569), we have shown that in this habituation paradigm with 4 min exposure either to the aversive Octanol, or the attractive Ethyl Acetate, spontaneous recovery occurs on or after 6 minutes after the habituated stimulus is withheld. This contrasts the immediate effect of the single dishabituating stimulus, delivered for a few seconds at the end of exposure to the habituator. Granted that per Thomson (Neurobiol Learn Mem. 2009), spontaneous recovery is a characteristic of habituation, we will work this point in the text.

(1.3) And more generally, the paper could perhaps improve by making a stronger case for why the results are important not just for flies but for neuroscience in general.

Thank you for the encouragement. We will try to rationally generalize our findings.

Reviewer #2:

(2.1) However, the claim that this represents a fundamental difference between homosensory and heterosensory pathways for dishabituation is overstated.

We had no intention of stating more than the fact that footshock and yeast odor dishabituators relay these stimuli to the mushroom bodies via distinct dopaminergic neurons, hence differentiating distinct dishabituating stimuli via the mechanosensory (footshock) and olfactory (yeast odor) modalities as they engage the mushroom bodies. As the reviewer suggests we will use more measured and specific language to state the above.

(2.2) The introductory section does not adequately present current broad models for habituation and dishabituation.

This was not done intentionally, but rather because we aimed at a less extended introductory section and ostensibly this resulted in brief and possibly inadequate presentation of current habituation models. We will present a much more detailed introduction and detail of habituation and dishabituation models in the revised manuscript (Also see reply to point 3.5 below).

(2.3) There are many different time scales, even for Drosophila olfactory habituation. These, as well as potential underlying mechanistic differences, need to be acknowledged; any claim should be specifically qualified for the time scales being studied here.

We understand and appreciate the point of the reviewer, as well as its significance and we will address this both in the revised text, but also by the paradigm figure we will add as stated above (point 1.1), where the time scales will be explicitly included and emphasized.

(2.4) Additionally, there are several unclear, vague, and inaccurate sections and statements. A more careful, precise, and considered presentation of current views, as well as more measured claims of the impact of the findings, would substantially enhance my enthusiasm.

We will address these concerns of course, though pointing out the specific offending parts would ascertain addressing them thoroughly. As stated above, we will incorporate current views in the introduction and when discussing our results and their impact.

Reviewer #3:

(3.1) The key issue is that the main concepts of this manuscript appear to be based on a misunderstanding/misinterpretation of the literature. As the authors set out to settle the debate "whether the novel dishabituating stimulus elicits sensitization of the habituated circuits, or it engages distinct neuronal routes to bypass habituation reinstating the naïve response", it seems that the authors based their investigation on the premise that "sensitization" is mediated by a facilitatory process within the S-R pathway, and "dishabituation" by a facilitatory process outside the S-R pathway. This is not the status quo in the field, particularly with the prevailing theory like the Dual-Process Theory.

We appreciate the reviewer’s comment and the opportunity to clarify the conceptual framework of our work. Our intention was in fact to test the Groves and Thomson hypothesis (Neurobiol Learn Mem. 2009), in our olfactory habituation system. As such, dishabituation could have been the result of a facilitatory process within the S-R pathway, or from mechanisms outside of it. Our experimental design allowed to distinguish these possibilities and our results clearly show that dishabituation involves circuitry outside the S-R pathway. We do thank the reviewer for pointing out that we have not articulated clearly this intention and we will take care to communicate this effectively in the revised manuscript.

(3.2) The original version of Dual-Process Theory (Groves and Thompson 1970, but also see Thompson 2008, Neurobiol Learn Mem) already hypothesized that habituation happens within the specific S-R pathway, and sensitization occurs separately in an "organism-wide" state system that modulates the output of all S-R pathways.

As mentioned above, we are aware of the Dual-Process hypothesis. In fact, our data demonstrate that activity outside the olfactory S-R pathway, engaging novel neuronal circuits, mediates dishabituation. Unlike habituation, these circuits mediating dishabituation include at minimum, the mushroom bodies, the dopaminergic system and the APL neurons. In our view this does not support the “organism-wide state” system, but rather particular circuits that in agreement with the Groves and Thomson hypothesis, are outside the S-R pathway and modulate its behavioral output. We will work these concepts in the discussion section of the revised manuscript.

(3.3) Dishabituation is recognized by the Dual-Process Theory as sensitization (organism-wide facilitation) manifested on top of existing habituation (depressed S-R pathway). This notion has been supported by a wide range of studies, including cat spinal cord reflex (e.g. Spencer et al. 1966) and work in Aplysia on heterosynaptic facilitation for both sensitization and dishabituation. Therefore, simply showing that the newly identified facilitatory pathways are outside the S-R habituation pathway is insufficient to demonstrate dishabituation.

We respectfully disagree with the concluding sentence here. In all of our experiments, we observe a clear recovery of olfactory avoidance after exposure to the footshock, or yeast odor dishabituators. Moreover, the dishabituators are emulated by (photo)activation of particular neuronal circuits and the recovery of olfactory avoidance is blocked when these circuits are silenced. Regardless of whether this recovery is classified as dishabituation via sensitization or another facilitatory process, the key point is that the habituated response is reliably reinstated contingent upon the dishabituating stimulus. We believe this meets the established criteria for dishabituation.

(3.4) As behavioral facilitation of a habituated response can be achieved by dishabituating (specific recovery of the S-R pathway) and/or superimposed sensitizing (organism-wide) processes, dishabituation and sensitization of this olfactory response must be first dissociated; however, the study provided no evidence for the dissociation. Without this piece of evidence, the claim of this paper that the newly identified pathways mediate dishabituation is not fully supported.

We agree with the reviewer that we have not provided specific evidence dissociating dishabituation and sensitization of the particular olfactory response beyond the evidence implicating particular circuitry in the outcome of facilitation of the olfactory response.

It should be noted that in photoactivation of the implicated circuitries in naïve flies, we do not observe enhanced octanol avoidance, suggesting that activation of these circuits alone does not induce sensitization. Moreover, our results show that neither footshock nor yeast odor drive an organism-wide sensitization, as silencing specific circuits was sufficient to block dishabituation—something that would not be expected if a global sensitization process was responsible of reinstating the olfactory response.

Nonetheless, we will also attempt to dissociate sensitization from dishabituation using mutants previously reported deficient in sensitization (Duerr and Quinn, PNAS 1982), assuming these mutants retain normal olfactory habituation. We will also try sensitization protocols in the case of within-modal dishabituation to further clarify the underlying mechanisms. In principle, this includes using diluted Octanol as the habituating stimulus and attempt dishabituation with concentrated octanol.

(3.5) The literature review of this manuscript has some discrepancies. In the introduction, the authors wrote "initial studies in Aplysia were consistent with the "dual-process theory" (Groves and Thompson 1979), where response recovery due to dishabituation appeared to result from sensitization superimposed on habituation, thus driving reversal of the attenuated response (Carew, Castellucci et al. 1971, Hochner, Klein et al. 1986, Marcus, Nolen et al. 1988, Ghirardi, Braha et al. 1992, Cohen, Kaplan et al. 1997, Antonov, Kandel et al. 1999, Hawkins, Cohen et al. 2006)." Hochner 1986 and Marcus 1988 in fact indicated otherwise. Hochner 1986 suggests that dishabituation and sensitization involve different molecular processes, while Marcus 1988 showed that dishabituation and sensitization have different behavioral characteristics. Therefore, the authors' statement is not supported by the cited literature.

We are grateful to the reviewer for pointing out these significant discrepancies, consequent of multiple rounds of edits followed by our own oversight. These important publications for this manuscript will be referenced properly in the revised version of the manuscript.

eLife Assessment

This manuscript presents a valuable computational tool for identifying 3-5 gene regulatory network topologies capable of generating oscillatory dynamics. The application of Monte Carlo Tree Search to circuit design is novel and effectively expands the scale at which non-linear behaviours can be explored in silico. The efficiency of the proposed algorithm is convincing, and the work will be of interest to the systems and synthetic biology communities. While the evolutionary implications remain unclear, the methodological contribution represents a significant advance in the field.

Joint Public Review:

This manuscript presents an algorithm for identifying network topologies that exhibit a desired qualitative behaviour, with a particular focus on oscillations. The approach is first demonstrated on 3-node networks, where results can be validated through exhaustive search, and then extended to 5-node networks, where the search space becomes intractable. Network topologies are represented as directed graphs, and their dynamical behaviour is classified using stochastic simulations based on the Gillespie algorithm. To efficiently explore the large design space, the authors employ reinforcement learning via Monte Carlo Tree Search (MCTS), framing circuit design as a sequential decision-making process.

This work meaningfully extends the range of systems that can be explored in silico to uncover non-linear dynamics and represents a valuable methodological advance for the fields of systems and synthetic biology.

Strengths

The evidence presented is strong and compelling. The authors validate their results for 3-node networks through exhaustive search, and the findings for 5-node networks are consistent with previously reported motifs, lending credibility to the approach. The use of reinforcement learning to navigate the vast space of possible topologies is both original and effective, and represents a novel contribution to the field. The algorithm demonstrates convincing efficiency, and the ability to identify robust oscillatory topologies is particularly valuable. Expanding the scale of systems that can be systematically explored in silico marks a significant advance for the study of complex gene regulatory networks.

Weaknesses

The principal weakness of the manuscript lies in the interpretation of biological robustness. The authors identify network topologies that sustain oscillatory behaviour despite perturbations to the system or parameters. However, in many cases, this persistence is due to the presence of partially redundant oscillatory motifs within the network. While this observation is interesting and of clear value for circuit design, framing it as evidence of evolutionary robustness may be misleading. The "mutant" systems frequently exhibit altered oscillatory properties, such as changes in frequency or amplitude. From a functional cellular perspective, mere oscillation is insufficient - preservation of specific oscillation characteristics is often essential. This is particularly true in systems like circadian clocks, where misalignment with environmental cycles can have deleterious effects. Robustness, from an evolutionary standpoint, should therefore be framed as the capacity to maintain the functional phenotype, not merely the qualitative behaviour.

A secondary limitation is that, despite the methodological advances, the scale of the systems explored remains modest. While moving from 3- to 5-node systems is non-trivial, five elements still represent a relatively small network. It is somewhat surprising that the algorithm does not scale further, particularly when considering the performance of MCTS in other domains - for instance, modern chess engines routinely explore far larger decision trees. A discussion on current performance bottlenecks and potential avenues for improving scalability would be valuable.

Finally, it is worth noting that the emergence of oscillations in a model often depends not only on the topology but also critically on parameter choices and the nature of the nonlinearities. The use of Hill functions and high Hill coefficients is a common strategy to induce oscillatory dynamics. Thus, the reported results should be interpreted within the context of the modelling assumptions and parameter regimes employed in the simulations.

eLife Assessment

This useful study reports findings that support the use of the Open Field Test in Drosophila as a model to study "emotion-like states", which are behavioral responses to several stressful or aversive treatments, and resilience upon their subsequent removal. Behavioral data, by employing established stress-causing treatments and genetic manipulations, are solid. While the results and conceptual framework of this work will be of interest to behaviorists regardless of animal models, the novelty of this work over previous studies could have been clearer.

Reviewer #1 (Public review):

Summary:

Animal behavior is continuously influenced by the internal state moment-by-moment, including emotion primitives, as the authors pointed out. Although emotion is a more human-related state, evolutionary conservation is undeniable, which can be inferred by the behavioral manifestation. To further elaborate on the neuronal mechanisms of emotion primitives, the simplest behavioral parameter related to emotional primitives should be well-characterized. In this study, the authors described in detail wall-following behavior (WAFO) and the total walking distance (TOWA) using flies after subjecting them to various conditions or flies being genetically manipulated according to the previous reports that could affect emotion primitives. Overall, the study is well designed and structured. In addition, the discussion on emotion primitives will be of value to the field.

Strengths:

The strength of this study is its use of a simple behavioral parameter, TOWA, and also a simple design of behavior, WAFO. The importance of the behavioral assay is reproducibility and comparability. In fact, the author demonstrated a summary of comparisons where different treatments result in scalable behavioral changes in WAFO and TOWA.

Weaknesses:

The weakness of the study is the lack of further experiments to support their assumption related to TOWA.

The authors suggested that TOWA can be interpreted as a behavioral proxy for exogenously induced arousal. However, it could be interpreted as higher activity, although the authors argued that the circadian clock increasing locomotor activity around ZT0 and ZT12 does not affect TOWA, and therefore TOWA is not related to the locomotor activity per se. As the author cited, flies lose locomotor activity in the circular arena of 6.6 cm in diameter, whereas they continuously move during a 1-h recording in the authors' arena of 1 cm in diameter.

I would agree that the arena of 1 cm in diameter, but not 6.6 cm in diameter, serves as an exogenous stimulus inducing arousal, and TOWA is manifested by arousal. However, TOWA would also be affected by other behavioral parameters, including the activity, motivation for exploration, or perception of the space. Therefore, it could be reasonable to re-examine some of the flies tested in this study in the circular arena of 6.6 cm in diameter. If arousal is biased by the components presented in Figure 6 and TOWA can assess mainly exogenously induced arousal, the treatment altering TOWA in the arena of 1 cm in diameter would not affect their behavior in the arena of 6.6 cm in diameter. My concern is that Figure 6 may demonstrate too simplistic a diagram to interpret the results. I would suggest adding the experiments using the arena of 6.6 cm diameter or softening the argument.

Reviewer #2 (Public review):

Summary:

This work seeks to establish the Open Field Test (OFT) as a paradigm to measure emotion-like states in the fruit fly Drosophila. To do this, the authors first applied various stressors and aversive stimuli to wild-type flies and tracked their locomotion. By measuring wall-following (WAFO) and total walking (TOWA), they showed that these behaviors are generally increased by stressors, but return to baseline levels after their removal. Then, they used the same approach to analyze the effects of pharmacological, genetic, and neuronal activity manipulations, showing that diazepam, serotonin, dopamine, and neuropeptide F affect locomotion in the OFT in largely expected ways that are consistent with their functions in rodents. Finally, the authors demonstrate that wild-type fly strains from the laboratory or caught in the wild differ significantly in their OFT behavior, with wild-caught flies generally behaving as if more 'stressed'. Given the numerous advantages of Drosophila, this study can form the foundation for using the OFT in conjunction with this animal model to elucidate the molecular and neuronal mechanisms that underlie emotion primitives.

Strengths:

The main strength of the paper is the rigorous use of several stressful or aversive treatments and their subsequent removal to show that WAFO is a robust proxy for stress-like emotional primitives across multiple stimuli. The pharmacological, molecular, and neuronal activity manipulations, although more limited in scope, lend further credence to the authors' central claim.

Weaknesses:

The conceptual advance of this research is unclear, as previous work (Mohammad et al., 2016, Curr Biol.) carried out similar treatments and manipulations and reached largely similar conclusions. Moreover, while WAFO is a good proxy for 'stress', I am not convinced that TOWA necessarily represents an emotional state in all cases. Indeed, as the authors themselves acknowledge, changes in total walking may be associated with other factors, such as starvation-induced hyperactivity, physical exhaustion after sleep deprivation, increased sex drive after mating, alcohol sedation, etc. Another unclear point is the interpretation of some unexpected results, such as the finding that both serotonin transporter overexpression and its knockdown give the same phenotype. Finally, there are some issues with the use of the OFT in rodent research (e.g., inconsistent effects of anxiolytic drugs; see Rosso et al., 2022, Neurosci Biobehav Rev., for a meta-analysis). These should be explained to place the Drosophila findings in their appropriate context.

Reviewer #3 (Public review):

Summary:

This is a valuable study providing solid evidence that the putative non-canonical initiation factor eIF2A has little or no role in the translation of any expressed mRNAs in cultured human (primarily HeLa) cells. Previous studies have implicated eIF2A in GTP-independent recruitment of initiator tRNA to the small (40S) ribosomal subunit, a function analogous to canonical initiation factor eIF2, and in supporting initiation on mRNAs that do not require scanning to select the AUG codon or that contain near-cognate start codons, especially upstream ORFs with non-AUG start codons, and may use the cognate elongator tRNA for initiation. Moreover, the detected functions for eIF2A were limited to, or enhanced by, stress conditions where canonical eIF2 is phosphorylated and inactivated, suggesting that eIF2A provides a back-up function for eIF2 in such stress conditions. CRISPR gene editing was used to construct two different knock-out cell lines that were compared to the parental cell line in a large battery of assays for bulk or gene-specific translation in both unstressed conditions and when cells were treated with inhibitors that induce eIF2 phosphorylation. None of these assays identified any effects of eIF2A KO on translation in unstressed or stressed cells, indicating little or no role for eIF2A as a back-up to eIF2 and in translation initiation at near-cognate start codons, in these cultured cells.

The study is very thorough and generally well executed, examining bulk translation by puromycin labeling and polysome analysis and translational efficiencies of all expressed mRNAs by ribosome profiling, with extensive utilization of reporters equipped with the 5'UTRs of many different native transcripts to follow up on the limited number of genes whose transcripts showed significant differences in translational efficiencies (TEs) in the profiling experiments. They also looked for differences in translation of uORFs in the profiling data and examined reporters of uORF-containing mRNAs known to be translationally regulated by their uORFs in response to stress, going so far as to monitor peptide production from a uORF itself. The high precision and reproducibility of the replicate measurements instil strong confidence that the myriad of negative results they obtained reflects the lack of eIF2A function in these cells rather than data that would be too noisy to detect small effects on the eIF2A mutations. They also tested and found no evidence for a recent claim that eIF2A localizes to the cytoplasm in stress and exerts a global inhibition of translation. Given the numerous papers that have been published reporting functions of eIF2A in specific and general translational control, this study is important in providing abundant, high-quality data to the contrary, at least in these cultured cells.

Strengths:

The paper employed two CRISPR knock-out cell lines and subjected them to a combination of high-quality ribosome profiling experiments, interrogating both main coding sequences and uORFs throughout the translatome, which was complemented by extensive reporter analysis, and cell imaging in cells both unstressed and subjected to conditions of eIF2 phosphorylation, all in an effort to test previous conclusions about eIF2A functioning as an alternative to eIF2.

Weaknesses:

No major issues were observed as the authors have provided additional evidence of the extent of ISR induction by tunicamycin. The discussion was also expanded to address concerns stemming from the previous version of the manuscript.

[Editors note: Reviewers and editors concluded that the authors revised the article in a satisfactory manner and no further concerns were raised]

Reviewer #2 (Public review):

Summary

Roiuk et al describe a work in which they have investigated the role of eIF2A in translation initiation in mammals without much success. Thus, the manuscript focuses on negative results. Further, the results, while original, are generally not novel, but confirmatory, since related claims have been made before independently in different systems with Haikwad et al study recently published in eLife being the most relevant.

Despite this, we find this work highly important. This is because of a massive wealth of unreliable information and speculations regarding eIF2A role in translation arising from series of artifacts that began at the moment of eIF2A discovery. This, in combination with its misfortunate naming (eIF2A is often mixed up with alpha subunit of eIF2, eIF2S1) has generated a widespread confusion among researchers who are not experts in eukaryotic translation initiation. Given this, it is not only justifiable but critical to make independent efforts to clear up this confusion and I very much appreciate the authors' efforts in this regard.

Strengths

The experimental investigation described in this manuscript is thorough, appropriate and convincing.

Weaknesses

No major weaknesses as the authors have improved their presentation.

eLife Assessment

In this valuable study, Roiuk et al combined ribosome profiling and reporter assays to provide compelling evidence that eIF2A does not have a major impact on mRNA translation in HeLa cells. These findings are consistent with several recent publications that disaffirm the previously proposed role of eIF2A in directing protein synthesis under stress. Considering that stress-dependent perturbations in translation play a major role in homeostasis and several pathological states (e.g., cancer and neurological disorders), this work should be of broad interest to researchers studying regulation of gene expression, stress-adaptation, cancer and neurobiology.

Reviewer #1 (Public review):

Summary:

Beyond what is stated in the title of this paper, not much needs to be summarized. eIF2A in HeLa cells promotes translation initiation of neither the main ORFs nor short uORFs under any of the conditions tested.

Strengths:

Very comprehensive, in fact, given the huge amount of purely negative data, an admirably comprehensive and well-executed analysis of the factor of interest.

Weaknesses:

The study is limited to the HeLa cell line, which is now addressed and clearly stated by the authors.

eLife Assessment

This important study demonstrates the significance of incorporating biological constraints in training neural networks to develop models that make accurate predictions under novel conditions. By comparing standard sigmoid recurrent neural networks (RNNs) with biologically constrained RNNs, the manuscript offers compelling evidence that biologically grounded inductive biases enhance generalization to perturbed conditions. This manuscript will appeal to a wide audience in systems and computational neuroscience.

Reviewer #1 (Public review):

I congratulate the authors on this beautiful work.

This manuscript introduces a biologically informed RNN (bioRNN) that predicts the effects of optogenetic perturbations in both synthetic and in vivo datasets. By comparing standard sigmoid RNNs (σRNNs) and bioRNNs, the authors make a compelling case that biologically grounded inductive biases improve generalization to perturbed conditions. This work is innovative, technically strong, and grounded in relevant neuroscience, particularly the pressing need for data-constrained models that generalize causally.

I have some suggestions for improvement, which I present in the order of re-reading the paper.

Major

(1) In line 76, the authors make a very powerful statement: 'σRNN simulation achieves higher similarity with unseen recorded trials before perturbation, but lower than the bioRNN on perturbed trials.' I couldn't find a figure showing this. This might be buried somewhere and, in my opinion, deserves some spotlight - maybe a figure or even inclusion in the abstract.

(2) It's mentioned in the introduction (line 84) and elsewhere (e.g., line 259) that spiking has some advantage, but I don't see any figure supporting this claim. In fact, spiking seems not to matter (Figure 2C, E). Please clarify how spiking improves performance, and if it does not, acknowledge that. Relatedly, in line 246, the authors state that 'spiking is a better metric but not significant' when discussing simulations. Either remove this statement and assume spiking is not relevant, or increase the number of simulations.

(3) The authors prefer the metric of predicting hits over MSE, especially when looking at real data (Figure 3). I would bring the supplementary results into the main figures, as both metrics are very nicely complementary. Relatedly, why not add Pearson correlation or R2, and not just focus on MSE Loss?

(4) I really like the 'forward-looking' experiment in closed loop! But I felt that the relevance of micro perturbations is very unclear in the intro and results. This could be better motivated: why should an experimentalist care about this forward-looking experiment? Why exactly do we care about micro perturbation (e.g., in contrast to non-micro perturbation)? Relatedly, I would try to explain this in the intro without resorting to technical jargon like 'gradients'.

Minor

(1) In the intro, the authors refer to 'the field' twice. Personally, I find this term odd. I would opt for something like 'in neuroscience'.

(2) Line 45: When referring to previous work using data-constrained RNN models, Valente et al. is missing (though it is well cited later when discussing regularization through low-rank constraints).

(3) Line 11: Method should be methods (missing an 's').

(4) In line 250, starting with 'So far', is a strange choice of presentation order. After interpreting the results for other biological ingredients, the authors introduce a new one. I would first introduce all ingredients and then interpret. It's telling that the authors jump back to 2B after discussing 2C.

(5) The black dots in Figure 3E are not explained, or at least I couldn't find an explanation.

Reviewer #2 (Public review):

Sourmpis et al. present a study in which the importance of including certain inductive biases in the fitting of recurrent networks is evaluated with respect to the generalization ability of the networks when exposed to untrained perturbations.

The work proceeds in three stages:<br /> (1) a simple illustration of the problem is made. Two reference (ground-truth) networks with qualitatively different connectivity, but similar observable network dynamics, are constructed, and recurrent networks with varying aspects of design similarity to the reference networks are trained to reproduce the reference dynamics. The activity of these trained networks during untrained perturbations is then compared to the activity of the perturbed reference networks. It is shown that, of the design characteristics that were varied, the enforced sign (Dale's law) and locality (spatial extent) of efference were especially important.<br /> (2) The intuition from the constructed example is then extended to networks that have been trained to reproduce certain aspects of multi-region neural activity recorded from mice during a detection task with a working-memory component. A similar pattern is demonstrated, in which enforcing the sign and locality of efference in the fitted networks has an influence on the ability of the trained networks to predict aspects of neural activity during unseen (untrained) perturbations.<br /> (3) The authors then illustrate the relationship between the gradient of the motor readout of trained networks with respect to the net inputs to the network units, and the sensitivity of the motor readout to small perturbations of the input currents to the units, which (in vivo) could be controlled optogenetically. The paper is concluded with a proposed use for trained networks, in which the models could be analyzed to determine the most sensitive directions of the network and, during online monitoring, inform a targeted optogenetic perturbation to bias behavior.

The authors do not overstate their claims, and in general, I find that I agree with their conclusions. A couple of points to be made:

(1) Some aspects of the methods are unclear. For comparisons between recurrent networks trained from randomly initialized weights, I would expect that many initializations were made for each model variant to be compared, and that the performance characteristics are constructed by aggregating over networks trained from multiple random initializations. I could not tell from the methods whether this was done or how many models were aggregated.

2) It is possible that including perturbation trials in the training sets would improve model performance across conditions, including held-out (untrained) perturbations (for instance, to units that had not been perturbed during training). It could be noted that if perturbations are available, their use may alleviate some of the design decisions that are evaluated here.

eLife Assessment

This useful study attempts to place an ancient maize sample from Bolivia, dated to the end of the Incan empire, in genetic and geographical context. The analyses show that this sample is most closely related to ancient Peruvian maize, but the data are inadequate to determine the direction of dispersal. There are additional deficiencies in the statistical analyses and selection inferences. The topic of the study would appeal to researchers studying maize dispersal and adaptation.

Reviewer #1 (Public review):

Summary:

In this manuscript, the authors describe a good-quality ancient maize genome from 15th-century Bolivia and try to link the genome characteristics to Inca influence. Overall, the manuscript is below the standard in the field. In particular, the geographic origin of the sample and its archaeological context is not well evidenced. While dating of the sample and the authentication of ancient DNA have been evidenced robustly, the downstream genetic analyses do not support the conclusion that genomic changes can be attributed to Inca influence. Furthermore, sections of the manuscript are written incoherently and with logical mistakes. In its current form, this paper is not robust and possibly of very narrow interest.

Strengths:

Technical data related to the maize sample are robust. Radiocarbon dating strongly evidenced the sample age, estimated to be around 1474 AD. Authentication of ancient DNA has been done robustly. Spontaneous C-to-T substitutions, which are present in all ancient DNA, are visible in the reported sample with the expected pattern. Despite a low fraction of C-to-T at the 1st base, this number could be consistent with the cool and dry climate in which the sample was preserved. The distribution of DNA fragment sizes is consistent with expectations for a sample of this age.

Weaknesses:

(1) Archaeological context for the maize sample is weakly supported by speculation about the origin and has unreasonable claims weighing on it. Perhaps those findings would be more convincing if the authors were to present evidence that supports their conclusions: i) a map of all known tombs near La Paz, ii) evidence supporting the stone tomb origins of this assemblage, and iii) evidence supporting non-Inca provenance of the tomb.

(2) Dismissal of the admixture in the reported samples is not evidenced correctly. Population f3 statistic with an outgroup is indeed one of the most robust metrics for sample relatedness; however, it should not be used as a test of admixture. For an admixture test, the population f3 statistic should be used in the form: i) target population, ii) one possible parental population, iii) another possible parental population. This is typically done iteratively with all combinations of possible parental populations. Even in such a form, the population f3 statistic is not very sensitive to admixture in cases of strong genetic drift, and instead population f4 statistic (with an outgroup) is a recommended test for admixture.

(3) The geographic placement of the sample based on genetic data is not robust. To make use of the method correctly, it would be necessary to validate that genetic samples in this region follow the assumption of the 'isolation-by-distance' with dense sampling, which has not been done. Additionally, the authors posit that "This suggests that aBM might not only be genetically related to the archaeological maize from ancient Peru, but also in the possible geographic location." The method used to infer the location is based on pure genetic estimation. The above conclusion is not supported by this method, and it directly contradicts the authors' suggestion that the sample comes from Bolivia.

(4) The conclusion that Ancient Andean maize is genetically similar to European varieties and hence shares a similar evolutionary history is not well supported. The PCA plot in Figure 4 merely represents sample similarity based on two components (jointly responsible for about 20% of the variation explained), and European samples could be very distant based on other components. Indeed, the direct test using the outgroup f3 statistic does not support that European varieties are particularly closely related to ancient Andean maize. Perhaps these are more closely related to Brazil? We do not know, as this has not been measured.

(5) The conclusion that long branches in the phylogenetic tree are due to selection under local adaptation has no evidence. Long branches could be the result of missing data, nucleotide misincorporations, genetic drift, or simply due to the inability of phylogenetic trees to model complex population-level relationships such as admixture or incomplete lineage sorting. Additionally, captions to Figure S3, do not explain colour-coding.

(6) The conclusion that selection detected in aBM sample is due to Inca influence has no support. Firstly, selection signature can be due to environmental or other factors. To disentangle those, the authors would need to generate the data for a large number of samples from similar cultural contexts and from a wide-ranging environmental context, followed by a formal statistical test. Secondly, allele frequency increase can be attributed to selection or demographic processes, and alone is not sufficient evidence for selection. The presented XP-EHH method seems more suitable. Overall, methods used in this paper raise some concerns: i) how accurate are allele-frequency tests of selection when only single individual is used as a proxy for a whole population, ii) the significance threshold has been arbitrary fixed to an absolute number based on other studies, but the standard is to use, for example, top fifth percentile. Finally, linking selection to particular GO terms is not strong evidence, as correlation does not imply causation, and links are unclear anyway.

In sum, this manuscript presents new data that seems to be of high quality, but the analyses are frequently inappropriate and/or over-interpreted.

Reviewer #2 (Public review):

Summary:

The manuscript presents valuable new datasets from two ancient maize seeds that contribute to our growing understanding of the maize evolution and biodiversity landscape in pre-colonial South America. Some of the analyses are robust, but the selection elements are not supported.

Strengths:

The data collection is robust, and the data appear to beof sufficiently high quality to carry out some interesting analytical procedures. The central finding that aBM maize is closely related to maize from the core Inca region is well supported, although the directionality of dispersal is not supported.

Weaknesses:

The selection results are not justified, see examples in the detailed comments below.

(1) The manuscript mentions cultural and natural selection (line 76), but then only gives a couple of examples of selecting for culinary/use traits. There are many examples of selection to tolerate diverse environments that could be relevant for this discussion, if desired.

(2) I would be extremely cautious about interpreting the observations of a Spanish colonizer (lines 95-99) without very significant caveats. Indigenous agriculture and foodways would have been far more nuanced than what could be captured in this context, and the genocidal activities of the Europeans would have impacted food production activities to a degree, and any contemporaneous accounts need to be understood through that lens.

(3) The f3 stats presented in Figure 2 are not set up to test any specific admixture scenarios, so it is unsupported to conclude that the aBM maize is not admixed on this basis (lines 201-202). The original f3 publication (Patterson et al, 2012) describes some scenarios where f3 characteristics associate with admixture, but in general, there are many caveats to this approach, and it's not the ideal tool for admixture testing, compared with e.g., f4 and D (abba-baba) statistics.

(4) I'm a little bit skeptical that the Locator method adds value here, given the small training sample size and the wide geographic spread and genetic diversity of the ancient samples that include Central America. The paper describing that method (Battey et al 2020 eLife) uses much larger datasets, and while the authors do not specifically advise on sample sizes, they caution about small sample size issues. We have already seen that the ancient Peruvian maize has the most shared drift with aBM maize on the basis of the f3 stats, and the Locator analysis seems to just be reiterating that. I would advise against putting any additional weight on the Locator results as far as geographic origins, and personally I would skip this analysis in this case.

(5) The overlap in PCA should not be used to confirm that aBM is authentically ancient, because with proper data handling, PCA placement should be agnostic to modern/ancient status (see lines 224-226). It is somewhat unexpected that the ancient Tehuacan maize (with a major teosinte genomic component) falls near the ancient South American maize, but this could be an artifact of sampling throughout the PCA and the lack of teosinte samples that might attract that individual.

(6) What has been established (lines 250-251) is genetic similarity to the Inca core area, not necessarily the directionality. Might aBM have been part of a cultural region supplying maize to the Inca core region, for example? Without a specific test of dispersal directionality, which I don't think is possible with the data at hand, this is somewhat speculative.

(7) Singleton SNPs are not a typical criterion for identifying selection; this method needs some citations supporting the exact approach and validation against neutral expectations (line 278). Without Datasets S2 and S3, which are not included with this submission, it is difficult to assess this result further. However, it is very unexpected that ~18,000 out of ~49,000 SNPs would be unique to the aBM lineage. This most likely reflects some data artifact (unaccounted damage, paralogs not treated for high coverage, which are extremely prevalent in maize, etc). I'm confused about unique SNPs in this context. How can they be unique to the aBM lineage if the SNPs used overlap the Grzybowski set? The GO results do not include any details of the exact method used or a statistical assessment of the results. It is not clear if the GO terms noted are statistically enriched.

(8) The use of XP-EHH with pseudohaplotype variant calls is not viable (line 293). It is not clear what exact implementation of XP-EHH was used, but this method generally relies on phased or sometimes unphased diploid genotype calls to observe shared haplotypes, and some minimum population size to derive statistical power. No implementation of XP-EHH to my knowledge is appropriate for application to this kind of dataset.

Reviewer #3 (Public review):

Summary:

The authors seek to place archaeological maize samples (2 kernels) from Bolivia into genetic and geographical context and to assess signatures of selection. The kernels were dated to the end of the Incan empire, just prior to European colonization. Genetic data and analyses were used to characterize the distance from other ancient and modern maize samples and to predict the origin of the sample, which was discovered in a tomb near La Paz, Bolivia. Given the conquest of this region by the Incan empire, it is possible that the sample could be genetically similar to populations of maize in Peru, the center of the Incan empire. Signatures of selection in the sample could help reveal various environmental variables and cultural preferences that shaped maize genetic diversity in this region at that time.

Strengths:

The authors have generated substantial genetic data from these archaeological samples and have assembled a data set of published archaeological and modern maize samples that should help to place these samples in context. The samples are dated to an interesting time in the history of South America during a period of expansion of the Incan empire and just prior to European colonization. Much could be learned from even this small set of samples.

Weaknesses:

(1) Sample preparation and sequencing:<br /> Details of the quality of the samples, including the percentage of endogenous DN,A are missing from the methods. The low percentage of mapped reads suggests endogenous DNA was low, and this would be useful to characterize more fully. Morphological assessment of the samples and comparison to morphological data from other maize varieties is also missing. It appears that the two kernels were ground separately and that DNA was isolated separately, but data were ultimately pooled across these genetically distinct individuals for analysis. Pooling would violate assumptions of downstream analysis, which included genetic comparison to single archaeological and modern individuals.

(2) Genetic comparison to other samples:<br /> The authors did not meaningfully address the varying ages of the other archaeological samples and modern maize when comparing the genetic distance of their samples. The archaeological samples were as old as >5000 BP to as young as 70 BP and therefore have experienced varying extents of genetic drift from ancestral allele frequencies. For this reason, age should explicitly be included in their analysis of genetic relatedness.

(3) Assessment of selection in their ancient Bolivian sample:<br /> This analysis relied on the identification of alleles that were unique to the ancient sample and inferred selection based on a large number of unique SNPs in two genes related to internode length. This could be a technical artifact due to poor alignment of sequence data, evidence supporting pseudogenization, or within an expected range of genetic differentiation based on population structure and the age of the samples. More rigor is needed to indicate that these genetic patterns are consistent with selection. This analysis may also be affected by the pooling of the Bolivian archaeological samples.

(4) Evidence of selection in modern vs. ancient maize: In this analysis, samples were pooled into modern and ancient samples and compared using the XP-EHH statistic. One gene related to ovule development was identified as being targeted by selection, likely during modern improvement. Once again, ancient samples span many millennia and both South, Central, and North America. These, and the modern samples included, do not represent meaningfully cohesive populations, likely explaining the extremely small number of loci differentiating the groups. This analysis is also complicated by the pooling of the Bolivian archaeological samples.

eLife Assessment

This important study provides a novel approach for delineating subcortical-cortical white matter bundles. The authors provide convincing evidence by harnessing state-of-the-art methods and cross-species data. Together, this effort will be of interest to scientists across multiple subfields.

Reviewer #1 (Public review):

Summary:

The authors note that it is challenging to perform diffusion MRI tractography consistently in both humans and macaques, particularly when deep subcortical structures are involved. The scientific advance described in this paper is effectively an update to the tracts that the XTRACT software supports. The claims of robustness are based on a very small selection of subjects from a very atypical dMRI acquisition (n=50 from HCP-Adult) and an even smaller selection of subjects from a more typical study (n=10 from ON-Harmony).

Strengths:

The changes to XTRACT are soundly motivated in theory (based on anatomical tracer studies) and practice (changes in seeding/masking for tractography), and I think the value added by these changes to XTRACT should be shared with the field. While other bundle segmentation software typically includes these types of changes in release notes, I think papers are more appropriate.

Weaknesses:

The demonstration of the new tracts does not include a large number of carefully selected scans and is only compared to the prior methods in XTRACT. The small n and limited statistical comparisons are insufficient to claim that they are better than an alternative. Qualitatively, this method looks sound.

Subject selection at each stage is unclear in this manuscript. On page 5 the data are described as "Using dMRI data from the macaque (𝑁 = 6) and human brain (𝑁 = 50)". Were the 50 HCP subjects selected to cover a range of noise levels or subject head motion? Figure 4 describes 72 pairs for each of monozygotic, dizygotic, non-twin siblings, and unrelated pairs - are these treated separately? Similarly, NH had 10 subjects, but each was scanned 5 times. How was this represented in the sample construction?

In the paper, the authors state "the mean agreement between HCP and NH reconstructions was lower for the new tracts, compared to the original protocols (𝑝 < 10^−10). This was due to occasionally reconstructing a sparser path distribution, i.e., slightly higher false negative rate," - how can we know this is a false negative rate without knowing the ground truth?

Reviewer #2 (Public review):

Summary:

In this article, Assimopoulos et al. expand the FSL-XTRACT software to include new protocols for identifying cortical-subcortical tracts with diffusion MRI, with a focus on tracts connecting to the amygdala and striatum. They show that the amygdalofugal pathway and divisions of the striatal bundle/external capsule can be successfully reconstructed in both macaques and humans while preserving large-scale topographic features previously defined in tract tracing studies. The authors set out to create an automated subcortical tractography protocol, and they accomplished this for a subset of specific subcortical connections for users of the FSL ecosystem.

Strengths:

A main strength of the current study is the translation of established anatomical knowledge to a tractography protocol for delineating cortical-subcortical tracts that are difficult to reconstruct. Diffusion MRI-based tractography is highly prone to false positives; thus, constraining tractography outputs by known anatomical priors is important. Key additional strengths include 1) the creation of a protocol that can be applied to both macaque and human data; 2) demonstration that the protocol can be applied to be high quality data (3 shells, > 250 directions, 1.25 mm isotropic, 55 minutes) and lower quality data (2 shells, 100 directions, 2 mm isotropic, 6.5 minutes); and 3) validation that the anatomy of cortical-subcortical tracts derived from the new method are more similar in monozygotic twins than in siblings and unrelated individuals.

Weaknesses:

Although this work validates the general organizational location and topographic organization of tractography-derived cortical-subcortical tracts against prior tract tracing studies (a clear strength), the validation is purely visual and thus only qualitative. Furthermore, it is difficult to assess how the current XTRACT method may compare to currently available tractography approaches to delineating similar cortical-subcortical connections. Finally, it appears that the cortical-subcortical tractography protocols developed here can only be used via FSL-XTRACT (yet not with other dMRI software), somewhat limiting the overall accessibility of the method.

Overall Appraisal:

This new method will accelerate research on anatomically validated cortical-subcortical white matter pathways. The work has utility for diffusion MRI researchers across fields.

eLife Assessment

This valuable work presents how PRDM16 plays a critical role during colloid plexus development, through regulating BMP signaling. Solid evidence supports the context-dependent gene regulatory mechanisms both in vivo and in vitro. The work will be of broad interest to researchers working on growth factor signaling mechanisms and vertebrate development.

Reviewer #1 (Public review):

Summary:<br /> This manuscript describes the role of PRDM16 in modulating BMP response during choroid plexus (ChP) development. The authors combine PRDM16 knockout mice and cultured PRDM16 KO primary neural stem cells (NSCs) to determine the interactions between BMP signaling and PRDM16 in ChP differentiation.<br /> They show PRDM16 KO affects ChP development in vivo and BMP4 response in vitro. They determine genes regulated by BMP and PRDM16 by ChIP-seq or CUT&TAG for PRDM16, pSMAD1/5/8, and SMAD4. They then measure gene activity in primary NSCs through H3K4me3 and find more genes are corepressed than coactivated by BMP signaling and PRDM16 and focus on the 31 genes found to be co-repressed by BMP and PRDM16. Wnt7b is in this set and the authors then provide evidence that PRDM16 and BMP signaling together repress Wnt activity in the developing choroid plexus.

Strengths:<br /> Understanding context-dependent response to cell signals during development is an important problem. The authors use a powerful combination of in vivo and in vitro systems to dissect how PRDM16 may modulate BMP response in early brain development.

Main weakness of the experimental setup:<br /> (1) Because the authors state that primary NSCs cultured in vitro lose endogenous Prdm16 expression, they drive expression by a constitutive promoter. However, this means the expression levels is very different from endogenous levels (as explicitly shown in Supp. Fig. 2B) and the effect of many transcription factors is strongly dose-dependent, likely creating differences between the PRDM16-dependent transcriptional response in the in vitro system and in vivo. Although the authors combine in vitro and in vivo evidence on the role of PRDM16 as a co-factor for MBP signaling and verified that BMP induces quiescence in their NSC model in a PRDM16-dependent manner, this experimental setup remains a weakness and likely affects the results of the various genomics experiments.

Other experimental weaknesses that make the evidence less convincing:

(1) It seems that the authors compare Prdm16_KO cells to Prdm16 WT cells overexpressing flag_Prdm16. Aside from the possible expression of endogenous Prdm16, other cell differences may have arisen between these cell lines. A properly controlled experiment would compare Prdm16_KO ctrl (possibly infected with a control vector without Prdm16) to Prdm16_KO_E (i.e. the Prdm16_KO cells with and without Prdm16 overexpression.) The authors acknowledged this problem in their rebuttal, stating that they were unable to overexpress PRDM16 in KO cells.

(2) The authors show in Fig.2E that Ttr is not upregulated by BMP4 in PRDM16_KO NSCs. This appears inconsistent with the presence of Ttr expression in the PRDM16_KO brain in Fig.1C. The authors explained in their rebuttal that the Ttr protein levels are not detectable in the NSCs with antibodies but the effect is still visible at the level of mRNA. The dramatic difference in protein expression is curious.

Reviewer #2 (Public review):

The authors have revised their manuscript in response to reviewer feedback, incorporating several modifications to improve clarity and provide additional supporting information. To address concerns about confusing terminology, they have standardized the reference to PRDM16 overexpressing cells as Prdm16_OE, clarifying its expression from a constitutive promoter. They also revised the text to resolve seemingly contradictory statements about ChP development in the mutant. New bioinformatic analysis comparing PRDM16 binding in E12.5 ChP cells to co-repressed versus BMP-only-repressed genes has been performed and included in Supplementary Figure 5C, providing a statistical assessment of PRDM16's regulatory role on co-repressed genes. Several figures were updated, including adding an illustration of the Prdm16 cGT allele to Figure 1B, providing a zoomed-in inset for Figure 1E, and including individual channels for Wnt2b and marking boundaries in Figure 7A. Full-view images and examples of spot segmentation for SCRINSHOT analysis are now available in a new supplementary figure, and the presentation of RT-qPCR data in Supplementary Figure 2B was improved by using separate graphs for overexpression samples to avoid a broken Y-axis. Furthermore, the authors have added more references to introductory statements, annotated structures like the ChP, CH, and fourth ventricle in figures, and clarified that the beta-Gal signal was used as a marker for mutant ChP cells in Figure 1D. Finally, the manuscript now includes a discussion of the recently published, related study by Hurwitz et al. (2023) in the discussion section, highlighting similarities and differences. Overall, the authors have satisfactorily addressed the reviewers' comments.

Reviewer #3 (Public review):

Summary:<br /> Bone morphogenetic protein (BMP) signaling instructs multiple processes during development including cell proliferation and differentiation. The authors set out to understand the role of PRDM16 in these various functions of BMP signaling. They find that PRDM16 and BMP co-operate to repress stem cell proliferation by regulating the genomic distribution of BMP pathway transcription factors. They additionally show that PRDM16 impacts choroid plexus epithelial cell specification. The authors provide evidence for a regulatory circuit (constituting of BMP, PRDM16 and Wnt) that influences stem cell proliferation/differentiation.

Strengths:<br /> I find the topics studied by the authors in this study of general interest to the field, the experiments well-controlled and the analysis in the paper sound. I have no major scientific concerns.

Weaknesses:<br /> I have some minor recommendations which will help improve the paper (regarding the discussion).

Comments on revised version:

The authors have addressed my concerns in the revised version of the manuscript.

Author response:

The following is the authors’ response to the original reviews

Reviewer #1 (Public review):

Summary:

This manuscript describes the role of PRDM16 in modulating BMP response during choroid plexus (ChP) development. The authors combine PRDM16 knockout mice and cultured PRDM16 KO primary neural stem cells (NSCs) to determine the interactions between BMP signaling and PRDM16 in ChP differentiation.

They show PRDM16 KO affects ChP development in vivo and BMP4 response in vitro. They determine genes regulated by BMP and PRDM16 by ChIP-seq or CUT&TAG for PRDM16, pSMAD1/5/8, and SMAD4. They then measure gene activity in primary NSCs through H3K4me3 and find more genes are co-repressed than co-activated by BMP signaling and PRDM16. They focus on the 31 genes found to be co-repressed by BMP and PRDM16. Wnt7b is in this set and the authors then provide evidence that PRDM16 and BMP signaling together repress Wnt activity in the developing choroid plexus.

Strengths:

Understanding context-dependent responses to cell signals during development is an important problem. The authors use a powerful combination of in vivo and in vitro systems to dissect how PRDM16 may modulate BMP response in early brain development.

We thank the reviewer for the thoughtful summary and positive feedback. We appreciate the recognition of our integrative in vivo and in vitro approach. We're glad the reviewer found our findings on context-dependent gene regulation and developmental signaling valuable.

Main weaknesses of the experimental setup:

(1) Because the authors state that primary NSCs cultured in vitro lose endogenous Prdm16 expression, they drive expression by a constitutive promoter. However, this means the expression levels are very different from endogenous levels (as explicitly shown in Supplementary Figure 2B) and the effect of many transcription factors is strongly dose-dependent, likely creating differences between the PRDM16-dependent transcriptional response in the in vitro system and in vivo.

We acknowledge that our in vitro experiments may not ideally replicate the in vivo situation, a common limitation of such experiments, our primary aim was to explore the molecular relationship between PRDM16 and BMP signaling in gene regulation. Such molecular investigations are challenging to conduct using in vivo tissues. In vitro NSCs treated with BMP4 has been used a model to investigate NSC proliferation and quiescence, drawing on previous studies (e.g., Helena Mira, 2010; Marlen Knobloch, 2017). Crucially, to ensure the relevance of our in vitro findings to the in vivo context, we confirmed that cultured cells could indeed be induced into quiescence by BMP4, and this induction necessitated the presence of PRDM16. Furthermore, upon identifying target genes co-regulated by PRDM16 and SMADs, we validated PRDM16's regulatory role on a subset of these genes in the developing Choroid Plexus (ChP) (Fig. 7 and Suppl.Fig7-8). Only by combining evidence from both in vitro and in vivo experiments could we confidently conclude that PRDM16 serves as an essential co-factor for BMP signaling in restricting NSC proliferation.

(2) It seems that the authors compare Prdm16_KO cells to Prdm16 WT cells overexpressing flag_Prdm16. Aside from the possible expression of endogenous Prdm16, other cell differences may have arisen between these cell lines. A properly controlled experiment would compare Prdm16_KO ctrl (possibly infected with a control vector without Prdm16) to Prdm16_KO_E (i.e. the Prdm16_KO cells with and without Prdm16 overexpression.)

We agree that Prdm16 KO cells carrying the Prdm16-expressing vector would be a good comparison with those with KO_vector. However, despite more than 10 attempts with various optimization conditions, we were unable to establish a viable cell line after infecting Prdm16 KO cells with the Prdm16-expressing vector. The overall survival rate for primary NSCs after viral infection is low, and we observed that KO cells were particularly sensitive to infection treatment when the viral vector was large (the Prdm16 ORF is more than 3kb).

As an alternative oo assess vector effects, we instead included two other control cell lines, wt and KO cells infected with the 3xNLS_Flag-tag viral vector, and presented the results in supplementary Fig 2.  When we compared the responses of the four lines — wt, KO, wt infected with the Flag vector, KO infected with the Flag vector — to the addition and removal of BMP4, we confirmed that the viral infection itself has no significant impacts on the responses of these cells to these treatments regarding changes in cell proliferation and Ttr induction.

Given that wt cells and the KO cells, with or without viral backbone infection behave quite similarly in terms of cell proliferation, we speculate that even if we were successful in obtaining a cell line with Prdm16-expressing vector in the KO cells, it may not exhibit substantial differences compared to wt cells infected with Prdm16-expressing vector.

Other experimental weaknesses that make the evidence less convincing:

(1) The authors show in Figure 2E that Ttr is not upregulated by BMP4 in PRDM16_KO NSCs. Does this appear inconsistent with the presence of Ttr expression in the PRDM16_KO brain in Figure1C?

The reviwer’s point is that there was no significant increase in Ttr expression in Prdm16_KO cells after BMP4 treatment (Fig. 2E), but there remained residule Ttr mRNA signals in the Prdm16 mutant ChP (Fig. 1C). We think the difference lies in the measuable level of Ttr expression between that induced by BMP4 in NSC culture and that in the ChP. This is based on our immunostaining expreriment in which we tried to detect Ttr using a Ttr antibody. This antibody could not detect the Ttr protein in BMP4-treated Prdm16_expressing NSCs but clearly showed Ttr signal in the wt ChP. This means that although Ttr expression can be significantly increased by BMP4 in vitro to a level measurable by RT-qPCR, its absolute quantity even in the Prdm16_expressing condition is much lower compared to that in vivo. Our results in Fig 1C and Fig 2E, as well as Fig 7B, all consistently showed that Prdm16 depletion significantly reduced Ttr expression in in vitro and in vivo.

(2) Figure 3: The authors use H3K4me3 to measure gene activity. This is however, very indirect, with bulk RNA-seq providing the most direct readout and polymerase binding (ChIP-seq) another more direct readout. Transcription can be regulated without expected changes in histone methylation, see e.g. papers from Josh Brickman. They verify their H3K4me3 predictions with qPCR for a select number of genes, all related to the kinetochore, but it is not clear why these genes were picked, and one could worry whether these are representative.

H3K4me3 has widely been used as an indicator of active transcription and is a mark for cell identity genes. And it has been demonstrated that H3K4me3 has a direct function in regulating transciption at the step of RNApolII pausing release. As stated in the text, there are advantages and disadvantages of using H3K4me3 compared to using RNA-seq. RNA-seq profiles all gene products, which are affected by transcription and RNA stability and turnover. In contrast, H3K4me3 levels at gene promoter reflects transcriptional activity. In our case, we aimed to identify differential gene expression between proliferation and quiescence states. The transition between these two states is fast and dynamic. RNA-seq may not be able to identify functionally relevant genes but more likely produces false positive and negative results. Therefore, we chose H3K4me3 profiling.

We agree that transcription may change without histone methylation changes. This may cause an under-estimation of the number of changed genes between the conditions. 

We validated 7 out of 31 genes (Wnt7b, Id3, Mybl2, Spc24, Spc25, Ndc80 and Nuf2). We chose these genes based on two critira: 1) their function is implicated in cell proliferation and cell-cycle regulation based on gene ontology analysis; 2) their gene products are detectable in the developing ChP based on the scRNA-seq data. Three of these genes (Wnt7b, Id3, Mybl2) are not related to the kinetochore. We now clarify this description in the revised text.

(3) Line 256: The overlap of 31 genes between 184 BMP-repressed genes and 240 PRDM16-repressed genes seems quite small.

This result indicates that in addition to co-repressing cell-cycle genes, BMP and PRDM16 have independent fucntions. For example, it was reported that BMP regulates neuronal and astrocyte differentiation (Katada, S. 2021), while our previous work demonstrated that Prdm16 controls temporal identity of NSCs (He, L. 2021).

(4) The Wnt7b H3K4me3 track in Fig. 3G is not discussed in the text but it shows H3K4me3 high in _KO and low in _E regardless of BMP4. This seems to contradict the heatmap of H3K4me3 in Figure 3E which shows H3K4me3 high in _E no BMP4 and low in _E BMP4 while omitting _KO no BMP4. Meanwhile CDKN1A, the other gene shown in 3G, is missing from 3E.

The track in Fig 3G shows the absolute signal of H3K4me3 after mapping the sequencing reads to the genome and normaliz them to library size. Compare the signal in Prdm16_E with BMP4 and that in Prdm16_E without BMP4, the one with BMP4 has a lower peak. The same trend can be seen for the pair of Prdm16_KO cells with or without BMP4.  The heatmap in Fig. 3E shows the relative level of H3K4me3 in three conditions. The Prdm16_E cells with BMP4 has the lowest level, while the other two conditions (Prdm16_KO with BMP4 and Prdm16_E without BMP4) display higher levels. These two graphs show a consistent trend of H3K4me3 changes at the Wnt7b promoter across these conditions. Figure 3E only includes genes that are co-repressed by PRDM16 and BMP. CDKN1A’s H3K4me3 signals are consistent between the conditions, and thus it is not a PRDM16- or BMP-regulated gene. We use it as a negative control. 

(5) The authors use PRDM16 CUT&TAG on dissected dorsal midline tissues to determine if their 31 identified PRDM16-BMP4 co-repressed genes are regulated directly by PRDM16 in vivo. By manual inspection, they find that "most" of these show a PRDM16 peak. How many is most? If using the same parameters for determining peaks, how many genes in an appropriately chosen negative control set of genes would show peaks? Can the authors rigorously establish the statistical significance of this observation? And why wasn't the same experiment performed on the NSCs in which the other experiments are done so one can directly compare the results? Instead, as far as I could tell, there is only ChIP-qPCR for two genes in NSCs in Supplementary Figure 4D.

In our text, we indicated the genes containing PRDM16 binding peaks in the figures and described them as “Text in black in Fig. 6A and Supplementary Fig. 5A”. We will add the precise number “25 of these genes” in the main text to clarify it. We used BMP-only repressed 184-31 =153 genes (excluding PRDM16-BMP4 co-repressed) as a negative control set of genes. By computationally determine the nearest TSS to a PRDM16 peak, we identified 24/31 co-repressed genes and 84/153 BMP-only-repressed genes, containing PRDM16 peaks in the E12.5 ChP data. Fisher’s Exact Test comparing the proportions yields the P-value = 0.015.

We are confused with the second part of the comment “And why wasn't the same experiment performed on the NSCs in which the other experiments are done so one can directly compare the results? Instead, as far as I could tell, there is only ChIP-qPCR for two genes in NSCs in Supplementary Figure 4D.” If the reviewer meant why we didn’t sequence the material from sequential-ChIP or validate more taget genes, the reason is the limitation of the material. Sequential ChIP requires a large quantity of the antibodies, and yields little material barely sufficient for a few qPCR after the second round of IP. This yielded amount was far below the minimum required for library construction. The PRDM16 antibody was a gift, and the quantity we have was very limited. We made a lot of efforts to optimize all available commercial antibodies in ChIP and Cut&Tag, but none of them worked in these assays.

(6) In comparing RNA in situ between WT and PRDM16 KO in Figure 7, the authors state they use the Wnt2b signal to identify the border between CH and neocortex. However, the Wnt2b signal is shown in grey and it is impossible for this reviewer to see clear Wnt2b expression or where the boundaries are in Figure 7A. The authors also do not show where they placed the boundaries in their analysis. Furthermore, Figure 7B only shows insets for one of the regions being compared making it difficult to see differences from the other region. Finally, the authors do not show an example of their spot segmentation to judge whether their spot counting is reliable. Overall, this makes it difficult to judge whether the quantification in Figure 7C can be trusted.

In the revised manuscript we have included an individal channel of Wnt2b and mark the boundaries. We also provide full-view images and examples of spot segmentation in the new supplementary figure 8. 

(7) The correlation between mKi67 and Axin2 in Figure 7 is interesting but does not convincingly show that Wnt downstream of PRDM16 and BMP is responsible for the increased proliferation in PRDM16 mutants.

We agree that this result (the correlation between mKi67 and Axin2) alone only suggests that Wnt signaling is related to the proliferation defect in the Prdm16 mutant, and does not necessarily mean that Wnt is downstream of PRDM16 and BMP. Our concolusion is backed up by two additional lines of evidences:  the Cut&Tag data in which PRDM16 binds to regulatory regions of Wnt7b and Wnt3a; BMP and PRDM16 co-repress Wnt7b in vitro.

An ideal result is that down-regulating Wnt signaling in Prdm16 mutant can rescue Prdm16 mutant phenotype. Such an experiment is technically challenging. Wnt plays diverse and essential roles in NSC regulation, and one would need to use a celltype-and stage-specific tool to down-regulate Wnt in the background of Prdm16 mutation. Moreover, Wnt genes are not the only targets regulated by PRDM16 in these cells, and downregulating Wnt may not be sufficient to rescue the phenotype. 

Weaknesses of the presentation:

Overall, the manuscript is not easy to read. This can cause confusion.

We have revised the text to improve clarity.

Reviewer #1 (Recommendations for the authors):

(1) Overall, the manuscript is not easy to read. Here are some causes of confusion for which the presentation could be cleaned up:

We are grateful for the reviewer’s suggestion. In the revised manuscript, we have made efforts to improve the clarity of the text.

(a) Part of the first section is confusing in that some statements seem contradictory, in particular:

"there is no overall patterning defect of ChP and CH in the Prdm16 mutant" (line 125)

"Prdm16 depletion disrupted the transition from neural progenitors into ChP epithelia" (line 144)

It would be helpful if the authors could reformulate this more clearly.

We modified the text to clarify that while the BMP-patterned domain is not affected, the transition of NSCs into ChP epithelial cells is compromised in the Prdm16 mutant.

(b) Flag_PRDM16, PRDM16_expressing, PRDM16_E, PRDM16 OE all seem to refer to the same PRDM16 overexpressing cells, which is very confusing. The authors should use consistent naming. Moreover, it would be good if they renamed these all to PRDM16_OE to indicate expression is not endogenous but driven by a constitutive promoter.

We appreciate the comment and agree that the use of multiple terms to refer to the same PRDM16-overexpressing condition was confusing. Our original intention in using Prdm16_E was to distinguish cells expressing PRDM16 from the two other groups: wild-type cells and Prdm16_KO cells, which both lack PRDM16 protein expression. However, we acknowledge that Prdm16_E could be misinterpreted as indicating expression from the endogenous Prdm16 promoter. To avoid this confusion and ensure consistency, we have now standardized the terminology and refer to this condition as Prdm16_OE, indicating Flag-tagged PRDM16 expression driven by a constitutive promoter.

(c) Line 179 states "generated a cell line by infecting Prdm16_KO cells with the same viral vector, expressing 3xNSL_Flag". Do the authors mean 3xNLS_Flag_Prdm16, so these are the Prdm16_KO_E cells by the notation suggested above? Or is this a control vector with Flag only? The following paragraph refers to Supplementary Figure 2C-F where the same construct is called KO_CDH, suggesting this was an empty CDH vector, without Flag, or Prdm16. This is confusing.

We appreciate the reviewer’s careful reading and helpful comment. We acknowledge the confusion caused by the inconsistent terminology. To clarify: in line 179, we intended to describe an attempt to generate a Prdm16_KO cell line expressing 3xNLS_Flag_Prdm16, not a control vector with Flag only. However, despite repeated attempts, we were unable to establish this line due to low viral efficiency and the vulnerability of Prdm16_KO cells to infection with the large construct. Therefore, these cells were not included in the subsequent analyses.

The term KO_CDH refers to Prdm16_KO cells infected with the empty CDH control vector, which lacks both Flag and Prdm16. This is the line used in the experiments shown in Supplementary Fig. 2C–F. We have revised the text throughout the manuscript to ensure consistent use of terminology and to avoid this confusion.

(2) The introductory statements on lines 53-54 could use more references.

Thanks for the suggestion. We have now included more references.

(3) It would be helpful if all structures described in the introduction and first section were annotated in Figure 1, or otherwise, if a cartoon were included. For example, the cortical hem, and fourth ventricle.

Thanks for the suggestion. We have now indicated the structures, ChP, CH and the fourth ventricle, in the images in Figure 1 and Supplementary Figure 1.

(4) In line 115, "as previously shown.." - to keep the paper self-contained a figure illustrating the genetics of the KO allele would be helpful.

Thanks for the suggestion. We have now included an illustration of the Prdm16 cGT allele in Figure 1B.

(5) In Figure 1D as costain for a ChP marker would be helpful because it is hard to identify morphologically in the Prdm16 KO.

Appoligize for the unclarity. The KO allele contains a b-geo reporter driven by Prdm16 endogenous promoter. The samples were co-stained for EdU, b-Gal and DAPI. To distingquish the ChP domain from the CH, we used the presence of b b-Gal as a marker. We indicated this in the figure legend, but now have also clarified this in the revised text.

(6) The details in Figure 1E are hard to see, a zoomed-in inset would help.

A zoomed-in inset is now included in the figure.

(7) Supplementary Figure 2A does not convincingly show that PRDM16 protein is undetectable since endogenous expression may be very low compared to the overexpression PRDM16_E cells so if the contrast is scaled together it could appear black like the KO.

We appreciate the reviewer’s point and have carefully considered this concern. We concluded that PRDM16 protein is effectively undetectable in cultured wild-type NSCs based on direct comparison with brain tissue. Both cultured NSCs and brain sections were processed under similar immunostaining and imaging conditions. While PRDM16 showed robust and specific nuclear localization in embryonic brain sections (Fig. 1B and Supplementary Fig. 1A), only a small subset of cultured NSCs exhibited PRDM16 signal, primarily in the cytoplasm (middle panel of Fig. 2A). This stark contrast supports our conclusion that endogenous PRDM16 protein is either absent or significantly downregulated in vitro. Because of this limitation, we turned to over-expressing Prdm16 in NSC culture using a constitutive promoter. 

(9) Line 182 "Following the washout step" - no such step had been described, maybe replace by "After washout of BMP".

Yes, we have revised the text.

(8) Line 214: "indicating a modest level" - what defines modest? Compared to what? Why is a few thousand moderate rather than low? Does it go to zero with inhibitors for pathways?

Here a modest level means a lower level than to that after adding BMP4. To clarify this, we revised the description to “indicating endogenous levels of …”

(9) The way qPCR data are displayed makes it difficult to appreciate the magnitude of changes, e.g. in Supplementary Figure 2B where a gap is introduced on the scale. Displaying log fold change / relative CT values would be more informative.

We used a segmented Y-axis in Supplementary Figure 2B because the Prdm16 overexpression samples exhibited much higher experssion levels compared to other conditions. In response to this suggestion, we explored alternative ways to present the result, including ploting log-transformed values and log fold changes. However, these methods did not enhance the clarity of the differences – in fact, log scaling made the magnitude of change appear less apparent. To address this, we now present the overexpression samples in a separate graph, thereby eliminating the need for a broken Y-axis and improving the overall readability of the data.

(10) Writing out "3 days" instead of 3D in Figure 2A would improve clarity. It would be good if the used time interval is repeated in other figures throughout the paper so it is still clear the comparison is between 0 and 3 days.

We have changed “3D” to “3 days”. All BMP4 treatments in this study were 3 days.

(11) Line 290: "we found that over 50% of SMAD4 and pSMAD1/5/8 binding peaks were consistent in Prdm16_E and Prdm16_KO cells, indicating that deletion of Prdm16 does not affect the general genomic binding ability of these proteins" - this only makes sense to state with appropriate controls because 50% seems like a big difference, what is the sample to sample variability for the same condition? Moreover, the next paragraph seems to contradict this, ending with "This result suggests that SMAD binding to these sites depends on PRDM16". The authors should probably clarify the writing.

We appreciate the reviwer’s comment and agree that clarification was needed. Our point was that SMAD4 and pSMAD1/5/8 retain the ability to bind DNA broadly in the Prdm16 KO cells, with more than half of the original binding sites still occupied. This suggests that deletion of Prdm16 does not globally impair SMAD genomic binding. Howerever, our primary interest lies in the subset of sites that show differential by SMAD binding between wt and Prdm16 KO conditions, as thse are likely to be PRDM16-dependent. 

In the following paragraph, we focused specifically on describing SMAD and PRDM16 co-bound sites. At these loci, SMAD4 and pSMAD1/5/8 showed reduced enrichment in the absence of PRDM16, suggesting PRDM16 facilitates SMAD binding at these particular regions. We have revised the text in the manuscript to more clearly distinguish between global SMAD binding and PRDM16-dependent sites.

(12) Much more convincing than ChIP-qPCR for c-FOS for two loci in Figures 5F-G would be a global analysis of c-FOS ChIP-seq data.

We agree that a global c-FOS ChIP-seq analysis would provide a more comprehensive view of c-FOS binding patterns. However, the primary focus of this study is the interaction between BMP signaling and PRDM16. The enrichment of AP-1 motifs at ectopic SMAD4 binding sites was an unexpected finding, which we validated using c-FOS ChIP-qPCR at selected loci. While a genome-wide analysis would be valuable, it falls beyond the current scope. We agree that future studies exploring the interplay among SMAD4/pSMAD, PRDM16, and AP-1 will be important and informative.

(13) Figure 6A is hard to read. A heatmap would make it much easier to see differences in expression. Furthermore, if the point is to see the difference between ChP and CH, why not combine the different subclusters belonging to those structures? Finally, why are there 28 genes total when it is said the authors are evaluating a list of 31 genes and also displaying 6 genes that are not expressed (so the difference isn't that unexpressed genes are omitted)?

For the scRNA-seq data, we chose violin plots because they display both gene expression levels and the number of cells that express each gene. However, we agree that the labels in Figure 6A were too small and difficult to read. We have revised the figure by increasing the font size and moved genes with low expression to  Supplementary Figure 5A. Figure 6A includes 17 more highly expressed genes together with three markers, and  Supplementary Figure 5A contains 13 lowly expressed genes. One gene Mrtfb is missing in the scRNA-seq data and thus not included. We have revised the description of the result in the main text and figure legends.

Reviewer #2 (Public review):

Summary:

This article investigates the role of PRDM16 in regulating cell proliferation and differentiation during choroid plexus (ChP) development in mice. The study finds that PRDM16 acts as a corepressor in the BMP signaling pathway, which is crucial for ChP formation.

The key findings of the study are:

(1) PRDM16 promotes cell cycle exit in neural epithelial cells at the ChP primordium.

(2) PRDM16 and BMP signaling work together to induce neural stem cell (NSC) quiescence in vitro.

(3) BMP signaling and PRDM16 cooperatively repress proliferation genes.

(4) PRDM16 assists genomic binding of SMAD4 and pSMAD1/5/8.

(5) Genes co-regulated by SMADs and PRDM16 in NSCs are repressed in the developing ChP.

(6) PRDM16 represses Wnt7b and Wnt activity in the developing ChP.

(7) Levels of Wnt activity correlate with cell proliferation in the developing ChP and CH.

In summary, this study identifies PRDM16 as a key regulator of the balance between BMP and Wnt signaling during ChP development. PRDM16 facilitates the repressive function of BMP signaling on cell proliferation while simultaneously suppressing Wnt signaling. This interplay between signaling pathways and PRDM16 is essential for the proper specification and differentiation of ChP epithelial cells. This study provides new insights into the molecular mechanisms governing ChP development and may have implications for understanding the pathogenesis of ChP tumors and other related diseases.

Strengths:

(1) Combining in vitro and in vivo experiments to provide a comprehensive understanding of PRDM16 function in ChP development.

(2) Uses of a variety of techniques, including immunostaining, RNA in situ hybridization, RT-qPCR, CUT&Tag, ChIP-seq, and SCRINSHOT.

(3) Identifying a novel role for PRDM16 in regulating the balance between BMP and Wnt signaling.

(4) Providing a mechanistic explanation for how PRDM16 enhances the repressive function of BMP signaling. The identification of SMAD palindromic motifs as preferred binding sites for the SMAD/PRDM16 complex suggests a specific mechanism for PRDM16-mediated gene repression.

(5) Highlighting the potential clinical relevance of PRDM16 in the context of ChP tumors and other related diseases. By demonstrating the crucial role of PRDM16 in controlling ChP development, the study suggests that dysregulation of PRDM16 may contribute to the pathogenesis of these conditions.

We thank the reviewer for the thorough and thoughtful summary of our study. We’re glad the key findings and significance of our work were clearly conveyed, particularly regarding the role of PRDM16 in coordinating BMP and Wnt signaling during ChP development. We also appreciate the recognition of our integrated approach and the potential implications for understanding ChP-related diseases.

Weaknesses:

(1) Limited investigation of the mechanism controlling PRDM16 protein stability and nuclear localization in vivo. The study observed that PRDM16 protein became nearly undetectable in NSCs cultured in vitro, despite high mRNA levels. While the authors speculate that post-translational modifications might regulate PRDM16 in NSCs similar to brown adipocytes, further investigation is needed to confirm this and understand the precise mechanism controlling PRDM16 protein levels in vivo.

While mechansims controlling PRDM16 protein stability and nuclear localization in the developing brain are interesting, the scope of this paper is revealing the function of PRDM16 in the choroid plexus and its interaction with BMP signaling. We will be happy to pursuit this direction in our next study.

(2) Reliance on overexpression of PRDM16 in NSC cultures. To study PRDM16 function in vitro, the authors used a lentiviral construct to constitutively express PRDM16 in NSCs. While this approach allowed them to overcome the issue of low PRDM16 protein levels in vitro, it is important to consider that overexpressing PRDM16 may not fully recapitulate its physiological role in regulating gene expression and cell behavior.

As stated above, we acknowledge that findings from cultured NSCs may not directly apply to ChP cells in vivo. We are cautious with our statements. The cell culture work was aimed to identify potential mechanisms by which PRDM16 and SMADs interact to regulate gene expression and target genes co-regulated by these factors. We expect that not all targets from cell culture are regulated by PRDM16 and SMADs in the ChP, so we validated expression changes of several target genes in the developing ChP and now included the new data in Fig. 7 and Supplementary Fig. 7. Out of the 31 genes identified from cultured cells, four cell cycle regulators including Wnt7b, Id3, Spc24/25/nuf2 and Mybl2, showed de-repression in Prdm16 mutant ChP. These genes can be relevant downstream genes in the ChP, and other target genes may be cortical NSC-specific or less dependent on Prdm16 in vivo.

(3) Lack of direct evidence for AP1 as the co-factor responsible for SMAD relocation in the absence of PRDM16. While the study identified the AP1 motif as enriched in SMAD binding sites in Prdm16 knockout cells, they only provided ChIP-qPCR validation for c-FOS binding at two specific loci (Wnt7b and Id3). Further investigation is needed to confirm the direct interaction between AP1 and SMAD proteins in the absence of PRDM16 and to rule out other potential co-factors.

We agree that the finding of the AP1 motif enriched at the PRDM16 and SMAD co-binding regions in Prdm16 KO cells can only indirectly suggest AP1 as a co-factor for SMAD relocation. That’s why we used ChIP-qPCR to examine the presence of C-fos at these sites. Although we only validated two targets, the result confirms that C-fos binds to the sites only in the Prdm16 KO cells but not Prdm16_expressing cells, suggesting AP1 is a co-factor.  Our results cannot rule out the presence of other co-factors.

Reviewer #2 (Recommendations for the authors):

Minor typo: [7, page 3] "sicne" should be "since".

We appreciate the reviewer’s careful reading. We have now corrected the typo and revised some part of the text to improve clarity.

Reviewer #3 (Public review):

Summary:

Bone morphogenetic protein (BMP) signaling instructs multiple processes during development including cell proliferation and differentiation. The authors set out to understand the role of PRDM16 in these various functions of BMP signaling. They find that PRDM16 and BMP co-operate to repress stem cell proliferation by regulating the genomic distribution of BMP pathway transcription factors. They additionally show that PRDM16 impacts choroid plexus epithelial cell specification. The authors provide evidence for a regulatory circuit (constituting of BMP, PRDM16, and Wnt) that influences stem cell proliferation/differentiation.

Strengths:

I find the topics studied by the authors in this study of general interest to the field, the experiments well-controlled and the analysis in the paper sound.

We thank the reviewer for their positive feedback and thoughtful summary. We appreciate the recognition of our efforts to define the role of PRDM16 in BMP signaling and stem cell regulation, as well as the soundness of our experimental design and analysis.

Weaknesses:

I have no major scientific concerns. I have some minor recommendations that will help improve the paper (regarding the discussion).

We have revised the discussion according to the suggestions.

Reviewer #3 (Recommendations for the authors):

Specific minor recommendations:

Page 18. Line 526: In a footnote, the authors point out a recent report which in parallel was investigating the link between PRDM16 and SMAD4. There is substantial non-overlap between these two papers. To aid the reader, I would encourage the authors to discuss that paper in the discussion section of the manuscript itself, highlighting any similarities/differences in the topic/results.

Thanks for the suggestion. We now included the comparison in the discussion. One conclusion between our study and this publication is consistent, that PRDM16 functions as a co-repressor of SMAD4. However, the mechanims are different. Our data suggests a model in which PRDM16 facilitates SMAD4/pSMAD binding to repress proliferation genes under high BMP conditions. However, the other report suggests that SMAD4 steadily binds to Prdm16 promoter and switches regulatory functions depending on the co-factors. Together with PRDM16, SMAD4 represses gene expression, while with SMAD3 in response to high levels of TGF-b1, it activates gene expression. These differences could be due to different signaling (BMP versus TGF-b), contexts (NSCs versus Pancreatic cancers) etc.

Page 3. Line 65: typo 'since'

We appreciate the reviewer’s careful reading. We have now corrected the typo and revised the text to improve clarity.

eLife Assessment

This important manuscript by Genzoni et al. reports the striking discovery of a regulatory role for trophic eggs in ant caste determination. Prior to this study, trophic eggs were widely assumed to play only a nutritional role in the colony, but this compelling study shows that trophic eggs can suppress queen development, and therefore regulate caste determination in specific social contexts.

Reviewer #1 (Public Review):

This manuscript describes a series of experiments documenting trophic egg production in a species of harvester ant, Pogonomyrmex rugosus. In brief, queens are the primary trophic egg producers, there is seasonality and periodicity to trophic egg production, trophic eggs differ in many basic dimensions and contents relative to reproductive eggs, and diets supplemented with trophic eggs had an effect on the queen/worker ratio produced (increasing worker production).

The manuscript is very well prepared and the methods are sufficient. The outcomes are interesting and help fill gaps in knowledge, both on ants as well as insects, more generally.

Reviewer #2 (Public review):

The revised manuscript by Genzoni et al. reports the striking discovery of a regulatory role for trophic eggs. Prior to this study, trophic eggs were widely assumed to play a nutritional role in the colony, but this study shows that trophic eggs can suppress queen development, and therefore, can play a role in regulating caste determination in specific social contexts. In this revised version of the manuscript, the authors have addressed many of the concerns raised in the first version regarding the lack of sufficient information and context in the Introduction and Discussion.

Author response:

The following is the authors’ response to the previous reviews

Public Reviews:

Reviewer #1 (Public Review):

This manuscript describes a series of experiments documenting trophic egg production in a species of harvester ant, Pogonomyrmex rugosus. In brief, queens are the primary trophic egg producers, there is seasonality and periodicity to trophic egg production, trophic eggs differ in many basic dimensions and contents relative to reproductive eggs, and diets supplemented with trophic eggs had an effect on the queen/worker ratio produced (increasing worker production).

The manuscript is very well prepared and the methods are sufficient. The outcomes are interesting and help fill gaps in knowledge, both on ants as well as insects, more generally. More context could enrich the study and flow could be improved.

We thank the reviewer for these comments. We agree that the paper would benefit from more context. We have therefore greatly extended the introduction.

Reviewer #2 (Public Review):

The manuscript by Genzoni et al. provides evidence that trophic eggs laid by the queen in the ant Pogonomyrmex rugosis have an inhibitory effect on queen development. The authors also compare a number of features of trophic eggs, including protein, DNA, RNA, and miRNA content, to reproductive eggs. To support their argument that trophic eggs have an inhibitory effect on queen development, the authors show that trophic eggs have a lower content of protein, triglycerides, glycogen, and glucose than reproductive eggs, and that their miRNA distributions are different relative to reproductive eggs. Although the finding of an inhibitory influence of trophic eggs on queen development is indeed arresting, the egg cross-fostering experiment that supports this finding can be effectively boiled down to a single figure (Figure 6). The rest of the data are supplementary and correlative in nature (and can be combined), especially the miRNA differences shown between trophic and reproductive eggs. This means that the authors have not yet identified the mechanism through which the inhibitory effect on queen development is occurring. To this reviewer, this finding is more appropriate as a short report and not a research article. A full research article would be warranted if the authors had identified the mechanism underlying the inhibitory effect on queen development. Furthermore, the article is written poorly and lacks much background information necessary for the general reader to properly evaluate the robustness of the conclusions and to appreciate the significance of the findings.

We thank the reviewer for these comments. We agree that the paper would benefit by having more background information and more discussion. We have followed this advice in the revision.

Reviewer #3 (Public Review):

In "Trophic eggs affect caste determination in the ant Pogonomyrmex rugosus" Genzoni et al. probe a fundamental question in sociobiology, what are the molecular and developmental processes governing caste determination? In many social insect lineages, caste determination is a major ontogenetic milestone that establishes the discrete queen and worker life histories that make up the fundamental units of their colonies. Over the last century, mechanisms of caste determination, particularly regulators of caste during development, have remained relatively elusive. Here, Genzoni et al. discovered an unexpected role for trophic eggs in suppressing queen development - where bi-potential larvae fed trophic eggs become significantly more likely to develop into workers instead of gynes (new queens). These results are unexpected, and potentially paradigm-shifting, given that previously trophic eggs have been hypothesized to evolve to act as an additional intracolony resource for colonies in potentially competitive environments or during specific times in colony ontogeny (colony foundation), where additional food sources independent of foraging would be beneficial. While the evidence and methods used are compelling (e.g., the sequence of reproductive vs. trophic egg deposition by single queens, which highlights that the production of trophic eggs is tightly regulated), the connective tissue linking many experiments is missing and the downstream mechanism is speculative (e.g., whether miRNA, proteins, triglycerides, glycogen levels in trophic eggs is what suppresses queen development). Overall, this research elevates the importance of trophic eggs in regulating queen and worker development but how this is achieved remains unknown.

We thank the reviewer for these comments and agree that future work should focus on identifying the substances in trophic eggs that are responsible for caste determination.  

Recommendations For The Authors:

Reviewer #1 (Recommendations For The Authors):

Introduction:

The context for this study is insufficiently developed in the introduction - it would be nice to have a more detailed survey of what is known about trophic eggs in insects, especially social insects. The end of the introduction nicely sets up the hypothesis through the prior work described by Helms Cahan et al. (2011) where they found JH supplementation increased trophic egg production and also increased worker size. I think that the introduction could give more context about egg production in Pogonomyrmex and other ants, including what is known about worker reproduction. For example, Suni et al. 2007 and Smith et al. 2007 both describe the absence of male production by workers in two different harvester ants. Workers tend to have underdeveloped ovaries when in the presence of the queen. Other species of ants are known to have worker reproduction seemingly for the purpose of nutrition (see Heinze and Hölldober 1995 and subsequent studies on Crematogaster smithi). Because some ants, including Pogonomyrmex, lack trophallaxis, it has been hypothesized that they distribute nutrients throughout the nest via trophic eggs as is seen in at least one other ant (Gobin and Ito 2000). Interestingly, Smith and Suarez (2009) speculated that the difference in nutrition of developing sexual versus worker larvae (as seen in their pupal stable isotope values) was due to trophic egg provisioning - they predicted the opposite as was found in this study, but their prediction was in line with that of Helms Cahan et al. (2011). This is all to say that there is a lot of context that could go into developing the ideas tested in this paper that is completely overlooked. The inclusion of more of what is known already would greatly enrich the introduction.

We agree that it would be useful to provide a larger context to the study. We now provide more information on the life-history of ants and explained under what situations queens and workers may produce trophic eggs. We also mentioned that some ants such as Crematogaster smithi have a special caste of “large workers” which are morphologically intermediate between winged queens and small workers and appear to be specialized in the production of unfertilized eggs. We now also mention the study of Goby and Ito (200) where the authors show that trophic eggs may play an important role in food distribution withing the colony, in particular in species where trophallaxis is rare or absent.

Methods:

L49: What lineage is represented in the colonies used? The collection location is near where both dependent-lineage (genetic caste determining) P. rugosus and "H" lineage exist. This is important to know. Further, depending on what these are, the authors should note whether this has relevance to the study. Not mentioning genetic caste determination in a paper that examines caste determination is problematic.

This is a good point. We have now provided information at the very beginning of the material and method section that the queens had been collected in populations known not to have dependentlineage (genetic caste determining) mechanisms of caste determination.

L63 and throughout: It would be more efficient to have a paragraph that cites R (must be done) and RStudio once as the tool for all analyses. It also seems that most model construction and testing was done using lme4 - so just lay this out once instead of over and over.

We agree and have updated the manuscript accordingly.

L95: 'lenght' needs to be 'length' in the formula.

Thanks, corrected.

L151: A PCA was used but not described in the methods. This should be covered here. And while a Mantel test is used, I might consider a permANOVA as this more intuitively (for me, at least) goes along with the PCA.

We added the PCA description in the Material and Method section.

Results:

I love Fig. 3! Super cool.

Thanks for this positive comment.

Discussion:

It would be good to have more on egg cannibalism. This is reasonably well-studied and could be good extra context.

We have added a paragraph in the discussion to mention that egg cannibalism is ubiquitous in ants.

Supp Table 1: P. badius is missing and citations are incorrectly attributed to P. barbatus.

P. badius was present in the Table but not with the other Pogonomyrmex species. For some genera the species were also not listed in alphabetic order. This has been corrected.

Reviewer #2 (Recommendations For The Authors):

Comments on introduction:

The introduction is missing information about caste determination in ants generally and Pogonomyrmex rugosis specifically. This is important because some colonies of Pogonomyrmex rugosis have been shown to undergo genetic caste determination, in which case the main result would be rendered insignificant. What is the evidence that caste determination in the lineages/colonies used is largely environmentally influenced and in what contexts/environmental factors? All of this should be made clear.

This is a good point. We have expanded the introduction to discuss previous work on caste determination in Pogonomyrmex species with environmental caste determination and now also provide evidence at the beginning of the Material and Method section that the two populations studied do not have a system of genetic caste determination.

Line 32 and throughout the paper: What is meant exactly by 'reproductive eggs'? Are these eggs that develop specifically into reproductives (i.e., queens/males) or all eggs that are non-trophic? If the latter, then it is best to refer to these eggs as 'viable' in order to prevent confusion.

We agree and have updated the manuscript accordingly.

Figure 1/Supp Table 1: It is surprising how few species are known to lay trophic eggs. Do the authors think this is an informative representation of the distribution of trophic egg production across subfamilies, or due to lack of study? Furthermore, the branches show ant subfamilies, not families. What does the question mark indicate? Also, the information in the table next to the phylogeny is not easy to understand. Having in the branches that information, in categories, shown in color for example, could be better and more informative. Finally, having the 'none' column with only one entry is confusing - discuss that only one species has been shown to definitely not lay trophic eggs in the text, but it does not add much to the figure.

Trophic eggs are probably very common in ants, but this has not been very well studied. We added a sentence in the manuscript to make this clear.

Thanks for noticing the error family/subfamily error. This has been corrected in Figure 1 and Supplementary Table 1.

The question mark indicates uncertainty about whether queens also contribute to the production of trophic eggs in one species (Lasius niger). We have now added information on that in the Figure legend.

We agree with the reviewer that it would be easier to have the information on whether queens and workers produce trophic on the branches of the Tree. However, having the information on the branches would suggest that the “trait” evolved on this part of the tree. As we do not know when worker or queen production of trophic eggs exactly evolved, we prefer to keep the figure as it is.

Finally, we have also removed the none in the figure as suggested by the reviewer and discussed in the manuscript the fact that the absence of trophic eggs has been reported in only one ant species (Amblyopone silvestrii: Masuko 2003_)._

Comments on materials and methods:

Why did they settle on three trophic eggs per larva for their experimental setup?

We used three trophic eggs because under natural conditions 50-65% of the eggs are trophic. The ratio of trophic eggs to viable eggs (larvae) was thus similar natural condition.

Line 50: In what kind of setup were the ants kept? Plaster nests? Plastic boxes? Tubes? Was the setup dry or moist? I think this information is important to know in the context of trophic eggs.

We now explain that colonies were maintained in plastic boxes with water tubes.

Line 60: Were all the 43 queens isolated only once, or multiple times?

Each of the 43 queens were isolated for 8 hours every day for 2 weeks, once before and once after hibernation (so they were isolated multiple times). We have changed the text to make clear that this was done for each of the 43 queens.

Could isolating the queen away from workers/brood have had an effect on the type of eggs laid?

This cannot be completely ruled out. However, it is possible to reliably determine the proportion of viable and trophic eggs only by isolating queens. And importantly the main aim of these experiments was not to precisely determine the proportion viable and trophic eggs, but to show that this proportion changes before and after hibernation and that queens do not lay viable and trophic eggs in a random sequence.

Since it was established that only queens lay trophic eggs why was the isolation necessary?

Yes this was necessary because eggs are fragile and very difficult to collect in colonies with workers (as soon as eggs are laid they are piled up and as soon as we disturb the nest, a worker takes them all and runs away with them). Moreover, it is possible that workers preferentially eat one type of eggs thus requiring to remove eggs as soon as queens would have laid them. This would have been a huge disturbance for the colonies.

Line 61: Is this hibernation natural or lab induced? What is the purpose of it? How long was the hibernation and at what temperature? Where are the references for the requirement of a diapause and its length?

The hibernation was lab induced. We hibernated the queens because we previously showed that hibernation is important to trigger the production of gynes in P. rugosus colonies in the laboratory (Schwander et al 2008; Libbrecht et al 2013). Hibernation conditions were as described in Libbrecht et al (2013).  

Line 73: If the queen is disturbed several times for three weeks, which effect does it have on its egg-laying rate and on the eggs laid? Were the eggs equally distributed in time in the recipient colonies with and without trophic eggs to avoid possible effects?

It is difficult to respond what was the effect of disturbance on the number and type of eggs laid. But again our aim was not to precisely determine these values but determine whether there was an effect of hibernation on the proportion of trophic eggs. The recipient colonies with and without trophic eggs were formed in exactly the same way. No viable eggs were introduced in these colonies, but all first instar larvae have been introduced in the same way, at the same time, and with random assignment. We have clarified this in the Material and Method section.

Line 77: Before placing the freshly hatched larvae in recipient colonies, how long were the recipient colonies kept without eggs and how long were they fed before giving the eggs? Were they kept long enough without the queen to avoid possible effects of trophic eggs, or too long so that their behavior changed?

The recipient colonies were created 7 to 10 days before receiving the first larvae and were fed ad libitum with grass seeds, flies and honey water from the beginning. Trophic eggs that would have been left over from the source colony should have been eaten within the first few days after creating the recipient colonies. However, even if some trophic eggs would have remained, this would not influence our conclusion that trophic eggs influence caste fate, given the fully randomized nature of our treatments and the considerable number of independent replicates. The same applies to potential changes in worker behavior following their isolation from the queen.

Line 77: Is it known at what stage caste determination occurs in this species? Here first instar larvae were given trophic eggs or not. Does caste-determination occur at the first instar stage? If not, what effect could providing trophic eggs at other stages have on caste-determination?

A previous study showed that there is a maternal effect on caste determination in the focal species (Schwander et al 2008). The mechanism underlying this maternal effect was hypothesized to be differential maternal provisioning of viable eggs. However, as we detail in the discussion, the new data presented in our study suggests that the mechanism is in fact a different abundance of trophic eggs laid by queens. There is currently no information when exactly caste determination occurs during development

Comments on results:

Line 65: How does investigating the order of eggs laid help to "inform on the mechanisms of oogenesis"?

We agree that the aim was not to study the mechanism of oogenesis. We have changed this sentence accordingly: “To assess whether viable and trophic eggs were laid in a random order, or whether eggs of a given type were laid in clusters, we isolated 11 queens for 10 hours, eight times over three weeks, and collected every hour the eggs laid”

Figure 2: There is no description/discussion of data shown in panels B, C, E, and F in the main text.

We have added information in the main text that while viable eggs showed embryonic development at 25 and 65 hours (Fig 12 B, C) there was no such development for trophic eggs (Fig. 2 E,F).

Line 172: Please explain hibernation details and its significance on colony development/life cycle.

We have added this information in the Material and Method section.

Figure 6: How is B plotted? How could 0% of gynes have 100% survival?

The survival is given for the larvae without considering caste. We have changed the de X axis of panel B and reworded the Figure legend to clarify this.

Is reduced DNA content just an outcome of reduced cell number within trophic eggs, i.e., was this a difference in cell type or cell number? Or is it some other adaptive reason?

It is likely to be due to a reduction in cell number (trophic eggs have maternal DNA in the chorion, while viable eggs have in addition the cells from the developing zygote) but we do not have data to make this point.

Is there a logical sequence to the sequence of egg production? The authors showed that the sequence is non-random, but can they identify in what way? What would the biological significance be?

We could not identify a logical sequence. Plausibly, the production of the two types of eggs implies some changes in the metabolic processes during egg production resulting in queens producing batches of either viable or trophic eggs. This would be an interesting question to study, but this is beyond the scope of this paper.

Figure 6b is difficult to follow, and more generally, legends for all figures can be made clearer and more easy to follow.

We agree. We have now improved the legends of Fig 6B and the other figures.

Lines 172-174: "The percentage of eggs that were trophic was higher before hibernation...than after. This higher percentage was due to a reduced number of reproductive eggs, the number of trophic eggs laid remained stable" - are these data shown? It would be nice to see how the total egglaying rate changes after hibernation. Also, is the proportion of trophic eggs laid similar between individual queens?

No the data were not shown and we do not have excellent data to make this point. We have therefore removed the sentence “This higher percentage was due to a reduced number of reproductive eggs, the number of trophic eggs laid remained stable” from the manuscript.

Figure 6B: Do several colonies produce 100% gynes despite receiving trophic eggs? It would be interesting if the authors discussed why this might occur (e.g., the larvae are already fully determined to be queens and not responsive to whatever signal is in the trophic eggs).

The reviewer is correct that 4 colonies produced 100% gynes despite receiving trophic eggs. However, the number of individuals produced in these four colonies was small (2,1,2,1, see supplementary Table 2). So, it is likely that it is just by chance that these colonies produced only gynes.

Figure 5: Why a separation by "size distribution variation of miRNA"? What is the relevance of looking at size distributions as opposed to levels?

We did that because there many different miRNA species, reflected by the fact that there is not just one size peak but multiple one. This is why we looked at size distribution

Figure 2: The image of the viable embryo is not clear. If possible, redo the viable to show better quality images.

Unfortunately, we do not anymore have colonies in the laboratory so this is not possible.

Comments on discussion:

Lines 236-247: Can an explanation be provided as to why the effect of trophic eggs in P. rugosus is the opposite of those observed by studies referenced in this section? Could P. rugosus have any life history traits that might explain this observation?

In the two mentioned studies there were other factors that co-varied with variation in the quantity of trophic eggs. We mentioned that and suggested that it would be useful to conduct experimental manipulation of the quantity of trophic eggs in the Argentine ant and P. barbatus (the two species where an effect of trophic eggs had been suggested).

The discussion should include implications and future research of the discovery.

We made some suggestions of experiments that should be performed in the future

The conclusion paragraph is too short and does not represent what was discussed.

We added two sentences at the end of the paragraph to make suggestions of future studies that could be performed.

Lines 231 to 247: Drastically reduce and move this whole part to the introduction to substantiate the assumption that trophic eggs play a nutritional role.

We moved most of this paragraph to the introduction, as suggested by the reviewer.

Reviewer #3 (Recommendations For The Authors):

I would like to commend the authors on their study. The main findings of the paper are individually solid and provide novel insight into caste determination and the nature of trophic eggs. However, the inferences made from much of the data and connections between independent lines of evidence often extend too far and are unsubstantiated.

We thank the reviewer for the positive comment. We made many changes in the manuscript to improve the discussion of our results.

eLife Assessment

This study reports useful information on the mechanisms by which a high-fat diet induces arrhythmias in the model organism Drosophila. Specifically, the authors propose that adipokinetic hormone (Akh) secretion is increased with this diet, and through binding of Akh to its receptor on cardiac neurons, arrhythmia is induced. The authors have revised their manuscript but the evidence remains incomplete. Nonetheless, the data presented will be helpful to those who wish to extend the research to a more complex model system, such as the mouse.

Reviewer #1 (Public review):

Summary:

In the manuscript submission by Zhao et al. entitled, "Cardiac neurons expressing a glucagon-like receptor mediate cardiac arrhythmia induced by high-fat diet in Drosophila" the authors assert that cardiac arrhythmias in Drosophila on a high fat diet is due in part to adipokinetic hormone (Akh) signaling activation. High fat diet induces Akh secretion from activated endocrine neurons, which activate AkhR in posterior cardiac neurons. Silencing or deletion of Akh or AkhR blocks arrhythmia in Drosophila on high fat diet. Elimination of one of two AkhR expressing cardiac neurons results in arrhythmia similar to high fat diet.

Strengths:

The authors propose a novel mechanism for high fat diet induced arrhythmia utilizing the Akh signaling pathway that signals to cardiac neurons.

Comments on revisions:

The authors have addressed my other concerns. The only outstanding issue is in regard to the following comment:

The authors state that "HFD led to increased heartbeat and an irregular rhythm." In representative examples shown, HFD resulted in pauses, slower heart rate, and increased irregularity in rhythm but not consistently increased heart rate (Figures 1B, 3A, and 4C). Based on the cited work by Ocorr et al (https://doi.org/10.1073/pnas.0609278104), Drosophila heart rate is highly variable with periods of fast and slow rates, which the authors attributed to neuronal and hormonal inputs. Ocorr et al then describe the use of "semi-intact" flies to remove autonomic input to normalize heart rate. Were semi-intact flies used? If not, how was heart rate variability controlled? And how was heart rate "increase" quantified in high fat diet compared to normal fat diet? Lastly, how does one measure "arrhythmia" when there is so much heart rate variability in normal intact flies?

- The authors state that 8 sec time windows were selected at the discretion of the imager for analysis. I don't know how to avoid bias unless the person acquiring the imaging is blinded to the condition and the analysis is also done blind. Can you comment whether data acquisition and analysis was done in a blinded fashion? If not, this should be stated as a limitation of the study.

Reviewer #3 (Public review):

Zhao et al. provide new insights into the mechanism by which a high-fat diet (HFD) induces cardiac arrhythmia employing Drosophila as a model. HFD induces cardiac arrhythmia in both mammals and Drosophila. Both glucagon and its functional equivalent in Drosophila Akh are known to induce arrhythmia. The study demonstrates that Akh mRNA levels are increased by HFD and both Akh and its receptor are necessary for high-fat diet-induced cardiac arrhythmia, elucidating a novel link. Notably, Zhao et al. identify a pair of AKH receptor-expressing neurons located at the posterior of the heart tube. Interestingly, these neurons innervate the heart muscle and form synaptic connections, implying their roles in controlling the heart muscle. The study presented by Zhao et al. is intriguing, and the rigorous characterization of the AKH receptor-expressing neurons would significantly enhance our understanding of the molecular mechanism underlying HFD-induced cardiac arrhythmia.

Many experiments presented in the manuscript are appropriate for supporting the conclusions while additional controls and precise quantifications should help strengthen the authors' arguments. The key results obtained by loss of Akh (or AkhR) and genetic elimination of the identified AkhR-expressing cardiac neurons do not reconcile, complicating the overall interpretation.

The most exciting result is the identification of AkhR-expressing neurons located at the posterior part of the heart tube (ACNs). The authors attempted to determine the function of ACNs by expressing rpr with AkhR-GAL4, which would induce cell death in all AkhR-expressing cells, including ACNs. The experiments presented in Figure 6 are not straightforward to interpret. Moreover, the conclusion contradicts the main hypothesis that elevated Akh is the basis of HFD-induced arrhythmia. The results suggest the importance of AkhR-expressing cells for normal heartbeat. However, elimination of Akh or AkhR restores normal rhythm in HFD-fed animals, suggesting that Akh and AkhR are not important for maintaining normal rhythms. If Akh signaling in ACNs is key for HFD-induced arrhythmia, genetic elimination of ACNs should unalter rhythm and rescue the HFD-induced arrhythmia. An important caveat is that the experiments do not test the specific role of ACNs. ACNs should be just a small part of the cells expressing AkhR. Specific manipulation of ACNs will significantly improve the study. Moreover, the main hypothesis suggests that HFD may alter the activity of ACNs in a manner dependent on Akh and AkhR. Testing how HFD changes calcium, possibly by CaLexA (Figure 2) and/or GCaMP, in wild-type and AkhR mutant could be a way to connect ACNs to HFD-induced arrhythmia. Moreover, optogenetic manipulation of ACNs may allow for specific manipulation of ACNs.

Interestingly, expressing rpr with AkhR-GAL4 was insufficient to eliminate both ACNs. It is not clear why it didn't eliminate both ACNs. Given the incomplete penetrance, appropriate quantifications should be helpful. Additionally, the impact on other AhkR-expressing cells should be assessed. Adding more copies of UAS-rpr, AkhR-GAL4, or both may eliminate all ACNs and other AkhR-expressing cells. The authors could also try UAS-hid instead of UAS-rpr.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public Review):

Summary:

In the manuscript submission by Zhao et al. entitled, "Cardiac neurons expressing a glucagon-like receptor mediate cardiac arrhythmia induced by high-fat diet in Drosophila" the authors assert that cardiac arrhythmias in Drosophila on a high-fat diet are due in part to adipokinetic hormone (Akh) signaling activation. High-fat diet induces Akh secretion from activated endocrine neurons, which activate AkhR in posterior cardiac neurons. Silencing or deletion of Akh or AkhR blocks arrhythmia in Drosophila on a high-fat diet. Elimination of one of two AkhR-expressing cardiac neurons results in arrhythmia similar to a high-fat diet.

Strengths:

The authors propose a novel mechanism for high-fat diet-induced arrhythmia utilizing the Akh signaling pathway that signals to cardiac neurons.

Weaknesses:

Major comments:

(1) The authors state, "Arrhythmic pathology is rooted in the cardiac conduction system." This assertion is incorrect as a blanket statement on arrhythmias. There are certain arrhythmias that have been attributable to the conduction system, such as bradycardic rhythms, heart block, sinus node reentry, inappropriate sinus tachycardia, AV nodal reentrant tachycardia, bundle branch reentry, fascicular ventricular tachycardia, or idiopathic ventricular fibrillation to name a few. However the etiological mechanism of many atrial and ventricular arrhythmias, such as atrial fibrillation or substrate-based ventricular tachycardia, are not rooted in the conduction system. The introduction should be revised to reflect a clear focus (away from?) on atrial fibrillation (AF). In addition, AF susceptibility is known to be modulated by autonomic tone, which is topically relevant (irrelevant?) to this manuscript.

Thank you for the helpful comment. We rephrased the sentence as “Arrhythmic pathology is often rooted in the cardiac conduction system”.

(2) The authors state that "HFD led to increased heartbeat and an irregular rhythm." In representative examples shown, HFD resulted in pauses, slower heart rate, and increased irregularity in rhythm but not consistently increased heart rate (Figures 1B, 3A, and 4C). Based on the cited work by Ocorr et al (https://doi.org/10.1073/pnas.0609278104), Drosophila heart rate is highly variable with periods of fast and slow rates, which the authors attributed to neuronal and hormonal inputs. Ocorr et al then describe the use of "semi-intact" flies to remove autonomic input to normalize heart rate. Were semi-intact flies used? If not, how was heart rate variability controlled? And how was heart rate "increase" quantified in high-fat diet compared to normal-fat diet? Lastly, how does one measure "arrhythmia" when there is so much heart rate variability in normal intact flies?

We also observed that fly heart rate is highly variable with periods of fast and slow rates. To control heart rate variability, Ocorr et al. used semi-intact flies to record the heartbeat  (https://doi.org/10.1073/pnas.0609278104). We consider it a rigorous method to get highly consistent results with high quality videos/images. Since our work has a focus on the neuronal inputs to the heart, we did not use the semi-intact method. Our concern is that it is likely to disrupt the neuronal processes during the dissection. Using OCT, we recorded the heartbeat of intact flies in an 8 s time window, when the heartbeat was relatively stable. The different groups of flies, which were fed on a high-fat diet or a normal-fat diet, were recorded using the same method. Thus, we could compare the differences in heart rate.

(3) The authors state, "to test whether the HFD-induced increase in Akh in the APC affects APC neuron activity, we used CaLexA (https://doi.org/10.3109/01677063.2011.642910)." According to the reference, CaLexA is a tool to map active neurons and would not indicate, as the authors state, whether Akh affects APC neuron activity specifically. It is equally possible that APC neurons may be activated by HFD and produce more Akh. Please clarify this language.

Thank you for clarifying the calcium reporter, CaLexA. We rephrased this sentence to “to test whether HFD affects APC neuron activity, we used CaLexA”.

(4) Are the AkhR+ neurons parasympathetic or sympathetic? Please provide additional experimentation that characterizes these neurons. The AkhR+ neurons appear to be anti-arrhythmic. Please expand the discussion to include a working hypothesis of the overall findings on Akh, AkhR, and AkhR+ neurons.

Noyes et al. showed that Akh treatment increases heartbeat (Noyes, B. E., F. N. Katz, and M. H. Schaffer. 1995. “Identification and Expression of the Drosophila Adipokinetic Hormone Gene.” Molecular and Cellular Endocrinology 109 (2): 133–41.), suggesting that AkhR+ neurons are sympathetic. We showed that high-fat diet induced Akh expression and secretion, which led to stimulation of AkhR+ neuron and increased heart rate, supporting the sympathetic role of the AkhR+ neurons. Additional explanation on the sympathetic & anti-arrhythmic role of the Akh, AkhR, and AkhR+ neurons were added to the discussion.

(5) The authors state, "Heart function is dependent on glucose as an energy source." However, the heart's main energy source is fatty acids with minimal use of glucose (doi: 10.1016/j.cbpa.2006.09.014). Glucose becomes more utilized by cardiomyocytes under heart failure conditions. Please amend/revise this statement.

Thank you for pointing this out and providing the reference. We rephrased this sentence “Heart function is dependent on continuous ATP production. Cardiac ATP in Drosophila might come from fatty acids, glucose, and lactate (Kodde et al., 2007), as well as trehalose.”

Reviewer #2 (Public Review):

This manuscript explores mechanisms underlying heart contractility problems in metabolic disease using Drosophila as a model. They confirm, as others have demonstrated, that a high-fat diet (HFD) induces cardiac problems in flies. They showed that a high-fat diet increased Akh mRNA levels and calcium levels in the Akh-producing cells (APC), suggesting there is increased production and release of this hormone in a HFD context. When they knock down Akh production in the APCs using RNAi they see that cardiac contractility problems are abolished. They similarly show that levels of the Akh receptor (Akhr) are increased on a HFD and that loss of Akhr also rescues contractility problems on a HFD.

One highlight of the paper was the identification of a pair of neurons that express a receptor for the metabolic hormone Akh, and showing initial data that these neurons innervate the cardiac muscle. They then overexpress cell death gene reaper (rpr) in all Akhr-positive cells with Akhr-GAL4 and see that cardiac contractility becomes abnormal.

However, this paper contains several findings that have been reported elsewhere and it contains key flaws in both experimental design and data interpretation. There is some rationale for doing the experiments, and the data and images are of good quality. However, others have shown that HFD induces cardiac contractility problems (Birse 2010), that Akh mRNA levels are changed with HFD (Liao 2021) that Akh modulates cardiac rhythms (Noyes 1995), so Figures 1-4 are largely a confirmation of what is already known. This limits the overall magnitude of the advances presented in these figures. Overall, the stated concerns limit the impact of the manuscript in advancing our understanding of heart contractility.

We thank the reviewer for the positive comments and appreciate the reviewer for the instructive suggestions. Birse 2010 (PMID: 21035763) was cited in our manuscript. Liao 2021 showed that Akh mRNA levels are changed with HFD. We added the reference to the revised manuscript and modified the text as: “In consistent with a previous work (Liao et al., 2020), we showed that the expression of Akh was significantly up-regulated in the flies fed a HFD, compared to NFD-fed flies (Figure 2B)”. Our qPCR verified Liao’s results. On top of this, we investigated the calcium levels in the Akh producing cells (APCs) and showed elevated calcium levels in the APC in HFD fed flies. In the revised version, we added more data to show that Akh protein levels were increased with HFD (Figure 2E-F). In line with Noyes' discovery, which showed that Akh injection caused cardioaccelation in prepupae, we showed that genetic manipulation of Akh expression affected heartbeat in the adults.   

Reviewer #3 (Public Review):

Zhao et al. provide new insights into the mechanism by which a high-fat diet (HFD) induces cardiac arrhythmia employing Drosophila as a model. HFD induces cardiac arrhythmia in both mammals and Drosophila. Both glucagon and its functional equivalent in Drosophila Akh are known to induce arrhythmia. The study demonstrates that Akh mRNA levels are increased by HFD and both Akh and its receptor are necessary for high-fat diet-induced cardiac arrhythmia, elucidating a novel link. Notably, Zhao et al. identify a pair of AKH receptor-expressing neurons located at the posterior of the heart tube. Interestingly, these neurons innervate the heart muscle and form synaptic connections, implying their roles in controlling the heart muscle. The study presented by Zhao et al. is intriguing, and the rigorous characterization of the AKH receptor-expressing neurons would significantly enhance our understanding of the molecular mechanism underlying HFD-induced cardiac arrhythmia.

Many experiments presented in the manuscript are appropriate for supporting the conclusions while additional controls and precise quantifications should help strengthen the authors' augments. The key results obtained by loss of Akh (or AkhR) and genetic elimination of the identified AkhR-expressing cardiac neurons do not reconcile, complicating the overall interpretation.

It is intriguing to see an increase in Akh mRNA levels in HFD-fed animals. This is a key result for linking HFD-induced arrhythmia to Akh. Thus, demonstrating that HFD also increases the Akh protein levels and Akh is secreted more should significantly strengthen the manuscript.

Thank you for the positive comments and the instructive suggestions. We performed immunostaining to show that Akh protein levels increased, which is consistent with elevated Akh mRNA expression in HFD-fed flies. The data was added to Figure 2, panels E and F. Akh secretion from the APCs is regulated by APC activity (https://doi.org/10.1038/s41586-019-1675-4). We used a calcium reporter CaLexA (https://doi.org/10.3109/01677063.2011.642910) to monitor APC activity and showed that HFD increased APC activity (Figure 2, C-D).

The experiments employing an AkhR null allele nicely demonstrate its requirement for HFD-induced cardiac arrhythmia. Depletion of Akh in Akh-expressing cells recapitulates the consequence of AkhR knockout, supporting that both Akh and its receptor are required for HFD-induced cardiac arrhythmia. Given that RNAi is associated with off-target effects and some RNAi reagents do not work, testing multiple independent RNAi lines is the standard procedure. It is also important to show the on-target effect of the RNAi reagents used in the study.

Indeed, RNAi approaches can suffer from off-target effects. For Akh experiments, we used an RNAi line BL_34960, which was generated using artificial microRNAs shRNA (DOI: 10.1038/nmeth.1592). In comparison to long-hairpin constructs, shRNA constructs are expected to be advantageous, e.g., more efficient and minimized off-target. We performed immunostaining to determine Akh-Gal4>UAS-Akh-RNAi efficiency. We showed that anti-Akh fluorescence diminished in Akh-Gal4>UAS-Akh-RNAi APCs. The data was added to Figure 3-figure supplement 1.

The most exciting result is the identification of AkhR-expressing neurons located at the posterior part of the heart tube (ACNs). The authors attempted to determine the function of ACNs by expressing rpr with AkhR-GAL4, which would induce cell death in all AkhR-expressing cells, including ACNs. The experiments presented in Figure 6 are not straightforward to interpret. Moreover, the conclusion contradicts the main hypothesis that elevated Akh is the basis of HFD-induced arrhythmia. The results suggest the importance of AkhR-expressing cells for normal heartbeat. However, elimination of Akh or AkhR restores normal rhythm in HFD-fed animals, suggesting that Akh and AkhR are not important for maintaining normal rhythms. If Akh signaling in ACNs is key for HFD-induced arrhythmia, genetic elimination of ACNs should unalter rhythm and rescue the HFD-induced arrhythmia. An important caveat is that the experiments do not test the specific role of ACNs. ACNs should be just a small part of the cells expressing AkhR. The experiments presented in Figure 6 cannot justify the authors' conclusion. Specific manipulation of ACNs will significantly improve the study. Moreover, the main hypothesis suggests that HFD may alter the activity of ACNs in a manner dependent on Akh and AkhR. Testing how HFD changes calcium, possibly by CaLexA (Figure 2) and/or GCaMP, in wild-type and AkhR mutants could be a way to connect ACNs to HFD-induced arrhythmia. Moreover, optogenetic manipulation of ACNs will allow for specific manipulation of ACNs, which is crucial for studying the specific role of ACNs in controlling cardiac rhythms.

Thank you for the insightful comments. We have been trying to find a way to only target the AkhR neurons using split-Gal4. Up to now, it’s not successful. Akh/AkhR signaling shall play a key role in the ACNs, however, we cannot rule out the possibility that ACNs also receive signals other than Akh in the modulation of heartbeat.

Interestingly, expressing rpr with AkhR-GAL4 was insufficient to eliminate both ACNs. It is not clear why it didn't eliminate both ACNs. Given the incomplete penetrance, appropriate quantifications should be helpful. Additionally, the impact on other AhkR-expressing cells should be assessed. Adding more copies of UAS-rpr, AkhR-GAL4, or both may eliminate all ACNs and other AkhR-expressing cells. The authors could also try UAS-hid instead of UAS-rpr.

We added more data to show that AkhR+ neurons are positive in anti-Akh staining, indicating the AkhR+ neurons indeed receive Akh.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Typo in line 765: "increased Akh section into the circulation." Section should be secretion.

Thank you for finding the typo. We changed section to secretion.

Reviewer #2 (Recommendations For The Authors):

One interesting extension to our knowledge in Figures 3 & 4 is that loss of Akhr and loss of Akh both block the cardiac contractility defects that accompany a HFD. The main concern I have with the Akh finding is that the authors use only a GAL4 control and no UAS alone control. Metabolic phenotypes often show strain-specific effects, so to make conclusions it is essential that the authors include a UAS alone control alongside the other genotypes to be sure it does not rescue the cardiac contractility defects that accompany a HFD by itself.

I am interested in the authors' identification of a pair of Akhr-positive neurons that innervate the cardiac muscle. I am not aware of any other studies identifying these neurons, or revealing their function. The contents of Figure 5 therefore represent the largest advance in the study. However, the characterization of these neurons is very superficial, and a lot more work to understand their regulation and function in a HFD context is needed to make conclusions about their role in any HFD-induced cardiac contractility problems. Or to determine how Akh influences the function of these specific neurons in an HFD context.

The reason I say this is that the authors ablate all Akhr-positive cells in Figure 6 and show that this disturbs normal cardiac contractility. While studies on the one pair of Akhr-positive neurons would be really interesting, ablating all Akhr-positive cells, which includes the fat and many other cell types in the fly, is not a scientifically rigorous approach to answering this question. As a result, the authors are only able to make the claim that ablating many cell types throughout the animal disrupts cardiac contractility, which does not advance our understanding of mechanisms underlying heart contractility problems. In addition, because the experiments they designed did not test whether it was Akh binding to Akhr on those neurons that regulate cardiac contractility problems in a HFD context, their experiments do not support their model in Figure 7.

The authors also make conclusions that are fairly speculative around Line 231 when describing their model in Figure 7. These claims are simply not supported by the data they present and must be removed. For example, the authors have not identified an endocrine-heart axis, they simply showed that changes in Akh can influence the heart, but this is not necessarily a direct effect on a specific cell type. They do not show data that Akh binds the newly identified Akhr-positive neuron pair to mediate the effects of HFD-induced contractility defects - they just ablate all Akhr-positive cells (fat, neurons, and other types) and show cardiac defects. If those neurons did mediate the abnormal cardiac rhythm promoted by Akh, then ablating those neurons (and not a large number of additional tissues) should rescue HFD-induced heart defects just like reducing Akhr or Akh did (but this is the opposite of what they see). Overall, concerns with experimental design, data interpretation, and relatively few findings that aren't reported elsewhere reduce the impact of this paper.

We appreciate the positive comments and helpful suggestions. Indeed, it is important to get clean genetic access to the cardiac neurons. We intended to use split Gal4 system to target the AkhR cardiac neurons. We have tried to build a split Gal4 driver AkhR-p65.AD. Two rounds of injection were carried out. However, we did not recover a transgenic line.

In the revised version, we performed immunostaining using Akh antibodies to show that anti-Akh fluorescence was observed in AkhR neurons (Figure 5-figure supplement 1), indicating an endocrine-heart axis.

eLife Assessment

This study provides fundamental information on how Arg-II participates in cardiac aging. The phenotypic data provide convincing evidence of non-cell-autonomous contributions to aging-related pathologies. Overall, the study highlights the importance of intercellular signaling in maintaining cardiac health during aging.

Reviewer #1 (Public review):

Summary:

The manuscript by Duilio M. Potenza et al. explores the role of Arginase II in cardiac aging, majorly using whole-body arg-ii knock-out mice. In this work, the authors have found that Arg-II exerts non-cell-autonomous effects on aging cardiomyocytes, fibroblasts, and endothelial cells mediated by IL-1b from aging macrophages. The authors have used arg II KO mice and an in vitro culture system to study the role of Arg II. Authors have also reported the cell-autonomous effect of Arg-II through mitochondrial ROS in fibroblasts that contribute to cardiac aging. These findings are sufficiently novel in cardiac aging and provide interesting insights. While the phenotypic data seem strong, the mechanistic details are unclear. How Arg II regulates the IL-1b and modulates cardiac aging is still being determined.

Strengths:

This study provides interesting information on the role of Arg II in cardiac aging.

The phenotypic data in the Arg II KO mice is convincing, and the authors have assessed most of the aging-related changes.

The data is supported by an in vitro cell culture system.

Weaknesses:

The manuscript needs more mechanistic details on how Arg II regulates IL-1b and modulates cardiac aging.

Reviewer #2 (Public review):

This study investigates the role of arginase-II (Arg-II) in cardiac aging. The authors challenge previous assumptions by demonstrating that Arg-II is not expressed in aged cardiomyocytes, but is upregulated in non-myocyte cells, specifically macrophages, fibroblasts, and endothelial cells. Using Arg-II knockout mice, they show protection against age-associated cardiac inflammation, fibrosis, apoptosis, endothelial-to-mesenchymal transition (EndMT), and ischemic injury. Mechanistically, Arg-II promotes IL-1β release from macrophages and increases mitochondrial ROS in fibroblasts, contributing to cardiac aging through both cell-autonomous and non-cell-autonomous mechanisms.

The study is well-structured and combines genetic models, molecular assays, and histological analyses to support its conclusions. Including both human and mouse samples strengthens the translational relevance of the findings. The authors have addressed most of the reviewers' comments and have made efforts to improve the manuscript by adding experimental data, explanations, and further discussion.

The data convincingly support their conclusions. This work provides valuable insights into the mechanisms of cardiac aging, aligns with growing evidence of non-cell-autonomous contributions to aging-related pathologies, and highlights the importance of intercellular signaling in maintaining cardiac health during aging.

Although the use of cell-specific knockout mouse models would enhance the depth and translational potential of the findings, it is understandable that such an approach would be beyond the scope of a single study. This work lays the groundwork for future investigations into conditional Arg-II knockouts in specific cell types to elucidate the cell-specific roles of Arg-II in cardiac aging.

Overall, this is a solid and impactful study with strong experimental support

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public Review):

Summary:

The manuscript by Duilio M. Potenza et al. explores the role of Arginase II in cardiac aging, majorly using whole-body arg-ii knock-out mice. In this work, the authors have found that Arg-II exerts non-cell-autonomous effects on aging cardiomyocytes, fibroblasts, and endothelial cells mediated by IL-1b from aging macrophages. The authors have used arg II KO mice and an in vitro culture system to study the role of Arg II. The authors have also reported the cell-autonomous effect of Arg-II through mitochondrial ROS in fibroblasts that contribute to cardiac aging. These findings are sufficiently novel in cardiac aging and provide interesting insights. While the phenotypic data seems strong, the mechanistic details are unclear. How Arg II regulates the IL-1b and modulates cardiac aging is still being determined. The authors still need to determine whether Arg II in fibroblasts and endothelial contributes to cardiac fibrosis and cell death. This study also lacks a comprehensive understanding of the pathways modulated by Arg II to regulate cardiac aging.

We sincerely appreciate the valuable feedback provided by the reviewer. It's gratifying to hear that our work provided novel information on the role of arginase-II in cardiac aging which is a complex process involving various cell types and mechanisms. We have devoted considerable effort by performing new experiments to address the reviewer's comments and to delineate more detailed mechanisms of Arg-II in cardiac aging. Please, see below our specific answers to each point of the reviewers.

Strengths:

This study provides interesting information on the role of Arg II in cardiac aging.

The phenotypic data in the arg II KO mice is convincing, and the authors have assessed most of the aging-related changes.

The data is supported by an in vitro cell culture system.

We appreciate this reviewer’s positive assessment on the strength of our study.

Weaknesses:

The manuscript needs more mechanistic details on how Arg II regulates IL-1b and modulates cardiac aging.

We made great effort and have performed new experiments in human monocyte cell line (THP1) in which iNOS is not expressed and not inducible by LPS and arg-ii gene was knocked out by CRISPR technology. Moreover, murine bone-marrow derived macrophages in which inos gene was ablated, is also use for this purpose. We found that in the human THP1 monocytes in which Arg-II but not iNOS is induced by LPS (100 ng/mL for 24 hours) (Suppl. Fig. 6A), mRNA and protein levels of IL-1b precursor are markedly reduced in arg-ii knockout THP1^{arg-ii^-/-} as compared to the THP1^wt cells (Suppl. Fig. 6B and 6C), further confirming that Arg-II promotes IL-1b production as also shown in RAW264.7 macrophages (Suppl. Fig. 5A and 5C). Moreover, in the mouse bone-marrow-derived macrophages, LPS-induced IL-1b production is inhibited by inos deficiency (BMDM^inos-/- vs BMDM^wt) (Suppl. Fig. 6D and 6E), while Arg-II levels are slightly enhanced in the BMDM^inos-/- cells (Suppl. Fig. 6D and 6F). All together, these results suggest that iNOS slightly reduces Arg-II expression. Arg-II and iNOS can be upregulated by LPS independently. Both Arg-II and iNOS are required for IL-1b production upon LPS stimulation as illustrated in Suppl. Fig. 6G. For detailed results and discussion, please see answers to the comments point 2 or point 6 raised by this reviewer.

The authors used whole-body KO mice, and the role of macrophages in cardiac aging is not studied in this model. A macrophage-specific arg II Ko would be a better model.

We fully agree with this comment of the reviewer. Unfortunately, this macrophage specific arg-ii knockout animal model is not available, yet. Future research shall develop the macrophage-specific arg-ii^-/- mouse model to confirm this conclusion with aging animals. Since Arg-II is also expressed in fibroblasts and endothelial cells and exerts cell-autonomous and paracrine functions, aging mouse models with conditional arg-ii knockout in the specific cell types would be the next step to elucidate cell-specific function of Arg-II in cardiac aging. We have pointed out this aspect for future research on page 19, lines 2 to 6.

Experiments need to validate the deficiency of Arg II in cardiomyocytes.

As pointed out by this reviewer in the comment point 10, Arg-II was previously reported to be expressed in isolated cardiomyocytes from in rats (PMID: 16537391). Unfortunately, negative controls. i.e., arg-ii^-/- samples were not included in the study to avoid any possible background signals. We made great effort to investigate whether Arg-II is present in the cardiomyocytes from different species including mice, rats and humans and have included old arg-ii^-/- mouse samples as a negative control. This allows to validate the antibody specificity and background noises beyond any reasonable doubt. The new experiments in Suppl. Fig. 4 confirms the specificity of the antibody against Arg-II in old mouse kidney which is known to express Arg-II in the S3 proximal tubular cells (Huang J, et al. 2021). To exclude the possible species-specific different expression of Arg-II in the cardiomyocytes, aged mouse and rat heart tissues were used for cellular localization of Arg-II by confocal immunofluorescence staining. As shown in Suppl. Fig. 4B and 4C, both species show Arg-II expression only in non-cardiomyocytes (cells between striated cardiomyocytes) (red arrows) but not in striated cardiomyocytes. Even in the rat myocardial infarction tissues, Arg-II was not found in cardiomyocytes but in endocardium cells (Suppl. Fig. 4B). In isolated cardiomyocytes exposed to hypoxia, a well know strong stimulus for Arg-II protein levels, no Arg-II signals could be detected, while in fibroblasts from the same animals, an elevated Arg-II levels under hypoxia is demonstrated (Fig. 5B). Furthermore, even RT-qPCR could not detect arg-ii mRNA in cardiomyocytes but in non-cardiomyocytes (Fig. 5C). All together, these results demonstrate that Arg-II are not expressed or at negligible levels in cardiomyocytes but expressed in non-cardiomyocytes. This new experiments with rat heart are included in the method section on page 20, the 1st paragraph. The results are described on page 7, the 1st paragraph, and discussed on page 12, the 2nd paragraph. Legend to Suppl. Fig. 4 is included in the file “Suppl. figure legend_R”.

The authors have never investigated the possibility of NO involvement in this mice model.

As above mentioned, we made great effort and have performed new experiments in human monocyte cell line (THP1) in which iNOS is not expressed and not inducible by LPS and arg-ii gene was knocked out by CRISPR technology. Moreover, murine bone-marrow derived macrophages in which inos gene was ablated, is also use for this purpose. The results show that Arg-II and iNOS can be upregulated by LPS independent of each other and iNOS slightly reduces Arg-II expression. However, both Arg-II and iNOS are required for IL-1b production upon LPS stimulation. For detailed results and discussion, please see answers to the comments point 2 or point 6 raised by this reviewer.

A co-culture system would be appropriate to understand the non-cell-autonomous functions of macrophages.

We appreciate the suggestion by this reviewer regarding the co-culture system to test the non-cell autonomous role of Arg-II. We think that our current model, which involves treating cells with conditioned media, is a well-established and effective method for demonstrating the non-cell autonomous role of Arg-II. This approach allows us to observe the effects of Arg-II on surrounding cells through the factors present in the conditioned media released from macrophages. The co-culture system could be considered, if the released factor in the conditioned medium is not stable. This is however not the case. Therefore, we are confident that our experimental model with conditioned medium is sufficiently enough to demonstrate a paracrine effect of cell-cell interaction (please also see answers to the comment point 16.

The Myocardial infarction data shown in the mice model may not be directly linked to cardiac aging.

As we have introduced and discussed in the manuscript, aging is a predominant risk factor for cardiovascular disease (CVD). Studies in experimental animal models and in humans provide evidence demonstrating that aging heart is more vulnerable to stressors such as ischemia/reperfusion injury and myocardial infarction as compared to the heart of young individuals. Even in the heart of apparently healthy individuals of old age, chronic inflammation, cardiomyocyte senescence, cell apoptosis, interstitial/perivascular tissue fibrosis, endothelial dysfunction and endothelial-mesenchymal transition (EndMT), and cardiac dysfunction either with preserved or reduced ejection fraction rate are observed. Our study is aimed to investigate the role of Arg-II in cardiac aging phenotype and age-associated cardiac vulnerability to stressors. Therefore, cardiac functional changes and myocardial infarction in response to ischemia/reperfusion injury are suitable surrogate parameters for the purpose.

Reviewer #2 (Public Review):

Summary:

The results from this study demonstrated a cell-specific role of mitochondrial enzyme arginase-II (Arg-II) in heart aging and revealed a non-cell-autonomous effect of Arg-II on cardiomyocytes, fibroblasts, and endothelial cells through the crosstalk with macrophages via inflammatory factors, such as by IL-1b, as well as a cell-autonomous effect of Arg-II through mtROS in fibroblasts contributing to cardiac aging phenotype. These findings highlight the significance of non-cardiomyocytes in the heart and bring new insights into the understanding of pathologies of cardiac aging. It also provides new evidence for the development of therapeutic strategies, such as targeting the ArgII activation in macrophages.

We're grateful for the reviewer's positive feedback, acknowledging the significant findings of our study on the role of arginase-II (Arg-II) in cardiac aging. We appreciate this reviewer’s insight into the therapeutic potential of targeting Arg-II activation in macrophages and are excited about the implications for future interventions in age-related cardiac pathologies. Thank you for recognizing the importance of our work in advancing our understanding of cardiac aging and potential therapeutic strategies.

Strengths:

This study targets an important clinical challenge, and the results are interesting and innovative. The experimental design is rigorous, the results are solid, and the representation is clear. The conclusion is logical and justified.

We thank this reviewer for the positive comment.

Weaknesses:

The discussion could be extended a little bit to improve the realm of the knowledge related to this study.

We appreciate this comment and have added and revised our discussion on this aspect accordingly at the end of the discussion section on page 19.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

I have several critical concerns, specifically about the mechanism of how Arg-II plays a role in cardiac aging.

My major concerns are:

(1) The authors have shown non-cell-autonomous effects on aging cardiomyocytes, fibroblasts, and endothelial cells mediated by IL-1b from aging macrophages. A macrophage-specific Arg-II knock-out mouse model is a suitable and necessary control to establish claims.

We fully agree with this comment of the reviewer. Unfortunately, this macrophage specific arg-ii knockout animal model is not available, yet. Future research shall develop the macrophage-specific arg-ii^-/- mouse model to confirm this conclusion with aging animals. Since Arg-II is also expressed in fibroblasts and endothelial cells and exerts cell-autonomous and paracrine functions, aging mouse models with conditional arg-ii knockout in the specific cell types would be the next step to elucidate cell-specific function of Arg-II in cardiac aging. We have pointed out this aspect for future research on page 19, lines 2 to 6.

(2) This study suggests that Arg-II exerts its effect through IL-1b in cardiac ageing. However, all experiments performed to demonstrate the link between ArgII and IL-1β are correlative at best. The underlying molecular mechanism, including transcription factors involved in the regulation of IL-1β by arg-ii, has not been demonstrated.

We sincerely appreciate this reviewer’s comment on the aspect! To make it clear, a causal role of Arg-II in promoting IL-1β production in macrophages is evidenced by the experimental results showing that old arg-ii^-/- mouse heart has lower IL-1β levels than the age-matched wt mouse heart (Fig. 6A to 6D). We further showed that the cellular IL-1β protein levels and release are reduced in old arg-ii^-/- mouse splenic macrophages as compared to the wt cells (Fig. 7A, 7C, and 7D). This result is further confirmed with the mouse macrophage cell line RAW264.7 (Suppl. Fig. 5A and suppl. Fig. 5C), in which we demonstrate that silencing arg-ii reduces IL-1β levels stimulated with LPS.

According to this reviewer’s comment (see comment point 6), we made further effort to investigate possible involvement of iNOS in Arg-II-regulated IL-1β production in macrophages stimulated with LPS. We performed new experiments in human monocyte cell line (THP1) in which iNOS is not expressed and not inducible by LPS and arg-ii gene was knocked out by CRISPR technology in the cells.

Moreover, murine bone-marrow derived macrophages in which inos gene was ablated, is also use for this purpose. We found that in the human THP1 monocytes in which Arg-II but not iNOS is induced by LPS (100 ng/mL for 24 hours) (Suppl. Fig. 6A), mRNA and protein levels of IL-1b are markedly reduced in arg-ii knockout THP1^{arg-ii^-/-} as compared to the THP1^wt cells (Suppl. Fig. 6B and 6C), further confirming that Arg-II promotes IL-1b production as also shown in RAW264.7 macrophages (Suppl. Fig. 5A and 5C). The results suggest that Arg-II promotes IL-1b production independently of iNOS. Moreover, the role of iNOS in IL-1b production was also studied in the mouse bone-marrow-derived macrophages in which inos gene is ablated. The results demonstrate that LPS-induced IL-1b production is inhibited by inos deficiency (BMDM^inos-/- vs BMDM^wt) (Suppl. Fig. 6D and 6E), while Arg-II levels are slightly enhanced in the BMDM^inos-/- cells (Suppl. Fig. 6D and 6F). Since arginase and iNOS share the same metabolic substrate L-arginine, ^inos-/- is expected to increase IL-1b production. This is however not the case. A strong inhibition of IL-1β production in ^inos-/- macrophages is observed. These results implicate that iNOS promotes IL-1β production independently of Arg-II and the inhibiting effect of IL-1β by inos deficiency is dominant and able to counteract Arg-II’s stimulating effect on IL-1β production. Hence, our results demonstrate that Arg-II promotes IL-1β production in macrophages independently of iNOS. All together, these results suggest that iNOS slightly reduces Arg-II expression. Arg-II and iNOS can be upregulated by LPS independently. Both Arg-II and iNOS are required for IL-1b production upon LPS stimulation (This concept is illustrated in the Suppl. Fig. 6G). The new results are described on page 8, the last paragraph and page 9, the 1st paragraph, presented in Suppl. Fig.6. The legend to Suppl. Fig. 6 is described in the file “Supplementary figure legend-R”. The related experimental methods are updated on page 23, the last two paragraphs and page 26 the last paragraph. The results are discussed o page 14, the last paragraph and page 15, the first two paragraphs.

(3) Figure 2: The authors have not validated the whole-body Arg-II knock-out mice for arg-ii ablation.

Thanks for pointing out this missing information! We have added the information regarding genotyping of the mice in the method section on page 20, first paragraph. Moreover, Fig. 5C also confirms the genotyping of the non-cardiomyocyte cells isolated from wt and arg-ii^-/- animals.

(4) It is unclear why the authors have chosen to focus on IL-1β specifically, among other pro-inflammatory cytokines that were also downregulated in Arg-II-/- mice as demonstrated in Fig. 2A-D.

We appreciate the reviewer's question, which provides an opportunity to delve deeper into our findings. In our investigation, we observed that aging is accompanied by elevated levels of various proinflammatory markers. Intriguingly, our data revealed that tnf-α remained unaffected by the ablation of arg-ii during aging in the heart tissues, while Il-1β showed a significant reduction in arg-ii^-/- animals compared to age-matched wild-type (wt) mice (Fig. 2). Mcp1 is however a chemoattractant for macrophages and F4-80 serves as a pan marker for macrophages. Moreover, our previous studies demonstrate a relationship between Arg-II and IL-1β in vascular disease and obesity and age-associated renal and pulmonary fibrosis. Finally, IL-1β has been shown to play a causal role in patients with coronary atherosclerotic heart disease as shown by CANTOS trials. Therefore, we have focused on IL-1β in this study. We have now explained and strengthened this aspect in the manuscript on page 7, the last two lines and page 8, the 1st paragraph as following:

“Taking into account that our previous studies demonstrated a relationship of Arg-II and IL-1β in vascular disease and obesity (Ming et al., 2012) and in age-associated organ fibrosis such as renal and pulmonary fibrosis (Huang et al., 2021; Zhu et al., 2023), and IL-1β has been shown to play a causal role in patients with coronary atherosclerotic heart disease as shown by CANTOS trials (Ridker et al., 2017), we therefore focused on the role of IL-1β in crosstalk between macrophages and cardiac cells such as cardiomyocytes, fibroblasts and endothelial cells”.

(5) Although macrophages are shown to be involved in cardiac ageing in the arg-ii mouse model, the authors have not estimated macrophage infiltration and expression of inflammatory or senescence markers in the hearts of these mice.

Thank you very much for raising this important point! Taking the comments of the reviewer into account, we have performed new experiments, i.e., multiple immunofluorescent staining to analyze the infiltrated (CCR2<sup>+</sip>/F4-80⁺) and resident (LYVE1⁺/F4-80⁺) macrophage populations and to investigate to which extent that Arg-II affects the infiltrated and resident macrophage populations in the aging heart and whether this is regulated by arg-ii^-/-. The results show an age-associated increase in the numbers of F4/80⁺ cells in the wt mouse heart, which is reduced in the age-matched arg-ii^-/- animals (Fig. 2G). This result is in accordance with the result of f4/80 gene expression shown in Fig. 2A, demonstrating that arg-ii gene ablation reduces macrophage accumulation in the aging heart. Interestingly, resident macrophages as characterized by LYVE1⁺/F4-80⁺ cells (Fig. 2E and 2H) are predominant in the aging heart as compared to the infiltrated CCR2⁺/F4-80⁺ cells (Fig. 2F and 2I). The increase in both LYVE1⁺/F4-80⁺ and CCR2⁺/F4-80⁺ macrophages in aging heart is reduced in arg-ii^-/- mice (Fig. 2E, 2F, 2H, and 2I). These new results are described on page 6, the 1st paragraph, presented in Fig. 2E to 2I, and discussed on page 13, the 2nd, paragraph. The legend to Fig. 2 is revised. The method for this additional experiment is included on page 22, the 1st paragraph.

Moreover, the aged-associated accumulation of the senescence cells as demonstrated by p16^ink4 positive cells is significantly reduced in arg-ii^-/- animals. This new result is incorporated in the Fig. 1 as Fig. 1G and 1H and described / discussed on page 5, the 2nd paragraph and page 14, the 2nd last sentences of the 1st paragraph. The method of p16^ink4 staining is included in the method section on page 22, the 1st paragraph, line 7. The legend to Fig. 1 is revised accordingly.

(6) Previously, Arg-II has been reported to serve a crucial role in ageing associated with reduced contractile function in rat hearts by regulating Nitric Oxide Synthase (PMID: 22160208). Elevated NO and superoxide have been shown to play crucial roles in the etiology of cardiovascular diseases (PMID: 24180388). Therefore, it is important to assess whether Nitric Oxide (NO) is involved in the aging-related phenotype in this mouse model.

Following the reviewer's suggestion, we conducted new experiments to investigate the role of nitric oxide (NO) in the context of the effect of Arg-II-induced IL-1b production in macrophages. We have addressed this question in the response to the comment point 2.

(7) Based on the results demonstrated in the study, ablation of Arg-II can be expected to cause a reduction in inflammation-associated phenotypes throughout the body at the multi-organ level. The observed improved cardiac phenotype could be an outcome of whole-body Arg-II ablation. It would be fruitful to develop a cardiac-specific Arg-II knockout mouse model to establish the role of Arg-II in the heart, independent of other organ systems.

We agree with the comment of the reviewer on this point. Unfortunately, as explained above (see point 1), it is currently not possible for us to perform the requested experiments, due to lack of cardiac specific arg-ii-knockout mouse model. Moreover, such an approach is complicated by the absence of Arg-II in cardiomyocytes and the expression of Arg-II in multiple cells including endothelial cells, fibroblasts and macrophage of different origin (resident and monocyte-derived infiltrating cells). It’s thus difficult to generate a cardiac-specific gene knockout mouse. One shall investigate roles of cell-specific Arg-II in cardiac aging by generating cell-specific arg-ii^-/- mice. We appreciate very this important aspect and have discussed issue on page 19, the lines 2 to 6.

(8) Contrary to the findings in this paper, Arg-II has previously been reported to be essential for IL-10-mediated downregulation of pro-inflammatory cytokines, including IL-1β (PMID: 33674584).

Thank you very much for mentioning this study! We have now discussed thoroughly the controversies as the following on page 15, the last paragraph and page 16, the 1st paragraph;

“It is of note that a study reported that Arg-II is required for IL-10 mediated-inhibition of IL-1b in mouse BMDM upon LPS stimulation (Dowling et al., 2021), which suggests an anti-inflammatory function of Arg-II. The results of our present study, however, demonstrate that LPS enhances Arg-II and IL-1b levels in macrophages and knockout or silencing Arg-II reduces IL-1b production and release, demonstrating a pro-inflammatory effect of Arg-II. Our findings are supported by the study from another group, which shows decreased pro-inflammatory cytokine production including IL-6 and IL-1b in arg-ii^-/- BMDM most likely through suppression of NFkB pathway, since arg-ii^-/- BMDM reveals decreased activation of NFkB and IL-1b levels upon LPS stimulation (Uchida et al., 2023). Most importantly, our previous study also showed that re-introducing arg-ii gene back to the arg-ii^-/- macrophages markedly enhances LPS-stimulated pro-inflammatory cytokine production (Ming et al., 2012), providing further evidence for a pro-inflammatory role of arg-ii under LPS stimulation. In support of this conclusion, chronic inflammatory diseases such as atherosclerosis and type 2 diabetes (Ming et al., 2012), inflammaging in lung (Zhu et al., 2023), kidney (Huang et al., 2021) and pancreas (Xiong, Yepuri, Necetin, et al., 2017) of aged animals or acute organ injury such as acute ischemic/reperfusion or cisplatin-induced renal injury are reduced in the arg-ii^-/- mice (Uchida et al., 2023). The discrepant findings between these studies and that with IL-10 may implicate dichotomous functions of Arg-II in macrophages, depending on the experimental context or conditions. Nevertheless, our results strongly implicate a pro-inflammatory role of Arg-II in macrophages in the inflammaging in aging heart”.

(9) The authors have only performed immunofluorescence-based experiments to show fibrotic and apoptotic phenotypes throughout this study. To verify these findings, we suggest that they additionally perform RT-PCR or western blotting analysis for fibrotic markers and apoptotic markers.

The fibrotic aspect was analyzed not only by microscopy but also by using a quantitative biochemical assay such as hydroxyproline content assessment. Hydroxyproline is a major component of collagen and largely restricted to collagen. Therefore, the measurement of hydroxyproline levels can be used as an indicator of collagen content as previous investigated in the lung (Zhu et al., 2023). We have also measured collagen genes expression by RT-qPCR as suggested by the reviewer and found an age-related decline of collagen mRNA expression levels in both wt and arg-ii^-/- mice, suggesting that the age-associated cardiac fibrosis and prevention in arg-ii^-/- mice is due to alterations of translational and/or post-translational regulations, including collagen synthesis and/or degradation. The results are in accordance with that reported by other studies published in the literature. We have pointed out this aspect on page 5, the 2nd paragraph:

“The increased cardiac fibrosis in aging is however, associated with decreased mRNA levels of collagen-Ia (col-Ia) and collagen-IIIa (col-IIIa), the major isoforms of pre-collagen in the heart (Suppl. Fig. 2A and 2B), which is a well-known phenomenon in cardiac fibrotic remodelling (Besse et al., 1994; Horn et al., 2016). The results demonstrate that age-associated cardiac fibrosis and prevention in arg-ii^-/- mice is due to alterations of translational and/or post-translational regulations including collagen synthesis and/or degradation”.

The results are presented in Suppl. Fig. 2, legend to Suppl. Fig. 2 is included in the file “Suppl. figure legend_R”. Suppl. table 2 for primers is revised accordingly.

We did not use additional markers to perform apoptotic assays with whole heart, since Fig. 3 shows good evidence that the aging is associated with increased apoptotic cells in the heart and significantly reduced in the arg-ii^-/- mice. The reduction of TUNEL positive (apoptotic) cells in aged arg-ii^-/- mice is mainly due to decrease in apoptotic cardiomyocytes. With the histological analysis, the apoptotic cell types can be well analysed. Moreover, biochemical assay for apoptosis such as caspase-3 cleavage with whole heart tissues can not distinguish apoptotic cell types and may not be sensitive enough for aging heart, due to relatively low numbers of apoptotic cells in aging heart as compared to myocardial infarct model.  

(10) Figure 4: arg-ii has previously been reported to be expressed in rat cardiomyocytes (PMID: 16537391). We strongly suggest the authors verify the expression of Arg-II via immunostaining in isolated cardiomyocytes (using published protocols), and by using multiple different cardiomyocyte-specific markers for colocalization studies to prove the lack of arg-ii expression beyond a reasonable doubt.

As pointed out by this reviewer, Arg-II was previously reported to be expressed in isolated cardiomyocytes from in rats (PMID: 16537391). Unfortunately, negative controls. i.e., arg-ii^-/- samples were not included in the study to avoid any possible background signals. We made great effort to investigate whether Arg-II is present in the cardiomyocytes from different species including mice, rats and humans and have included old arg-ii^-/- mouse samples as a negative control. This allows to validate the antibody specificity and background noises beyond any reasonable doubt. The new experiments in Suppl. Fig. 4 confirms the specificity of the antibody against Arg-II in old mouse kidney which is known to express Arg-II in the S3 proximal tubular cells (Huang J, et al. 2021). To exclude the possible species-specific different expression of Arg-II in the cardiomyocytes, aged mouse and rat heart tissues were used for cellular localization of Arg-II by confocal immunofluorescence staining. As shown in Suppl. Fig. 4B and 4C, both species show Arg-II expression only in non-cardiomyocytes (cells between striated cardiomyocytes) (red arrows) but not in striated cardiomyocytes. Even in the rat myocardial infarction tissues, Arg-II was not found in cardiomyocytes but in endocardium cells (Suppl. Fig. 4B). In isolated cardiomyocytes exposed to hypoxia, a well know strong stimulus for Arg-II protein levels, no Arg-II signals could be detected, while in fibroblasts from the same animals, an elevated Arg-II levels under hypoxia is demonstrated (Fig. 5B). Furthermore, RT-qPCR could not detect arg-ii mRNA in cardiomyocytes but in non-cardiomyocytes (Fig. 5C). All together, these results demonstrate that Arg-II are not expressed or at negligible levels in cardiomyocytes but expressed in non-cardiomyocytes. This new experiments with rat heart are included in the method section on page 20, the 1st paragraph. The results are described on page 7, the 1st paragraph, and discussed on page 12, the 2nd paragraph. Legend to Suppl. Fig. 4 is included in the file “Suppl. figure legend_R”.

(11) Figure 6G: It may be worthwhile to supplement arg-ii^-/- old cells with IL-1beta to see if there is an increase in TUNEL-positive cells.

IL-1b is a well known pro-inflammatory cytokine that causes apoptosis in various cell types including cardiomyocytes (Shen Y., et al., Tex Heart Inst J. 2015;42:109–116. doi: 10.14503/THIJ-14-4254; Liu Z. et. al., Cardiovasc Diabetol 2015;14,125. doi: 10.1186/s12933-015-0288-y; Li. Z., et al., Sci Adv 2020;6:eaay0589. doi: 10.1126/sciadv.aay0589). We appreciate very much the interesting idea of this reviewer to investigate the apoptotic responses of cardiomyocytes from arg-ii^-/- mice to IL-1b. We agree that it is possible that cardiomyocytes from wt from arg-ii^-/- mice react differently to IL-1b, although the cardiomyocytes do not express Arg-II as demonstrated in our present study. If this is true, it must be due to non-cell autonomous effects of different aging microenvironment in the heart or epigenetic modulations of the myocytes. We found that this is a very interesting aspect and requires further extensive investigation. Since our current study focused on the effect of wt and arg-ii^-/- macrophages on cardiomyocytes and non-cardiomyocytes, we prefer not to include this suggested aspect in our manuscript and would like to explore it in the following study.

(12) Figures 4-9: It would be interesting to see if the effect of ArgII in cardiac ageing is gender-specific. It is recommended to include experimental data with male mice in addition to the results demonstrated in female mice.

As pointed out in the manuscript, we have focused on female mice, because an age-associated increase in arg-ii expression is more pronounced in females than in males (Fig. 1A). As suggested by this reviewer, we performed additional experiments investigating effects of arg-ii deficiency in male mice during aging, focusing on pathophysiological outcomes of ischemia/reperfusion injury in ex vivo experiments. The ex vivo functional analytic experiments with Langendorff system were performed in aged male mice (see Suppl. Fig. 9). Following ischemia/reperfusion injury, wt male mice display reduced left ventricular developed pressure (LVDP), as well as the inotropic and lusitropic states (expressed as dP/dt max and dP/dt min, respectively). As previously reported (Murphy et al., 2007), we also found that old male mice are more prone to I/R injury than age-matched female animals. Specifically, 15 minutes of ischemia are enough to significantly affect the left ventricle contractile function in the male mice (Suppl. Fig. 9). As opposite, age-matched old female mice are relatively resistant to I/R injury, and at least 20 min of ischemia are necessary to induce a significant impairment of the contractile function (Fig. 10). Similar to females, the post I/R recovery of cardiac function is also significantly improved in the male arg-ii^-/- mice as compared to age-matched wt animals. In addition to functional recovery, triphenyl tetrazolium chloride (TTC) staining (myocardial infarction) upon I/R-injury in males is significantly reduced in the age-matched male arg-ii^-/- animals (Suppl. Fig. 9C and 9D). All together, these results reveal a role for Arg-II in heart function impairment during aging in both genders with a higher vulnerability to stress in the males. These new results are presented in Suppl. Fig. 9, described on page 10, the last paragraph and page 11. The results are discussed on page 18, the 2nd paragraph as following:

“The fact that aged females have higher Arg-II but are more resistant to I/R injury seems contradictory to the detrimental effect of Arg-II in I/R injury. It is presumable that cardiac vulnerability to injuries stressors depends on multiple factors/mechanisms in aging. Other factors/mechanisms associated with sex may prevail and determine the higher sensitivity of male heart to I/R injury, which requires further investigation. Nevertheless, the results of our study show that Arg-II plays a role in cardiac I/R injury also in males”.

The information on the experimental methods in the male animals is included on page 20, the last paragraph and page 21, the 1st paragraph. Legend to Suppl. Fig. 9 is included in the file “Suppl. figure legend_R”.

(13) Figure 6G: cardiomyocytes from wild-type mice, when treated with macrophages, show 0% TUNEL-positive cells. Since it is unlikely to obtain no TUNEL staining in a cell population, there may be an experimental or analytical error.

Now it is Fig. 7F and 7G. This is due to our specific experimental procedure. After tissue digestion, cardiomyocytes were plated on laminin-coated dishes. Laminin promotes the adhesion of survived cells. Following plating, we conducted a deep washing process to remove damaged and partially adherent cells. This step ensures that only well-shaped, viable, and strongly adherent cells remain as bioassay cells. These “healthy” cells are then selected for the experiments. the apoptotic cells are removed by washing out, reflecting the high viability of the bioassay cells. We have added this detailed information in the method section on page 24, the 2nd paragraph.

(14) Figure 7J: Please assess whether arg-ii depletion also affects the mtROS phenotype.

According to the suggestion of this reviewer, we performed new experiments which show that human cardiac fibroblasts (HCFs) exposed to hypoxia (1% O₂, 48 hours), a known physiological trigger of Arg-II up-regulation, exhibit increased mtROS generation, which involves Arg-II (new Fig. 8M to 8P). We found that Arg-II protein level as well as mtROS (assessed by mitoSOX staining) were both enhanced, accompanied by increased levels of HIF1α (Fig 8M). Moreover, mito-TEMPO pre-incubation reduces mtROS, confirming the mitochondrial origin of the ROS. Silencing of arg-ii with rAd-mediated shRNA, significantly reduces mtROS levels demonstrating a role of Arg-II in the production of mitochondrial ROS in cardiac fibroblasts (Fig 8M to 8P). We have included these results on page 9, the last paragraph and discussed the results on page 17, the 1st paragraph. The related method is described on page 26, the 2nd paragraph. Legend to Fig. 8 is updated on page 32.

(15) Figure 8A-E: The authors have treated human-origin endothelial cells with mice-origin macrophage-conditioned media. It would be more suitable to treat the endothelial cells with human-origin macrophage-conditioned media.

We acknowledge the concern regarding the use of mouse-origin macrophage-conditioned media on human-origin endothelial cells. It is to note, the biological cross-reactivity of cytokines from one species on cells from a different species has been reported in the literature. It was observed that there is quite a strict threshold of 60% amino acid identity, above which cytokines tend to cross-react and statistically, cytokines would tend to cross-react more often as their % amino acid identity increases (Scheerlinck JPY. Functional and structural comparison of cytokines in different species. Vet Immunol Immunopathol. 1999; 72:39-44. https://doi.org/10.1016/S0165-2427(99)00115-4). Taking IL-1b as an example, the 17.5 kDa mature mouse and human IL-1b share 92% aa sequence identity, suggesting a high cross-reactivity. Indeed, human IL-1b has shown biological cross-reactivity in mouse cells (Ledesma E., et al. Interleukin-1 beta (IL-1β) induces tumor necrosis factor alpha (TNF-α) expression on mouse myeloid multipotent cell line 32D cl3 and inhibits their proliferation. Cytokine. 2004; 26:66-72. https://doi.org/10.1016/j.cyto.2003.12.009). Moreover, our results also support the reported cross-reactivity between human and mouse IL-1b. The CM from mouse macrophage indeed showed biological function in human endothelial cells. The observed effects of the conditioned media from aged wild-type macrophages on endothelial cells were specifically mediated through IL-1β. This conclusion is supported by our data showing that the upregulation induced by the conditioned media was significantly reduced by the addition of an IL-1β receptor blocker.

(16) The co-culture system would be more interesting to test the non-cell autonomous role of Arg II.

We appreciate the suggestion by this reviewer regarding the co-culture system to test the non-cell autonomous role of Arg-II. We believe that our current model, which involves treating cells with conditioned media, is a well-established and effective method for demonstrating the non-cell autonomous role of Arg-II. This approach allows us to observe the effects of Arg-II on surrounding cells through the factors present in the conditioned media. The co-culture system could be considered, if the released factor in the conditioned medium is not stable. This is however not the case. So we are confident that our experimental model with conditioned medium is good enough to demonstrate a paracrine effect of cell-cell interaction.

Reviewer #2 (Recommendations For The Authors):

Some minor comments may be considered to improve the realm of the knowledge related to this study.

We appreciate this comment and have added and revised our discussion on this aspect accordingly at the end of the discussion section on page 19, the last 6 lines.

(1) The current study showed strong evidence demonstrating the key role of cardiac macrophages in pathologies of cardiac aging, particularly, the macrophages (MФ) from the circulating blood (hematogenous). It is known that the heart is among the minority of organs in which substantial numbers of yolk-sac MФ persist in adulthood and play a crucial role in maintaining cardiac function. Thus, the adult mammalian heart contains two separate and discrete cardiac MФ subgroups, i.e., the resident MФs originated from yolk sac-derived progenitors and the hematogenous MФs recruited from circulating blood monocytes. These two subtypes of MФs may play distinctive roles in the aging heart and the response to cardiac injury. The author could extend the discussion on the possibility of the resident MФs in aging hearts, which could be further investigated in the future.

We appreciate the suggestion and agree that it provides valuable insight into the study. Taking the comments of the reviewer 1 into account, we have performed new experiments, i.e., co- immunostaining to analyze the infiltrated (CCR2⁺/F4-80⁺) and resident (LYVE1⁺/F4-80⁺) macrophage populations and to investigate to which extent that Arg-II affects infiltrated and resident macrophage populations in the aging heart. We found that in line with the gene expression of f4/80, immunofluorescence staining reveals an age-associated increase in the numbers of F4/80⁺ cells in the wt mouse heart, which is reduced in the age-matched arg-ii^-/- animals (Fig. 2E, F, G), demonstrating that arg-ii gene ablation reduces macrophage accumulation in the aging heart. Interestingly, resident macrophages as characterized by LYVE1⁺/F4-80⁺ cells (Fig. 2E and 2H) are predominant in the aging heart as compared to the infiltrated CCR2⁺/F4-80⁺ cells (Fig. 2F and 2I). The increase in both LYVE1⁺/F4-80⁺ and CCR2⁺/F4-80⁺ macrophages in aging heart is reduced in arg-ii^-/- mice (Fig. 2E, 2F, 2H, and 2I). These new results are described on page 6, the 1st paragraph, presented in Fig. 2E to 2I, and discussed on page 13, the 2nd, paragraph. The legend to Fig. 2 is revised. The method for this additional experiment is included on page 22, the 1st paragraph.

(2) It would be beneficial to the readers if the author could provide some explanation about why ArgII could not be detected in VSMCs in the mouse heart and the species difference between humans and mice. In addition, the author may provide an assumption on the possibility that there may also be a cross-talk between macrophages and VSMCs in the aging heart. A little bit more explanation in the Discussion will be helpful.

We acknowledge and appreciate the suggestion and have discussed these points on page 19 as the following:

“In this context, another interesting aspect is the cross-talk between macrophages and vascular SMC in the aging heart. In our present study, we could not detect Arg-II in vascular SMC of mouse heart but in that of human heart. This could be due to the difference in species-specific Arg-II expression in the heart or related to the disease conditions in human heart which is harvested from patients with cardiovascular diseases. Indeed, in the apoe^-/- mouse atherosclerosis model, aortic SMCs do express Arg-II (Xiong et al., 2013). It is interesting to note that rodents hardly develop atherosclerosis as compared to humans. Whether this could be partly contributed by the different expression of Arg-II in vascular SMC between rodents and humans requires further investigation. In our present study, the aspect of the cross-talk between macrophages and vascular SMC is not studied. Since the crosstalk between macrophages and vascular SMC has been implicated in the context of atherogenesis as reviewed (Gong et al., 2025), further work shall investigate whether Arg-II expressing macrophages could interact with vascular SMC in the coronary arteries in the heart and contribute to the development of coronary artery disease and/or vascular remodelling and the underlying mechanisms“.

(3) Please clarify the arrows in Figure 9C that indicate the infarct area in each splicing section from one heart.

The arrows in Figure 9C (now Fig. 10C) are indeed utilized to indicate the sections displaying the infarcted area within each splicing section from one heart. We have explained the arrow in the figure legend (now Fig. 10 and also new Suppl. Fig. 9).

Joint Public Review:

Summary:

The authors have conducted the largest to date Mendelian Randomization (MR) analysis of the association between genetically predicted measures of adiposity and risk of head and neck cancer (HNC) overall and by subsites within HNC. MR uses genetic predictors of an exposure, such as gene variants associated with high BMI or tobacco use, rather than data from individual physical exams or questionnaires, and if it can be done in its idealized state, there should be no problems with confounding. Traditional epidemiologic studies have reported a variety of associations between BMI (and a few other measures of adiposity) and risk of HNC that typically differ by the smoking status of the subjects. Those findings are controversial given the complex relationship between tobacco and both BMI and HNC risk. Tobacco smokers are often thinner than non-smokers, so this could create an artificial ('confounded') association that may not be fully adjusted away in risk models. The findings of a BMI-HNC association are often attributed to residual confounding, and this seems ripe for an MR approach if suitable genetic instrumental variables can be created. Here, the authors built a variety of genetic instrumental variables for BMI and other measures of adiposity, as well as two instrumental variables for smoking habits, and then tested their hypotheses in a large case-control set of HNC and controls with genetic data.

The authors found that the genetic model for BMI was associated with HNC risk in simple models, but this association disappeared when using models that better accounted for pleiotropy, the condition when genetic variants are associated with more than one trait, such as both BMI and tobacco use. When they used both adiposity and tobacco use genetic instruments in a single model, there was a strong association with genetically predicted tobacco use (as is expected), but there was no remaining association with genetic predictors of adiposity. They conclude that high BMI/adiposity is not a risk factor for HNC.

Strengths:

The primary strength was the expansive use of a variety of different genetic instruments for BMI/adiposity/body size, along with employing a variety of MR model types, several of which are known to be less sensitive to pleiotropy. They also used the largest case-control sample size to date.

Weaknesses:

The lack of pleiotropy is an unconfirmable assumption of MR, and the addition of those models is therefore quite important, as this is a primary weakness of the MR approach. Given that concern, I read the sensitivity analyses using pleiotropy-robust models as the main result, and in that case, they can't test their hypotheses as these models do not show a BMI instrumental variable association. The other weakness, which might be remedied, is that the power of the tests here is not described. When a hypothesis is tested with an under-powered model, the apparent lack of association could be due to inadequate sample size rather than a true null. Typically, when a statistically significant association is reported, power concerns are discounted as long as the study is not so small as to create spurious findings. That is the case with their primary BMI instrumental variable model - they find an association so we can presume it was adequately powered. But the primary models they share are not the pleiotropy-robust methods MR-Egger, weighted median, and weighted mode. The tests for these models are null, and that could mean a couple of things: (1) the original primary significant association between the BMI genetic instrument was due to pleiotropy, and they therefore don't have a robust model to explore the effects of the tobacco genetic instrument. (2) The power for the sensitivity analysis models (the pleiotropy-robust methods) is inadequate, and the authors share no discussion about the relative power of the different MR approaches. If they do have adequate power, then again, there is no need to explore the tobacco instrument.

Reviewing Editor Comments:

We suggest that the authors add power estimates to assess whether the sample size is sufficient, given the strength and variability of the genetic instruments. It would also be helpful to present effect estimates for the tobacco instruments alone, to clarify their independent contribution and improve the interpretation of the joint models. In addition, the role of pleiotropy should be addressed more clearly, including which model is considered primary. Stratified analyses by smoking status are encouraged, as prior studies indicate that BMI-HNC associations may differ between smokers and non-smokers. Finally, the comparison with previous studies should be revised, as most reported null findings without accounting for tobacco instruments. If this study finds an association, it should not be framed as a replication.

Author response:

Our response aims to address the following:

The lack of pleiotropy is an unconfirmable assumption of MR, and the addition of those models is therefore quite important, as this is a primary weakness of the MR approach. Given that concern, I read the sensitivity analyses using pleiotropy-robust models as the main result, and in that case, they can't test their hypotheses as these models do not show a BMI instrumental variable association. The other weakness, which might be remedied, is that the power of the tests here is not described. When a hypothesis is tested with an under-powered model, the apparent lack of association could be due to inadequate sample size rather than a true null. Typically, when a statistically significant association is reported, power concerns are discounted as long as the study is not so small as to create spurious findings. That is the case with their primary BMI instrumental variable model - they find an association so we can presume it was adequately powered. But the primary models they share are not the pleiotropy-robust methods MR-Egger, weighted median, and weighted mode. The tests for these models are null, and that could mean a couple of things: (1) the original primary significant association between the BMI genetic instrument was due to pleiotropy, and they therefore don't have a robust model to explore the effects of the tobacco genetic instrument. (2) The power for the sensitivity analysis models (the pleiotropy-robust methods) is inadequate, and the authors share no discussion about the relative power of the different MR approaches. If they do have adequate power, then again, there is no need to explore the tobacco instrument.

We would like to highlight that post-hoc power calculations are often considered redundant since the statistical power estimated for an observed association is directly related to its p-value[1]. In other words, the uncertainty of the association is already reflected in its 95% confidence interval. However, we understand power calculations may still be of interest to the reader, so we will incorporate them in the revised manuscript.

The reason we use inverse variance weighted (IVW) Mendelian randomization (MR) to obtain our main results rather than the pleiotropy-robust methods mentioned by the reviewer/editors (i.e., MR-Egger, weighted median and weighted mode) is that the former has greater statistical power than the latter[2]. Hence, instead of focussing on the statistical significance of the pleiotropy-robust analyses, we consider it is of more value to compare the consistency of the effect sizes and direction of the effect estimates across methods. Any evidence of such consistency increases our confidence in our main findings, since each method relies on different assumptions. As we cannot be sure about the presence and nature of horizontal pleiotropy, it is useful to compare results across methods even though they are not equally powered. It is true that our results for the genetically predicted effects of body mass index (BMI) on the risk of head and neck cancer (HNC) differ across methods. This is precisely what led us to question the validity of our main finding (suggesting a positive effect of BMI on HNC risk). We will clarify this in the discussion section of the revised manuscript as advised.

We understand that the reviewer/editors are concerned that we do not have a robust model to explore the role of tobacco consumption in the link between BMI and HNC. However, we have a different perspective on the matter. If indeed, the main IVW finding for BMI and HNC is due to pleiotropy (since some of the pleiotropy-robust methods suggest conflicting results), then the IVW multivariable MR method is a way to explore the potential source of this bias[3]. We were particularly interested in exploring the role of smoking in the observed association because smoking and adiposity are known to influence each other [4-9] and share a genetic basis[10, 11].

References:

(1) Heinsberg LW, Weeks DE: Post hoc power is not informative. Genet Epidemiol 2022, 46(7):390-394.

(2) Burgess S, Butterworth A, Thompson SG: Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013, 37(7):658-665.

(3) Burgess S, Davey Smith G, Davies NM, Dudbridge F, Gill D, Glymour MM, Hartwig FP, Kutalik Z, Holmes MV, Minelli C et al: Guidelines for performing Mendelian randomization investigations: update for summer 2023. Wellcome Open Res 2019, 4:186.

(4) Morris RW, Taylor AE, Fluharty ME, Bjorngaard JH, Asvold BO, Elvestad Gabrielsen M, Campbell A, Marioni R, Kumari M, Korhonen T et al: Heavier smoking may lead to a relative increase in waist circumference: evidence for a causal relationship from a Mendelian randomisation meta-analysis. The CARTA consortium. BMJ Open 2015, 5(8):e008808.

(5) Taylor AE, Morris RW, Fluharty ME, Bjorngaard JH, Asvold BO, Gabrielsen ME, Campbell A, Marioni R, Kumari M, Hallfors J et al: Stratification by smoking status reveals an association of CHRNA5-A3-B4 genotype with body mass index in never smokers. PLoS Genet 2014, 10(12):e1004799.

(6) Taylor AE, Richmond RC, Palviainen T, Loukola A, Wootton RE, Kaprio J, Relton CL, Davey Smith G, Munafo MR: The effect of body mass index on smoking behaviour and nicotine metabolism: a Mendelian randomization study. Hum Mol Genet 2019, 28(8):1322-1330.

(7) Asvold BO, Bjorngaard JH, Carslake D, Gabrielsen ME, Skorpen F, Smith GD, Romundstad PR: Causal associations of tobacco smoking with cardiovascular risk factors: a Mendelian randomization analysis of the HUNT Study in Norway. Int J Epidemiol 2014, 43(5):1458-1470.

(8) Carreras-Torres R, Johansson M, Haycock PC, Relton CL, Davey Smith G, Brennan P, Martin RM: Role of obesity in smoking behaviour: Mendelian randomisation study in UK Biobank. BMJ 2018, 361:k1767.

(9) Freathy RM, Kazeem GR, Morris RW, Johnson PC, Paternoster L, Ebrahim S, Hattersley AT, Hill A, Hingorani AD, Holst C et al: Genetic variation at CHRNA5-CHRNA3-CHRNB4 interacts with smoking status to influence body mass index. Int J Epidemiol 2011, 40(6):1617-1628.

(10) Thorgeirsson TE, Gudbjartsson DF, Sulem P, Besenbacher S, Styrkarsdottir U, Thorleifsson G, Walters GB, Consortium TAG, Oxford GSKC, consortium E et al: A common biological basis of obesity and nicotine addiction. Transl Psychiatry 2013, 3(10):e308.

(11) Wills AG, Hopfer C: Phenotypic and genetic relationship between BMI and cigarette smoking in a sample of UK adults. Addict Behav 2019, 89:98-103.

eLife Assessment

The findings represent an important contribution to understanding whether BMI influences head and neck cancer (HNC) risk after accounting for tobacco use. Within the context of the Mendelian Randomization (MR) field, the strength of evidence appears convincing, supported by rigorous methods and a thorough exploration of multiple genetic models of adiposity using diverse MR approaches. Limitations include the absence of associations in sensitivity models designed to better account for pleiotropy, which prevents evaluation of whether incorporating an instrumental variable for tobacco use would alter the findings. Additionally, the lack of a formal power assessment for detecting associations with the instrumental variables employed limits the interpretability and reach of the results.

eLife Assessment

This study identifies novel approaches to improving transgene expression in the injured mammalian myocardium through a combination of a tissue regeneration enhancer element and engineered AAVs - specifically, a liver-detargeting capsid, AAV.cc84, and an in vivo library screen-selected AAV-IR41. The evidence is convincing, and the AAV vectors are of fundamental value to the field of cardiac gene therapy. Future research exploring how to combine the features of AAV.cc84 and AAV-IR41 could yield an even more promising vector for therapeutic use.

Reviewer #1 (Public review):

In this manuscript, Wolfson and co-authors demonstrate a combination of an injury-specific enhancer and engineered AAV that enhances transgene expression in injured myocardium. The authors characterize spatiotemporal dynamics of TREE-directed AAV expression in the injured heart using a non-invasive longitudinal monitoring system. They show that transgene expression is drastically increased 3 days post-injury, driven by 2ankrd1a. They reported a liver-detargeted capsid, AAV cc.84, with decreased viral entry into the liver while maintaining TREE transgene specificity. They further identified the IR41 serotype with enhanced transgene expression in injured myocardium from AAV library screening. This is an interesting study that optimizes the potential application of TREE delivery for cardiac repair. However, several concerns were raised prior to publication:

Major Concerns:

(1) In Figure 1, the authors demonstrated that 2andkrd1aEN is not responsive to sham injury after AAV delivery, but Figure 3 shows a strong response to sham when AAV is delivered after injury. The authors do not provide an explanation for this observation.

(2) In Figure 4, a higher GFP signal is observed in all areas of the heart of the IR41-treated mouse compared to AAV9. The authors should compare GFP expression between AAV9 and IR41 in uninjured hearts and provide insights into enhanced cardiac tropism to confirm that IR41 is MI injury enriched, not Sham as well.

(3) The authors should clarify which model is being used between myocardial infarction (MI) and Ischemia-reperfusion (IR) throughout the figures, as the experimental schemes and figure legends did not match with each other (MI or IR in Figure 1A, 1D, 3A, and 3E). Both models cause different types of injuries. The authors should explain the difference in TREE expression in both models.

(4) In Figure 2, the authors use REN instead of 2ankrd1aEN to demonstrate liver-detargeting using AAV cc.84. Is there a specific reason?