eLife Assessment

This study by Lapao et al. uncovers a novel role for the Rab27A effector SYTL5 in regulating mitochondrial function and mitophagy under hypoxic conditions. Using a range of imaging and functional assays, the authors demonstrate that SYTL5 localizes to mitochondria in a Rab27A-dependent manner and impacts mitochondrial respiration and metabolic reprogramming. While the findings are solid and valuable in the area of cancer biology, further mechanistic clarity and improved imaging would strengthen the conclusions.

Reviewer #1 (Public review):

Summary:

In this study, Ana Lapao et al. investigated the roles of Rab27 effector SYTL5 in cellular membrane trafficking pathways. The authors found that SYTL5 localizes to mitochondria in a Rab27A-dependent manner. They demonstrated that SYTL5-Rab27A positive vesicles containing mitochondrial material are formed under hypoxic conditions, thus they speculate that SYTL5 and Rab27A play roles in mitophagy. They also found that both SYTL5 and Rab27A are important for normal mitochondrial respiration. Cells lacking SYTL5 undergo a shift from mitochondrial oxygen consumption to glycolysis which is a common process known as the Warburg effect in cancer cells. Based on cancer patient database, the author noticed that low SYTL5 expression is related to reduced survival for adrenocortical carcinoma patients, indicating SYTL5 could be a negative regulator of the Warburg effect and potentially tumorigenesis.

Strengths:

The authors take advantages of multiple techniques and novel methods to perform the experiments.

(1) Live-cell imaging revealed that stably inducible expression of SYTL5 co-localized with filamentous structures positive for mitochondria. This result was further confirmed by using correlative light and EM (CLEM) analysis and western blotting from purified mitochondrial fraction.

(2) In order to investigate whether SYTL5 and RAB27A are required for mitophagy in hypoxic conditions, two established mitophagy reporter U2OS cell lines were used to analyze the autophagic flux.

Weaknesses:

This study revealed a potential function of SYTL5 in mitophagy and mitochondrial metabolism. However, the mechanistic evidence that establishes the relationship between SYTL5/Rab27A and mitophagy is insufficient. The involvement of SYTL5 in ACC needs more investigation. Furthermore, images and results supporting the major conclusions need to be improved.

Comments on revisions: The authors did not revise the paper as suggested.

Reviewer #2 (Public review):

Summary:

The authors provide convincing evidence that Rab27 and STYL5 work together to regulate mitochondrial activity and homeostasis.

Strengths:

The development of models which allow the function to be dissected, and the rigous approach and testing of mitochondrial activity.

This work is carefully done, and supports the importance of the roles of Rab27A and STYL5.

Reviewer #3 (Public review):

In the manuscript by Lapao et al., the authors uncover a role for the RAB27A effector protein SYTL5 in regulating mitochondrial function and apparent selective turnover of mitochondrial components. The authors find that SYTL5 localizes to mitochondria in a RAB27A dependent way and that loss of SYTL5 (or RAB27A) impairs lysosomal turnover of MTCO1 (but not a matrix-based reporter/other mitochondrial proteins). The authors go on to show that loss of SYTL5 impacts mitochondrial respiration and ECAR and as such may influence the Warburg effect and tumorigenesis. Of relevance here, the authors go on to show that SYTL5 expression is reduced in adrenocortical carcinomas and this correlates with reduced survival rates.

As previously reviewed, this is a very intriguing body of work and reveals a new role for SYTL5/RAB27A at the mitochondria. Unfortunately, it appears that SYTL5 is challenging protein to detect endogenously and the authors' cell lines "comprise a heterogenous pool with high variability", which means that a lot of my original concerns remain. It is still also not clear if the conventional autophagy machinery is required for this pathway, especially if SYTL5/RAB27A mitochondrial recruitment is upstream of this. Hopefully, in future work, the authors (and/or others) will be able to address this and build on the mechanisms of this interesting and potentially important pathway.

Author response:

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

Reviewer #1 (Public review):

Summary:

In this study, Ana Lapao et al. investigated the roles of Rab27 effector SYTL5 in cellular membrane trafficking pathways. The authors found that SYTL5 localizes to mitochondria in a Rab27A-dependent manner. They demonstrated that SYTL5-Rab27A positive vesicles containing mitochondrial material are formed under hypoxic conditions, thus they speculate that SYTL5 and Rab27A play roles in mitophagy. They also found that both SYTL5 and Rab27A are important for normal mitochondrial respiration. Cells lacking SYTL5 undergo a shift from mitochondrial oxygen consumption to glycolysis which is a common process known as the Warburg effect in cancer cells. Based on the cancer patient database, the author noticed that low SYTL5 expression is related to reduced survival for adrenocortical carcinoma patients, indicating SYTL5 could be a negative regulator of the Warburg effect and potentially tumorigenesis.

Strengths:

The authors take advantage of multiple techniques and novel methods to perform the experiments.

(1) Live-cell imaging revealed that stably inducible expression of SYTL5 co-localized with filamentous structures positive for mitochondria. This result was further confirmed by using correlative light and EM (CLEM) analysis and western blotting from purified mitochondrial fraction.

(2) In order to investigate whether SYTL5 and Rab27A are required for mitophagy in hypoxic conditions, two established mitophagy reporter U2OS cell lines were used to analyze the autophagic flux.

Weaknesses:

This study revealed a potential function of SYTL5 in mitophagy and mitochondrial metabolism. However, the mechanistic evidence that establishes the relationship between SYTL5/Rab27A and mitophagy is insufficient. The involvement of SYTL5 in ACC needs more investigation. Furthermore, images and results supporting the major conclusions need to be improved.

We thank the reviewer for their constructive comments. We agree that a complete understanding of the mechanism by which SYTL5 and Rab27A are recruited to the mitochondria and subsequently involved in mitophagy requires further investigation. Here, we have shown that SYTL5 recruitment to the mitochondria requires both its lipid-binding C2 domains and the Rab27A-binding SHD domain (Figure 1G-H). This implies a coincidence detection mechanism for mitochondrial localisation of SYTL5.  Additionally, we find that mitochondrial recruitment of SYTL5 is dependent on the GTPase activity and mitochondrial localisation of Rab27A (Figure 2D-E). We also identified proteins linked to the cellular response to oxidative stress, reactive oxygen species metabolic process, regulation of mitochondrion organisation and protein insertion into mitochondrial membrane to be enriched in the SYTL5 interactome (Figure 3A and C).

However, less details regarding the mitochondrial localisation of Rab27A are understood. To investigate this, we have now performed a mass spectrometry analysis to identify the interactome of Rab27A (see Author response table 1 below,). U2OS cells with stable expression of mScarlet-Rab27A or mScarlet only, were subjected to immunoprecipitation, followed by MS analysis.  Of the 32 significant Rab27A-interacting hits (compared to control), two of the hits are located in the inner mitochondrial membrane (IMM); ATP synthase F(1) complex subunit alpha (P25705), and mitochondrial very long-chain specific acyl-CoA dehydrogenase (VLCAD)(P49748). However, as these IMM proteins are not likely involved in mitochondrial recruitment of Rab27A, observed under basal conditions, we choose not to include these data in the manuscript. 

It is known that other RAB proteins are recruited to the mitochondria. During parkin-mediated mitophagy, RABGEF1 (a guanine nucleotide exchange factor) is recruited through its ubiquitin-binding domain and directs mitochondrial localisation of RAB5, which subsequently leads to recruitment of RAB7 by the MON1/CCZ1 complex[1]. As already mentioned in the discussion (p. 12), ubiquitination of the Rab27A GTPase activating protein alpha (TBC1D10A) is reduced in the brain of Parkin KO mouse compared to controls[35], suggesting a possible connection of Rab27A with regulatory mechanisms that are linked with mitochondrial damage and dysfunction. While this an interesting avenue to explore, in this paper we will not follow up further on the mechanism of mitochondrial recruitment of Rab27A. 

Author response table 1.

Rab27A interactome. Proteins co-immunoprecipitated with mScarlet-Rab27A vs mScarlet expressing control. The data show average of three replicates. 

To investigate the role of SYTL5 in the context of ACC, we acquired the NCI-H295R cell line isolated from the adrenal gland of an adrenal cancer patient. The cells were cultured as recommended from ATCC using DMEM/F-12 supplemented with NuSerum and ITS +premix. It is important to note that the H295R cells were adapted to grow as an adherent monolayer from the H295 cell line which grows in suspension. However, there can still be many viable H295R cells in the media. 

We attempted to conduct OCR and ECAR measurements using the Seahorse XF upon knockdown of SYTL5 and/or Rab27A in H295R cells. For these assays, it is essential that the cells be seeded in a monolayer at 70-90% confluency with no cell clusters[4]. Poor adhesion of the cells can cause inaccurate measurements by the analyser. Unfortunately, the results between the five replicates we carried out were highly inconsistent, the same knockdown produced trends in opposite directions in different replicates. This is likely due to problems with seeding the cells. Despite our best efforts to optimise seeding number, and pre-coating the plate with poly-D-lysine[5] we observed poor attachment of cells and inability to form a monolayer. 

To study the localisation of SYTL5 and Rab27A in an ACC model, we transduced the H295R cells with lentiviral particles to overexpress pLVX-SV40-mScarlet-I-Rab27A and pLVX-CMV-SYTL5-EGFP-3xFLAG. Again, this proved unsuccessful after numerous attempts at optimising transduction. 

These issues limited our investigation into the role of SYTL5 in ACC to the cortisol assay (Supplementary Figure 6). For this the H295R cells were an appropriate model as they are able to produce an array of adrenal cortex steroids[6] including cortisol[7]. In this assay, measurements are taken from cell culture supernatants, so the confluency of the cells does not prevent consistent results as the cortisol concentration was normalised to total protein per sample. With this assay we were able to rule out a role for SYTL5 and Rab27A in the secretion of cortisol.  

Another consideration when investigating the involvement of SYTL5 in ACC, is that in general ACC cells should have a low expression of SYTL5 as is seen from the patient expression data (Figure 6B).

The reviewer also writes “Furthermore, images and results supporting the major conclusions need to be improved.”. We have tried several times, without success, to generate U2OS cells with CRISPR/Cas9-mediated C-terminal tagging of endogenous SYTL5 with mNeonGreen, using an approach that has been successfully implemented in the lab for other genes. This is likely due to a lack of suitable sgRNAs targeting the C-terminal region of SYTL5, which have a low predicted efficiency score and a large number of predicted off-target sites in the human genome including several other gene exons and introns (see Author response image 2). 

We have also included new data (Supplementary Figure 4B) showing that some of the hypoxia-induced SYTL5-Rab27A-positive vesicles stain positive for the autophagy markers p62 and LC3B when inhibiting lysosomal degradation, further strengthening our data that SYTL5 and Rab27A function as positive regulators of mitophagy.  

Reviewer #2 (Public review): 

Summary:

The authors provide convincing evidence that Rab27 and STYL5 work together to regulate mitochondrial activity and homeostasis.

Strengths:

The development of models that allow the function to be dissected, and the rigorous approach and testing of mitochondrial activity.

Weaknesses:

There may be unknown redundancies in both pathways in which Rab27 and SYTL5 are working which could confound the interpretation of the results.

Suggestions for revision:

Given that Rab27A and SYTL5 are members of protein families it would be important to exclude any possible functional redundancies coming from Rab27B expression or one of the other SYTL family members. For Rab27 this would be straightforward to test in the assays shown in Figure 4 and Supplementary Figure 5. For SYTL5 it might be sufficient to include some discussion about this possibility.

We thank the reviewer for pointing out the potential redundancy issue for Rab27A and SYTL5. There are multiple studies demonstrating the redundancy between Rab27A and Rab27B. For example, in a study of the disease Griscelli syndrome, caused by Rab27A loss of function, expression of either Rab27A or Rab27B rescues the healthy phenotype indicating redundancy[8]. This redundancy however applies to certain function and cell types. In fact, in a study regarding hair growth, knockdown of Rab27B had the opposite effect to knockdown of Rab27A[9].

In this paper, we conducted all assays in U2OS cells, in which the expression of Rab27B is very low. Human Protein Atlas reports expression of 0.5nTPM for Rab27B, compared to 18.4nTPM for Rab27A. We also observed this low level of expression of Rab27B compared to Rab27A by qPCR in U2OS cells. Therefore, there would be very little endogenous Rab27B expression in cells depleted of Rab27A (with siRNA or KO). In line with this, Rab27B peptides were not detected in our SYTL5 interactome MS data (Table 1 in paper). Moreover, as Rab27A depletion inhibits mitochondrial recruitment of SYTL5 and mitophagy, it is not likely that Rab27B provides a functional redundancy. It is possible that Rab27B overexpression could rescue mitochondrial localisation of SYTL5 in Rab27A KO cells, but this was not tested as we do not have any evidence for a role of Rab27B in these cells. Taken together, we believe our data imply that Rab27B is very unlikely to provide any functional redundancy to Rab27A in our experiments. 

For the SYTL family, all five members are Rab27 effectors, binding to Rab27 through their SHD domain. Together with Rab27, all SYTL’s have been implicated in exocytosis in different cell types. For example, SYTL1 in exocytosis of azurophilic granules from neutrophils[10], SYTL2 in secretion of glucagon granules from pancreatic α cells[11], SYTL3 in secretion of lytic granules from cytotoxic T lymphocytes[12], SYTL4 in exocytosis of dense hormone containing granules from endocrine cells[13] and SYTL5 in secretion of the RANKL cytokine from osteoblasts[14]. This indicates a potential for redundancy through their binding to Rab27 and function in vesicle secretion/trafficking. However, one study found that different Rab27 effectors have distinct functions at different stages of exocytosis[15].

Very little known about redundancy or hierarchy between these proteins. Differences in function may be due to the variation in gene expression profile across tissues for the different SYTL’s (see Author response image 1 below). SYTL5 is enriched in the brain unlike the others, suggesting possible tissue specific functions. There are also differences in the binding affinities and calcium sensitivities of the C2iA and C2B domains between the SYTL proteins[16].

Author response image 1.

GTEx Multi Gene Query for SYTL1-5

All five SYTL’s are expressed in the U2OS cell line with nTPMs according to Human Protein Atlas of SYTL1: 7.5, SYTL2: 13.4, SYTL3:14.2, SYTL4: 8.7, SYTL5: 4.8. In line with this, in the Rab27A interactome, when comparing cells overexpressing mScarlet-Rab27A with control cells, we detected all five SYTL’s as specific Rab27A-interacting proteins (see Author response table 1 above). Whereas, in the SYTL5 interactome we did not detect any other SYTL protein (table 1 in paper), confirming that they do not form a complex with SYTL5. 

We have included the following text in the discussion (p. 12): “SYTL5 and Rab27A are both members of protein families, suggesting possible functional redundancies from Rab27B or one of the other SYTL isoforms. While Rab27B has a very low expression in U2OS cells, all five SYTL’s are expressed. However, when knocking out or knocking down SYTL5 and Rab27A we observe significant effects that we presume would be negated if their isoforms were providing functional redundancies. Moreover, we did not detect any other SYTL protein or Rab27B in the SYTL5 interactome, confirming that they do not form a complex with SYTL5.”

Suggestions for Discussion: 

Both Rab27A and STYL5 localize to other membranes, including the endolysosomal compartments. How do the authors envisage the mechanism or cellular modifications that allow these proteins, either individually or in complex to function also to regulate mitochondrial funcYon? It would be interesYng to have some views.

We agree that it would be interesting to better understand the mechanism involved in modulation of the localisation and function of SYTL5 and Rab27A at different cellular compartments, including the mitochondria. Here, we have shown that SYTL5 recruitment to the mitochondria involves coincidence detection, as both its lipid-binding C2 domains and the Rab27A-binding SHD domain are required (Figure 1G-H). Both these domains also seem required for localisation of SYTL5 to vesicles, and we can only speculate that binding to different lipids (Figure 1F) may regulate SYTL5 localisation. Additionally, we find that mitochondrial recruitment of SYTL5 is dependent on the GTPase activity and mitochondrial localisation of Rab27A (Figure 2D-E). However, this seems also the case for vesicular recruitment of SYTL5, although a few SYTL5-Rab27A (T23N) positive vesicles were seen (Figure 2E). 

To characterise the mechanisms involved in mitochondrial localisation of Rab27A, we have performed mass spectrometry analysis to identify the interactome of Rab27A (see Author response table 1 above). U2OS cells with stable expression of mScarlet-Rab27A or mScarlet only were subjected to immunoprecipitation, followed by MS analysis.  Of the 32 significant Rab27A-interacting hits (compared to control), two of the hits localise in the inner mitochondrial membrane (IMM); ATP synthase F(1) complex subunit alpha (P25705), and mitochondrial very long-chain specific acyl-CoA dehydrogenase (VLCAD)(P49748). However, as these IMM proteins are not likely involved in mitochondrial recruitment of Rab27A, observed under basal conditions, we chose not to include these data in the manuscript. 

It is known that other RAB proteins are recruited to the mitochondria by regulation of their GTPase activity. During parkin-mediated mitophagy, RABGEF1 (a guanine nucleotide exchange factor) is recruited through its ubiquitin-binding domain and directs mitochondrial localisation of RAB5, which subsequently leads to recruitment of RAB7 by the MON1/CCZ1 GEF complex[1]. As already mentioned in the discussion (p.12), ubiquitination of the Rab27A GTPase activating protein alpha (TBC1D10A) is reduced in the brain of Parkin KO mouse compared to controls[35], suggesting a possible connection of Rab27A with regulatory mechanisms that are linked with mitochondrial damage and dysfunction. While this an interesting avenue to explore, it is beyond the scope of this paper. 

Our data suggest that SYTL5 functions as a negative regulator of the Warburg effect, the switch from OXPHOS to glycolysis. While both SYTL5 and Rab27A seem required for mitophagy of selective mitochondrial components, and their depletion leading to reduced mitochondrial respiration and ATP production, only depletion of SYTL5 caused a switch to glycolysis. The mechanisms involved are unclear, but we found several proteins linked to the cellular response to oxidative stress, reactive oxygen species metabolic process, regulation of mitochondrion organisation and protein insertion into mitochondrial membrane to be enriched in the SYTL5 interactome (Figure 3A and C).

We have addressed this comment in the discussion on p.12 

Reviewer #3 (Public review):

Summary:

In the manuscript by Lapao et al., the authors uncover a role for the Rab27A effector protein SYTL5 in regulating mitochondrial function and turnover. The authors find that SYTL5 localizes to mitochondria in a Rab27A-dependent way and that loss of SYTL5 (or Rab27A) impairs lysosomal turnover of an inner mitochondrial membrane mitophagy reporter but not a matrix-based one. As the authors see no co-localization of GFP/mScarlet tagged versions of SYTL5 or Rab27A with LC3 or p62, they propose that lysosomal turnover is independent of the conventional autophagy machinery. Finally, the authors go on to show that loss of SYTL5 impacts mitochondrial respiration and ECAR and as such may influence the Warburg effect and tumorigenesis. Of relevance here, the authors go on to show that SYTL5 expression is reduced in adrenocortical carcinomas and this correlates with reduced survival rates.

Strengths:

There are clearly interesting and new findings here that will be relevant to those following mitochondrial function, the endocytic pathway, and cancer metabolism.

Weaknesses:

The data feel somewhat preliminary in that the conclusions rely on exogenously expressed proteins and reporters, which do not always align.

As the authors note there are no commercially available antibodies that recognize endogenous SYTL5, hence they have had to stably express GFP-tagged versions. However, it appears that the level of expression dictates co-localization from the examples the authors give (though it is hard to tell as there is a lack of any kind of quantitation for all the fluorescent figures). Therefore, the authors may wish to generate an antibody themselves or tag the endogenous protein using CRISPR.

We agree that the level of SYTL5 expression is likely to affect its localisation. As suggested by the reviewer, we have tried hard, without success, to generated U2OS cells with CRISPR knock-in of a mNeonGreen tag at the C-terminus of endogenous SYTL5, using an approach that has been successfully implemented in the lab for other genes. This is likely due to a lack of suitable sgRNAs targeting the C-terminal region of SYTL5, which have a low predicted efficiency score and a large number of predicted off-target sites in the human genome including several other gene exons and introns (see Author response image 2). 

Author response image 2.

Overview of sgRNAs targeting the C-terminal region of SYTL5 

Although the SYTL5 expression level might affect its cellular localization, we also found the mitochondrial localisation of SYTL5-EGFP to be strongly increased in cells co-expressing mScarletRab27A, supporting our findings of Rab27A-mediated mitochondrial recruitment of SYTL5. We have also included new data (Supplementary Figure 4B) showing that some of the hypoxia-induced SYTL5Rab27A-positive vesicles stain positive for the autophagy markers p62 and LC3B when inhibiting lysosomal degradation, further strengthening our data that SYTL5 and Rab27A function as positive regulators of mitophagy.  

In relation to quantitation, the authors found that SYTL5 localizes to multiple compartments or potentially a few compartments that are positive for multiple markers. Some quantitation here would be very useful as it might inform on function. 

We find that SYTL5-EGFP localizes to mitochondria, lysosomes and the plasma membrane in U2OS cells with stable expression of SYTL5-EGFP and in SYTL5/Rab27A double knock-out cells rescued with SYTL5EGFP and mScralet-Rab27A. We also see colocalization of SYTL5-EGFP with endogenous p62, LC3 and LAMP1 upon induction of mitophagy. However, as these cell lines comprise a heterogenous pool with high variability we do not believe that quantification of the overexpressing cell lines would provide beneficial information in this scenario. As described above, we have tried several times to generate SYTL5 knock-in cells without success.  

The authors find that upon hypoxia/hypoxia-like conditions that punctate structures of SYTL5 and Rab27A form that are positive for Mitotracker, and that a very specific mitophagy assay based on pSu9-Halo system is impaired by siRNA of SYTL5/Rab27A, but another, distinct mitophagy assay (Matrix EGFP-mCherry) shows no change. I think this work would strongly benefit from some measurements with endogenous mitochondrial proteins, both via immunofluorescence and western blot-based flux assays. 

In addition to the western blotting for different endogenous ETC proteins showing significantly increased levels of MTCO1 in cells depleted of SYTL5 and/or Rab27A (Figure 5E-F), we have now blotted for the endogenous mitochondrial proteins, COXIV and BNIP3L, in DFP and DMOG conditions upon knockdown of SYTL5 and/or Rab27A (Figure 5G and Supplementary Figure 5A). Although there was a trend towards increased levels, we did not see any significant changes in total COXIV or BNIP3L levels when SYTL5, Rab27A or both are knocked down compared to siControl. Blotting for endogenous mitochondrial proteins is however not the optimum readout for mitophagy. A change in mitochondrial protein level does not necessarily result from mitophagy, as other factors such as mitochondrial biogenesis and changes in translation can also have an effect. Mitophagy is a dynamic process, which is why we utilise assays such as the HaloTag and mCherry-EGFP double tag as these indicate flux in the pathway. Additionally, as mitochondrial proteins have different half-lives, with many long-lived mitochondrial proteins[17], differences in turnover rates of endogenous proteins make the results more difficult to interpret. 

A really interesting aspect is the apparent independence of this mitophagy pathway on the conventional autophagy machinery. However, this is only based on a lack of co-localization between p62or LC3 with LAMP1 and GFP/mScarlet tagged SYTL5/Rab27A. However, I would not expect them to greatly colocalize in lysosomes as both the p62 and LC3 will become rapidly degraded, while the eGFP and mScarlet tags are relatively resistant to lysosomal hydrolysis. -/+ a lysosome inhibitor might help here and ideally, the functional mitophagy assays should be repeated in autophagy KOs. 

We thank the reviewer for this suggestion. We have now repeated the colocalisation studies in cells treated with DFP with the addition of bafilomycin A1 (BafA1) to inhibit the lysosomal V-ATPase. Indeed, we find that a few of the SYTL5/Rab27A/MitoTracker positive structures also stain positive for p62 and LC3 (Supplementary Figure 4B). As expected, the occurrence of these structures was rare, as BafA1 was only added for the last 4 hrs of the 24 hr DFP treatment. However, we cannot exclude the possibility that there are two different populations of these vesicles.

The link to tumorigenesis and cancer survival is very interesYng but it is not clear if this is due to the mitochondrially-related aspects of SYTL5 and Rab27A. For example, increased ECAR is seen in the SYTL5 KO cells but not in the Rab27A KO cells (Fig.5D), implying that mitochondrial localization of SYTL5 is not required for the ECAR effect. More work to strengthen the link between the two sections in the paper would help with future direcYons and impact with respect to future cancer treatment avenues to explore. 

We agree that the role of SYTL5 in ACC requires future investigation. While we observe reduced OXPHOS levels in both SYTL5 and Rab27A KO cells (Figure 5B), glycolysis was only increased in SYTL5 KO cells (Figure 5D). We believe this indicates that Rab27A is being negatively regulated by SYTL5, as ECAR was unchanged in both the Rab27A KO and Rab27A/SYTL5 dKO cells. This suggests that Rab27A is required for the increase in ECAR when SYTL5 is depleted, therefore SYTL5 negatively regulates Rab27A. The mechanism involved is unclear, but we found several proteins linked to the cellular response to oxidative stress, reactive oxygen species metabolic process, regulation of mitochondrion organisation and protein insertion into mitochondrial membrane to be enriched in the SYTL5 interactome (Figure 3A and C).

To investigate the link to cancer further, we tested the effect of knockdown of SYTL5 and/or Rab27A on the levels of mitochondrial ROS. ROS levels were measured by flow cytometry using the MitoSOX Red dye, together with the MitoTracker Green dye to normalise ROS levels to the total mitochondria. Cells were treated with the antioxidant N-acetylcysteine (NAC)[18] as a negative control and menadione as a positive control, as menadione induces ROS production via redox cycling[19]. We must consider that there is also a lot of autofluorescence from cells that makes it impossible to get a level of ‘zero ROS’ in this experiment. We did not see a change in ROS with knockdown of SYTL5 and/or Rab27A compared to the NAC treated or siControl samples (see Author response image 3 below). The menadione samples confirm the success of the experiment as ROS accumulated in these cells. Thus, based on this, we do not believe that low SYTL5 expression would affect ROS levels in ACC tumours.

Author response image 3.

Mitochondrial ROS production normalised to total mitochondria

As discussed in our response to Reviewer #1, we tried hard to characterise the role of SYTL5 in the context of ACC using the NCI-H295R cell line isolated from the adrenal gland of an adrenal cancer patient. We attempted to conduct OCR and ECAR measurements using the Seahorse XF upon knockdown of SYTL5 and/or Rab27A in H295R cells without success, due to poor attachment of the cells and inability to form a monolayer. We also transduced the H295R cells with lentiviral particles to overexpress pLVX-SV40-mScarlet-I-Rab27A and pLVX-CMV-SYTL5-EGFP-3xFLAG to study the localisation of SYTL5 and Rab27A in an ACC model. Again, this proved unsuccessful after numerous attempts at optimising the transduction. These issues limited our investigation into the role of SYTL5 in ACC to the cortisol assay (Supplementary Figure 6). For this the H295R cells were an appropriate model as they are able to produce an array of adrenal cortex steroids[6] including cortisol[7] In this assay, measurements are taken from cell culture supernatants, so the confluency of the cells does not prevent consistent results as the cortisol concentration was normalised to total protein per sample. With this assay we were able to rule out a role for SYTL5 and Rab27A in the secretion of cortisol.  

Another consideration when investigating the involvement of SYTL5 in ACC, is that in general ACC cells should have a low expression of SYTL5 as is seen from the patient expression data (Figure 6B).

Further studies into the link between SYTL5/Rab27A and cancer are beyond the scope of this paper as we are limited to the tools and expertise available in the lab.

References

(1) Yamano, K. et al. Endosomal Rab cycles regulate Parkin-mediated mitophagy. eLife 7 (2018). https://doi.org:10.7554/eLife.31326

(2) Carré, M. et al. Tubulin is an inherent component of mitochondrial membranes that interacts with the voltage-dependent anion channel. The Journal of biological chemistry 277, 33664-33669 (2002). https://doi.org:10.1074/jbc.M203834200

(3) Hoogerheide, D. P. et al. Structural features and lipid binding domain of tubulin on biomimetic mitochondrial membranes. Proceedings of the National Academy of Sciences 114, E3622-E3631 (2017). https://doi.org:10.1073/pnas.1619806114

(4) Plitzko, B. & Loesgen, S. Measurement of Oxygen Consumption Rate (OCR) and Extracellular Acidification Rate (ECAR) in Culture Cells for Assessment of the Energy Metabolism. Bio Protoc 8, e2850 (2018). https://doi.org:10.21769/BioProtoc2850

(5) Yavin, E. & Yavin, Z. Attachment and culture of dissociated cells from rat embryo cerebral hemispheres on polylysine-coated surface. The Journal of cell biology 62, 540-546 (1974). https://doi.org:10.1083/jcb.62.2.540

(6) Wang, T. & Rainey, W. E. Human adrenocortical carcinoma cell lines. Mol Cell Endocrinol 351, 5865 (2012). https://doi.org:10.1016/j.mce.2011.08.041

(7) Rainey, W. E. et al. Regulation of human adrenal carcinoma cell (NCI-H295) production of C19 steroids. J Clin Endocrinol Metab 77, 731-737 (1993). https://doi.org:10.1210/jcem.77.3.8396576

(8) Barral, D. C. et al. Functional redundancy of Rab27 proteins and the pathogenesis of Griscelli syndrome. J. Clin. Invest. 110, 247-257 (2002). https://doi.org:10.1172/jci15058

(9) Ku, K. E., Choi, N. & Sung, J. H. Inhibition of Rab27a and Rab27b Has Opposite Effects on the Regulation of Hair Cycle and Hair Growth. Int. J. Mol. Sci. 21 (2020). https://doi.org:10.3390/ijms21165672

(10) Johnson, J. L., Monfregola, J., Napolitano, G., Kiosses, W. B. & Catz, S. D. Vesicular trafficking through cortical actin during exocytosis is regulated by the Rab27a effector JFC1/Slp1 and the RhoA-GTPase–activating protein Gem-interacting protein. Mol. Biol. Cell 23, 1902-1916 (2012). https://doi.org:10.1091/mbc.e11-12-1001

(11) Yu, M. et al. Exophilin4/Slp2-a targets glucagon granules to the plasma membrane through unique Ca2+-inhibitory phospholipid-binding activity of the C2A domain. Mol. Biol. Cell 18, 688696 (2007). https://doi.org:10.1091/mbc.e06-10-0914

(12) Kurowska, M. et al. Terminal transport of lyXc granules to the immune synapse is mediated by the kinesin-1/Slp3/Rab27a complex. Blood 119, 3879-3889 (2012). https://doi.org:10.1182/blood-2011-09-382556

(13) Zhao, S., Torii, S., Yokota-Hashimoto, H., Takeuchi, T. & Izumi, T. Involvement of Rab27b in the regulated secretion of pituitary hormones. Endocrinology 143, 1817-1824 (2002). https://doi.org:10.1210/endo.143.5.8823

(14) Kariya, Y. et al. Rab27a and Rab27b are involved in stimulation-dependent RANKL release from secretory lysosomes in osteoblastic cells. J Bone Miner Res 26, 689-703 (2011). https://doi.org:10.1002/jbmr.268

(15) Zhao, K. et al. Functional hierarchy among different Rab27 effectors involved in secretory granule exocytosis. Elife 12 (2023). https://doi.org:10.7554/eLife.82821

(16) Izumi, T. Physiological roles of Rab27 effectors in regulated exocytosis. Endocr J 54, 649-657 (2007). https://doi.org:10.1507/endocrj.kr-78

(17) Bomba-Warczak, E. & Savas, J. N. Long-lived mitochondrial proteins and why they exist. Trends in cell biology 32, 646-654 (2022). https://doi.org:10.1016/j.tcb.2022.02.001

(18) Curtin, J. F., Donovan, M. & Cotter, T. G. Regulation and measurement of oxidative stress in apoptosis. Journal of Immunological Methods 265, 49-72 (2002). https://doi.org:https://doi.org/10.1016/S0022-1759(02)00070-4

(19) Criddle, D. N. et al. Menadione-induced Reative Oxygen Species Generation via Redox Cycling Promotes Apoptosis of Murine Pancreatic Acinar Cells. Journal of Biological Chemistry 281, 40485-40492 (2006). https://doi.org:https://doi.org/10.1074/jbc.M607704200

eLife Assessment

This work provides one of the first important attempts to look at Drosophila immune responses against bacterial, viral, and fungal pathogens in a way that combines the roles of four major arms in immunity (Imd signaling, Toll signaling, phagocytosis, and melanization) rather than studying them separately. The findings are compelling and the tools provided can be used as they are, or built upon, in various contexts.

Reviewer #1 (Public review):

Summary:

The innate immune system serves as the first line of defense against invading pathogens. Four major immune-specific modules-the Toll pathway, the Imd pathway, melanization, and phagocytosis-play critical roles in orchestrating the immune response. Traditionally, most studies have focused on the function of individual modules in isolation. However, in recent years, it has become increasingly evident that effective immune defense requires intricate interactions among these pathways.

Despite this growing recognition, the precise roles, timing, and interconnections of these immune modules remain poorly understood. Moreover, addressing these questions represents a major scientific undertaking.

Strengths:

In this manuscript, Ryckebusch et al. systematically evaluate both the individual and combined contributions of these four immune modules to host defense against a range of pathogens. Their findings significantly enhance our understanding of the layered architecture of innate immunity.

Reviewer #2 (Public review):

Summary:

In this work, the authors take a holistic view at the Drosophila immunity by selecting four major components of fly immunity often studied separately (Toll signaling, Imd signaling, phagocytosis and melanization), and studying their combinatory effects on the efficiency of the immune response. They achieve this by using fly lines mutant for one of these components, or modules, as well as for a combination of them, and testing the survival of these flies upon infection with a plethora of pathogens (bacterial, viral and fungal).

Strengths:

It is clear that this manuscript has required a large amount of hands-on work, considering the number of pathogens, mutations and timepoints tested. In my opinion, this work is a very welcome addition to the literature on fly immune responses, which obviously do not occur one type of a response at a time, but in parallel, subsequently and/or are interconnected. I find that the major strength of this work is the overall concept, which is made possible by the mutations designed to target the specific immune function of each module, without effects on other functions. I believe that the combinatory mutants will be of use for the fly community and enable further studies of interplay of these components of immune response in various settings.

To control for the effects arising from the genetic variation other than the intended mutations, the mutants have been backcrossed into a widely used, isogenized Drosophila strain called w1118. Therefore, the differences accounted for by the genotype are controlled.

I also appreciate that the authors have investigated the two possible ways of dealing with an infection: tolerance and resistance, and how the modules play into those.

Weaknesses:

While controlling for the background effects is vital, the w1118 background is problematic (an issue not limited to this manuscript) because of the wide effects of the white mutation on several phenotypes (also other than eye color/eyesight). It is a possibility that the mutation influences the functionality of the immune response components. I acknowledge that it is not reasonable to ask for data in different backgrounds better representing a "wild type" fly, but I think this matter should be brought up and discussed.

The whole study has been conducted on male flies. Immune responses show quite extensive sex-specific variation across a variety of species studied, also in the fly. But the reasons for this variation are not fully understood. Therefore, I suggest that the authors would conduct a subset of experiments on female flies to see if the findings apply to both sexes, especially the infection-specificity of the module combinations.

Comments on the revised manuscript:

I appreciate the author's responses to the points I raised and the additional work they have conducted. The authors have now discussed the possible background effect and added an experiment on female flies showing that the module function is applicable to both sexes.

Author response:

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

Reviewer #1 (Public review): 

Summary: 

The innate immune system serves as the first line of defense against invading pathogens. Four major immune-specific modules - the Toll pathway, the Imd pathway, melanization, and phagocytosis- play critical roles in orchestrating the immune response. Traditionally, most studies have focused on the function of individual modules in isolation. However, in recent years, it has become increasingly evident that effective immune defense requires intricate interactions among these pathways. 

Despite this growing recognition, the precise roles, timing, and interconnections of these immune modules remain poorly understood. Moreover, addressing these questions represents a major scientific undertaking. 

Strengths: 

In this manuscript, Ryckebusch et al. systematically evaluate both the individual and combined contributions of these four immune modules to host defense against a range of pathogens. Their findings significantly enhance our understanding of the layered architecture of innate immunity. 

We thank the reviewer for their kind assessment.

Weaknesses: 

While I have no critical concerns regarding the study, I do have several suggestions to offer that may help further strengthen the manuscript. These include: 

(1) Have the authors validated the efficiency of the mutants used in this study? It would be helpful to include supporting data or references confirming that the mutations effectively disrupted the intended immune pathways. 

We have done so in Figure 1.

(2) Given the extensive use of double, triple, and quadruple mutants, a more detailed description of the mutant construction process is warranted. 

We now provide a supplement (File S1) that details the successive genetic crosses and recombinations that were required to generate these compound fly stocks carrying multiple mutations. We also provide some information regarding rapid screening of stocks for phenotypes. Of note some of these fly stocks have been deposited at VDRC as they will be useful to fly community to assess immune modules in a controlled background, and complete stock information will be tied to these stocks there.

Reviewer #2 (Public review): 

Summary: 

In this work, the authors take a holistic view of Drosophila immunity by selecting four major components of fly immunity often studied separately (Toll signaling, Imd signaling, phagocytosis, and melanization), and studying their combinatory effects on the efficiency of the immune response. They achieve this by using fly lines mutant for one of these components, or modules, as well as for a combination of them, and testing the survival of these flies upon infection with a plethora of pathogens (bacterial, viral, and fungal). 

Strengths: 

It is clear that this manuscript has required a large amount of hands-on work, considering the number of pathogens, mutations, and timepoints tested. In my opinion, this work is a very welcome addition to the literature on fly immune responses, which obviously do not occur in one type of response at a time, but in parallel, subsequently, and/or are interconnected. I find that the major strength of this work is the overall concept, which is made possible by the mutations designed to target the specific immune function of each module (at least seemingly) without major effects on other functions. I believe that the combinatory mutants will be of use for the fly community and enable further studies of the interplay of these components of immune response in various settings. 

To control for the effects arising from the genetic variation other than the intended mutations, the mutants have been backcrossed into a widely used, isogenized Drosophila strain called w1118. Therefore, the differences accounted for by the genotype are controlled. 

I also appreciate that the authors have investigated the two possible ways of dealing with an infection: tolerance and resistance, and how the modules play into those. 

We thank the reviewer for their kind assessment. 

Weaknesses: 

While controlling for the background effects is vital, the w1118 background is problematic (an issue not limited to this manuscript) because of the wide effects of the white mutation on several phenotypes (also other than eye color/eyesight). It is a possibility that the mutation influences the functionality of the immune response components, for example, via effects of the faulty tryptophan handling on the metabolism of the animal. 

I acknowledge that it is not reasonable to ask for data in different backgrounds better representing a "wild type" fly (however, that is defined is another question), but I think this matter should be brought up and discussed. 

We agree with the reviewer and have included caveats on the different genetic effects brought about the combinatory mutant approach including differences in white gene status, insertion of GFP or DsRed markers, and nature of genetic mutations (Line 142-on).

“Of note, the strains used in this study differ in their presence/absence of the white⁺ gene, present in the PPO1^∆, NimC1¹ and eater¹ mutations.  In addition to its well established function in eye pigmentation, the white gene can also impact host neurology and intestinal stem cell proliferation (Ferreiro et al., 2017; Sasaki et al., 2021). We did not observe any obvious correlations between white⁺ gene status and susceptibilities in this study. Moreover,  in a previous study looking at the cumulative effects of AMP mutations on lifespan, white gene status and fluorescent markers did not readily explain differences in longevity (Hanson and Lemaitre, 2023). We therefore believe that the extreme immune susceptibility we have created through deficiencies for pathways regulating hundreds of genes, or major immune modules, overwhelms the potential effects of white⁺ and other transgenic markers. For additional information on which stocks bear which markers, see discussion in Supplementary file 1.”

Of interest, we were highly conscious of this concern in working with combinatory AMP mutants which differed in white, GFP, and DsRed copies. However, even over the many weeks of snowballing effects on microbiota community composition and structure, we found no trends tied strictly to white+ or to other genetic insertions on lifespan (Hanson and Lemaitre, 2023; DMM).

The whole study has been conducted on male flies. Immune responses show quite extensive sex-specific variation across a variety of species studied, also in the fly. But the reasons for this variation are not fully understood. Therefore, I suggest that the authors conduct a subset of experiments on female flies to see if the findings apply to both sexes, especially the infection-specificity of the module combinations.  

We thank the reviewer for this suggestion. We have performed the requested experiments, and include female survival trends in Figure 4supp1. We have added the following text to the main manuscript (Line 554):

“All survival experiments to this point were done with males. We therefore assessed key survival trends for these infections in females to learn whether the dynamics we observed were consistent across sexes (Figure 4supp1). For all three pathogens (Pr rettgeri, Sa aureus, C. albicans) the rank order of susceptibility was broadly similar between males and females, with higher rates of mortality in females overall. Thus, we found no marked sex-bygenotype interaction. Interestingly, the greater susceptibility of females in our hands is true even for ∆ITPM flies, although there are only a few surviving flies on which we can base these conclusions. However, these data may suggest the sexual dimorphism in defense against infection that we see against these pathogens is due to factors independent of the immune modules we disrupted.”

It is worth noting that male-female sex dichotomies in infection are inconsistent across the literature, with strong lab-specific effects (Belmonte et al., 2020 and personal observation). In our lab setting, we consistently see female mortality higher than males when compared, independent of pathogen and mutant background. We have not seen notable interaction terms of sex and genotype for most immune deficient mutants. It is quite interesting to have done these experiments with ITPM, however, which reveals that there is at least a trend suggesting this dichotomy is independent of the four immune modules we deleted. Still, our infection conditions kill most males, and so it would be good to replicate this sex-specific ∆ITPM result in a dedicated study with doses chosen to improve the resolution of male-female differences. For now, we prefer to use conservative language and avoid overinterpreting this trend, but do feel it merits mentioning.  

Recommendations for the authors:

Comment on statistical requests

Both reviewers requested further clarity on the statistical analyses supplemental to Figure 3. We haved address these comments as follows.

First, we now provide an additional supplementary .zip file containing summary statistics for all survival data in Figure 3 (Supplementary File 3). We have additionally added this text to line 226 to make this data treatment more clear:

…” we chose to focus on major differences apparent in summary statistics,Highlighting”…

And we highlight that all survival data are also provided as Kaplan-Meier survival curves in the main or supplementary figures in Line 233:

“Kaplan-Meier survival curves for all experiments are provided in the main text or supplementary information”.

Second, as outlined in the main text, we were unable to sample across all pathogenby-genotype interactions systematically, and this unfortunately obfuscates robust statistical modelling. We addressed the challenge of finding meaningful statistical differences by focusing on trends only if they were i) consistent across experimental replicates, ii) of a consistent logic across comparable genotypes, ensuring random inter-experimental noise was not unduly shaping interpretations, and iii) of a mean lifespan difference ≥1.0 days compared to wild-type, and compared to relevant unchallenged or clean-injury controls. This last choice was especially important because not all experimental replicates included all genotypes due to challenges of animal husbandry and coordination among multiple researchers over five years of data collection. As a result, our initial analyses using a cox mixed-effects model found it to be rather useless, being insensitive to important experiment batch effects visible to the eye because statistically-affected genotypes were not present in all experiments.

We therefore ensured that behaviour relative to controls within* experiments was consistent, rather than the comparison of genotypes to controls across the sum of experiments with a post-hoc treatment attempting to apportion variance to experiment batch (but unable to do so for some genotypes and some batches). Due to differeces in baseline health and the dynamics explained by studies like Duneau et al. (2017; eLife, there is an expected unequal variance of genotype*pathogen interactions across experiment batches. Unfortunately, this unequal variance, coupled with incomplete sampling across experiment batches, means “highly significant” differences can emerge that don’t hold up to scrutiny of comparisons to controls taken only from within an experiment batch. Thus, we chose to forego a cox mixed effect model approach entirely. Instead, our highly conservative approach, focusing on only very large effects with a mean lifespan difference ≥1.0 days, mitigates these issues. We have taken great care to ensure that any results we highlight stand up to inter-experiment batch effects. We would further draw the reviewers’ attention to our response to Reviewer 2 relating to Figure 3, which emphasizes the level of conservativism that we are applying.

At the end of the Discussion, we have added the following sentence to emphasize these limitations:

“…a combinatorial mutation approach to deciphering immune function can be extended even to the broad level of whole immune modules. Of note, we were unable to systematically sample all genotype-bypathogen interactions equally. We have therefore been highly conservative in our reporting of major effects. There are likely many important interactions” not discussed in our study. Future investigations may highlight important biology that is apparent in our data, but which we may not have mentioned here. To this end, we have deposited our isogenic immunity fly stocks in the Vienna Drosophila Resource Centre to facilitate their use. Beyond immunity, our tools can also be of use to study various questions at the cutting edge of aging, memory, neurodegeneration, cancer, and more, where immune genes are repeatedly implicated. We hope that this set of lines will be useful to the community to better characterize the Drosophila host defense.”

We recognise this response may not fully satisfy the reviewers’ requests. While use of summary statistics is simple, our rules for highlighting interactions of importance are defined, readily understood and interpreted, and draw attention to key trends in that are backed by a solid understanding of the data and its limitations. We have taken this approach out of a responsibility to avoid making spurious assertions that stem from underpowered statistical models rather than from the biology itself.

Reviewer #1 (Recommendations for the authors): 

(1) Lines 1092-1093 - Please double-check the labeling of the panels in Figure 2. It appears that panels A and C correspond to single-module mutants, whereas panels B and D refer to compound-module mutants. 

We have modified Figure 2 and Figure 2supp1 labelling. We also realise there was an error in the column titling that contributed to the confusion. We hope the new layout is clear, and thank the reviewers for noting this issue.

(2) Lines 347-377 - Figure 2D is not cited in the text. 

We now cite Fig2D in Line 356.

(3) P values should be indicated in Figure 2 and Figure 3 for all relevant comparisons. Additionally, "ns" (not significant) should be added in Figure 5A-B. 

We make the effort to show key uninfected survival trends in Figure 2, and list the total flies (n_flies) in Fig3 to provide the reader with the underlying confidence in the trends observed. We focus on differences of mean lifespan of at least 1 day, and which are consistent in direction across combinatory mutations.  We have avoided the multiple comparisons of cox proportional hazard survival analyses throughout this study because they are overly sensitive for our purposes, as we have previously when systematically comparing many genotypes to each other (see Hanson and Lemaitre, 2023; DMM).

(4) Minor points: Hml-Gal4, UAS-GFP should be italic; Line 192-- "uL" and "uM"; Line 596: P>.05.

We have made these changes. We’re unsure what the comment regarding P>.05 referred to, but have removed spaces and made it non-italics. 

Reviewer #2 (Recommendations for the authors): 

Statistical analyses and their outcomes are clearly indicated only for the data in Figure 1 and Figure 5 and in the supplement for Figure 1, while they are not reported/not easily accessible for other data. For the main figures, statistics should be indicated in the figure for an easier assessment of the data. In case of multiple comparisons potentially crowding the plots too much, statistics may be in a supplementary file/table. 

See response above.

In case of the hemocytes, besides phagocytosis, I would think that ROS generation via the DUOX/NOX system is also an integral part of the immune response against pathogens, and that has not been included here. That might be an interesting addition for future experiments. As the NimC1, eater double mutant flies are said to have fewer hemocytes, it is possible that this function of the hemocytes is affected as well. This could be commented on in the text. 

The reviewer raises a good point. The role of DUOX and NOX in ROS responses is not assessed in our study. To our knowledge, DUOX and NOX participate primarily in the wound repair response, or in epithelial renewal at damage sites or in the gut. In our study on systemic immunity, we did not assess the role of clotting, the precise function of ROS, and we have missed other host defense or stress response mechanisms as well (e.g. constitutively-expressed AMP-like genes, TEPs, JAK-STAT) that likely play a role in the systemic immune defense. Considering the lethality caused by Nox and Duox mutation, there would be inherent genetic difficulties to recombine these as multiple mutations. Unfortunately, this makes it  difficult to include these processes in our analysis in a systematic manner.  We are already happy to have generated fly lines lacking four immune modules simultaneously, even if they are not fully immune deficient. We have mentioned this point in the discussion (Line 613-on).

Of note, the NimC1, eater double mutants actually have decreased hemocyte counts at the adult stage (Melcarne et al,. 2019). Thus NimC1, eater double mutants are not impaired only in phagocytosis, but the overall cellular response. We make a point to outline this in Line 225-257, and 607.

I think it could be mentioned that the melanization response at larval stage (against parasitoids) functions differently from the melanization described here (requiring hemocyte differentiation and PPO3).

A good point. We have added this mention in Line 97:

“In addition, a third PPO gene (PPO3) is specifically expressed by lamellocytes, specialized hemocytes that differentiate in larvae responding to and enveloping invading parasites (Dudzic et al., 2015)”.

Overall, the clarity of the figures and figure legends could be worked on to make them a bit easier to follow. Below are some of my suggestions: 

(1) In Figure 2, adding headings to parts C & D (similarly to A & B) would make it easier to follow what is happening in the figure at a glance. Also, it is rather difficult to visually follow which strain is which in the plots. I'd suggest adding the key/legend for single mutants below 2A & B, and the key for the double mutants below C & D. If a mutant is present in A & B and in C & D, it could be included in both keys. I also think that it would be intuitive to present the single mutants by dashed lines and double mutants by continuous lines (or vice versa), so that one would easily distinguish between them. Of note, the figure legend says that A & B are single mutants, but for example in B there are also some double mutants (?). 

We have modified Figure 2 and Figure 2supp1 labelling. We also realise there was an error in the column titling that contributed to the confusion. We hope the new layout is clear, and thank the reviewers for noting this issue.

(2) In Figure 3, it looks like ΔMel is almost identical to controls in the clean injury survival, but in Figure 2C, it is clearly doing worse. I might be missing something here, but would like the authors to clarify the matter. Also, the meaning of the numbers in the heat map could be explained in the figure legend and/or added to the figure (color key). 

The reviewer is correct. We thank the reviewer for this astute observation. Inadvertently, we used an old version of the Figure 2 preparation where only a subset of experiments was entered in the Prism data file rather than the total data used to inform Figure 3. This issue affected all genotypes.

We have reviewed the data in Figure 2, Figure 2supp1, and Figure 3, and updated these figures accordingly to ensure they represent the full survival data. We have also incorporated new experiments into the sum data related to male-female differences and to fill gaps in the data from the 1^st submission. We will also note due to the nature of 1^st decimal rounding that the difference between WT and ΔMel appears slightly underrepresented: the true difference (over the 7-day lifespan) is 0.37. We’ve provided a version of this figure rounded to 2 decimal places below, but prefer the simpler 1 decimal place in the main text for readability. The updated Figure 2 shows the full data in Figure 3 accurately.

We will also take this opportunity to highlight how conservative our ≥1.0 days difference approach is. Breaking down survival curve patterns in Figure 2 relative to mean differences in Figure 3, for clean injury, approximately ~75% of ΔMel flies survive to day 7 with mortality mostly taking place between days 3-7. The result is a mean lifespan of 6.37 days. On a survival curve, this difference appears quite strong, but in our mean lifespan table the difference is rather muted (WT vs. ΔMel difference = 0.37 days). Thus, differences of ≥1.0 days reflect very strong trends in survival data that are near-guaranteed to be independent of experimental noise. While we note issues that prevented us from a fully systematic sampling for all experiments, we are confident that the ≥1.0 day differences we highlight, using the rules explained in the main text, are robust. While this approach could be seen as overly conservative, it is our preference in this initial study, containing combinations of 25 treatments and 14 genotypes, to be highly conservative. Future studies may investigate other strong differences we have not highlighted, and the data we provide here can help generate expectations and guide those studies.

Author response image 1.

Figure 3 with 2 decimals places of rounding for mean lifespans. The 7-day clean injury mean lifespan of WT is 6.74 days, and of ΔMel is 6.37 days. Due to rounding, in the 1 decimal Figure 3 this difference appears as if it is only 0.3 days, but it closer to 0.4 days. Regardless, this level of difference, which appears rather clearly in a survival curve, is well below the level of difference we have chosen to highlight in our study.

(1) Figure 4: I find it very tedious to compare CFUs among different mutants from the plots. As the idea is to compare bacterial loads among the mutants at different timepoints, it would be easier to compare them if the data were shown within a timepoint (CFUs of each mutant at 2h, at 6h, and so on). This is also how the results are written in the text (within a time point). Would it also be clearer if the CFU plots were named, for example: " A', B', and C'"? 

We appreciate this note. We feel both representations have merits and pitfalls, but prefer our original design showing the progression of bacterial growth within genotype first. However, we have added dotted lines representing the wild-type bacterial loads at 2hpi, 12hpi, and 24hpi to assist the reader in making acrossgenotype comparisons at key time points. Like this, the reader can see if the error bars (StDev) overlap the mean of the wild-type, and so make more intuitive judgements about whether these differences are meaningful.

(2) Figure 2D is not referred to in the text. 

We now cite Fig2D in Line 356.

eLife Assessment

This potentially valuable study presents claims of evidence for coordinated membrane potential oscillations in E. coli biofilms that can be linked to a putative K+ channel and that may serve to enhance photo-protection. The finding of waves of membrane potential would be of interest to a wide audience from molecular biology to microbiology and physical biology. Unfortunately, a major issue is that it is unclear whether the dye used can act as a Nernstian membrane potential dye in E. coli. The arguments of the authors, who largely ignore previously published contradictory evidence, are not adequate in that they do not engage with the fact that the dye behaves in their hands differently than in the hands of others. In addition, the lack of proper validation of the experimental method including key control experiments leaves the evidence incomplete.

Reviewer #1 (Public Review):

(1) Significance of the findings:

Cell-to-cell communication is essential for higher functions in bacterial biofilms. Electrical signals have proven effective in transmitting signals across biofilms. These signals are then used to coordinate cellular metabolisms or to increase antibiotic tolerance. Here, the authors have reported for the first time coordinated oscillation of membrane potential in E. coli biofilms that may have a functional role in photoprotection.

(2) Strengths of the manuscript:

- The authors report original data.<br /> - For the first time, they showed that coordinated oscillations in membrane potential occur in E. Coli biofilms.<br /> - The authors revealed a complex two-phase dynamic involving distinct molecular response mechanisms.<br /> - The authors developed two rigorous models inspired by 1) Hodgkin-Huxley model for the temporal dynamics of membrane potential and 2) Fire-Diffuse-Fire model for the propagation of the electric signal.<br /> - Since its discovery by comparative genomics, the Kch ion channel has not been associated with any specific phenotype in E. coli. Here, the authors proposed a functional role for the putative gated-voltage-gated K+ ion channel (Kch channel) : enhancing survival under photo-toxic conditions.

(3) Weakness:

- Contrarily to what is stated in the abstract, the group of B. Maier has already reported collective electrical oscillations in the Gram-negative bacterium Neisseria gonorrhoeae (Hennes et al., PLoS Biol, 2023).<br /> - The data presented in the manuscript are not sufficient to conclude on the photo-protective role of the Kch channel. The authors should perform the appropriate control experiments related to Fig4D,E, i.e. reproduce these experiments without ThT to rule out possible photo-conversion effects on ThT that would modify its toxicity. In addition, it looks like the data reported on Fig 4E are extracted from Fig 4D. If this is indeed the case, it would be more conclusive to report the percentage of PI-positive cells in the population for each condition. This percentage should be calculated independently for each replicate. The authors should then report the average value and standard deviation of the percentage of dead cells for each condition.<br /> - Although Fig 4A clearly shows that light stimulation has an influence on the dynamics of ThT signal in the biofilm, it is important to rule out possible contributions of other environmental variations that occur when the flow is stopped at the onset of light stimulation. I understand that for technical reasons, the flow of fresh medium must be stopped for the sake of imaging. Therefore, I suggest to perform control experiments consisting in stopping the flow at different time intervals before image acquisition (30min or 1h before). If there is no significant contribution from environmental variations due to medium perfusion arrest, the dynamics of ThT signal must be unchanged regardless of the delay between flow stop and the start of light stimulation.<br /> - To precise the role of K+ in the habituation response, I suggest using the ionophore valinomycin at sub-inhibitory concentrations (5 or 10µM). It should abolish the habituation response. In addition, the Kch complementation experiment exhibits a sharp drop after the first peak but on a single point. It would be more convincing to increase the temporal resolution (1min->10s) to show that there are indeed a first and a second peak. Finally, the high concentration (100µM) of CCCP used in this study completely inhibits cell activity. Therefore, it is not surprising that no ThT dynamics was observed upon light stimulation at such concentration of CCCP.<br /> - Since TMRM signal exhibits a linear increase after the first response peak (Supp Fig1D), I recommend to mitigate the statement at line 78.<br /> - Electrical signal propagation is an important aspect of the manuscript. However, a detailed quantitative analysis of the spatial dynamics within the biofilm is lacking. At minima, I recommend to plot the spatio-temporal diagram of ThT intensity profile averaged along the azimuthal direction in the biofilm. In addition, it is unclear if the electrical signal propagates within the biofilm during the second peak regime, which is mediated by the Kch channel: I have plotted the spatio-temporal diagram for Video S3 and no electrical propagation is evident at the second peak. In addition, the authors should provide technical details of how R^2(t) is measured in the first regime (Fig 7E).<br /> - In the series of images presented in supplementary Figure 4A, no wavefront is apparent. Although the microscopy technics used in this figure differs from other images (like in Fig2), the wavefront should be still present. In addition, there is no second peak in confocal images as well (Supp Fig4B) .<br /> - Many important technical details are missing (e.g. biofilm size, R^2, curvature and 445nm irradiance measurements). The description of how these quantitates are measured should be detailed in the Material & Methods section.<br /> - Fig 5C: The curve in Fig 5D seems to correspond to the biofilm case. Since the model is made for single cells, the curve obtained by the model should be compared with the average curve presented in Fig 1B (i.e. single cell experiments).<br /> - For clarity, I suggest to indicate on the panels if the experiments concern single cell or biofilm experiments. Finally, please provide bright-field images associated to ThT images to locate bacteria.<br /> - In Fig 7B, the plateau is higher in the simulations than in the biofilm experiments. The authors should add a comment in the paper to explain this discrepancy.

Reviewer #2 (Public Review):

The authors use ThT dye as a Nernstian potential dye in E. coli. Quantitative measurements of membrane potential using any cationic indicator dye are based on the equilibration of the dye across the membrane according to Boltzmann's law.

Ideally, the dye should have high membrane permeability to ensure rapid equilibration. Others have demonstrated that E.coli cells in the presence of ThT do not load unless there is blue light present, that the loading profile does not look like it is expected for a cationic Nernstian dye. They also show that the loading profile of the dye is different for E.coli cells deleted for the TolC pump. I, therefore, objected to interpreting the signal from the ThT as a Vm signal when used in E.coli. Nothing the authors have said has suggested that I should be changing this assessment.

Specifically, the authors responded to my concerns as follows:

(1) 'We are aware of this study, but believe it to be scientifically flawed. We do not cite the article because we do not think it is a particularly useful contribution to the literature.' This seems to go against ethical practices when it comes to scientific literature citations. If the authors identified work that handles the same topic they do, which they believe is scientifically flawed, the discussion to reflect that should be included.

(2)'The Pilizota group invokes some elaborate artefacts to explain the lack of agreement with a simple Nernstian battery model. The model is incorrect not the fluorophore.'<br /> It seems the authors object to the basic principle behind the usage of Nernstian dyes. If the authors wish to use ThT according to some other model, and not as a Nernstian indicator, they need to explain and develop that model. Instead, they state 'ThT is a Nernstian voltage indicator' in their manuscript and expect the dye to behave like a passive voltage indicator throughout it.

(3)'We think the proton effect is a million times weaker than that due to potassium i.e. 0.2 M K+<br /> versus 10-7 M H+. We can comfortably neglect the influx of H+ in our experiments.'<br /> I agree with this statement by the authors. At near-neutral extracellular pH, E.coli keeps near-neutral intracellular pH, and the contribution from the chemical concentration gradient to the electrochemical potential of protons is negligible. The main contribution is from the membrane potential. However, this has nothing to do with the criticism to which this is the response of the authors. The criticism is that ThT has been observed not to permeate the cell without blue light. The blue light has been observed to influence the electrochemical potential of protons (and given that at near-neutral intracellular and extracellular pH this is mostly the membrane potential, as authors note themselves, we are talking about Vm effectively). Thus, two things are happening when one is loading the ThT, not just expected equilibration but also lowering of membrane potential. The electrochemical potential of protons is coupled via the membrane potential to all the other electrochemical potentials of ions, including the mentioned K+.

(4) 'The vast majority of cells continue to be viable. We do not think membrane damage is dominating.' In response to the question on how the authors demonstrated TMRM loading and in which conditions (and while reminding them that TMRM loading profile in E.coli has been demonstrated in Potassium Phosphate buffer). The request was to demonstrate TMRM loading profile in their condition as well as to show that it does not depend on light. Cells could still be viable, as membrane permeabilisation with light is gradual, but the loading of ThT dye is no longer based on simple electrochemical potential (of the dye) equilibration.

(5) On the comment on the action of CCCP with references included, authors include a comment that consists of phrases like 'our understanding of the literature' with no citations of such literature. Difficult to comment further without references.

(6) 'Shielding would provide the reverse effect, since hyperpolarization begins in the dense centres of the biofilms. For the initial 2 hours the cells receive negligible blue light. Neither of the referee's comments thus seem tenable.'<br /> The authors have misunderstood my comment. I am not advocating shielding (I agree that this is not it) but stating that this is not the only other explanation for what they see (apart from electrical signaling). The other I proposed is that the membrane has changed in composition and/or the effective light power the cells can tolerate. The authors comment only on the light power (not convincingly though, giving the number for that power would be more appropriate), not on the possible changes in the membrane permeability.

(7) 'The work that TolC provides a possible passive pathway for ThT to leave cells seems slightly niche. It just demonstrates another mechanism for the cells to equilibrate the concentrations of ThT in a Nernstian manner i.e. driven by the membrane voltage.' I am not sure what the authors mean by another mechanism. The mechanism of action of a Nernstian dye is passive equilibration according to the electrochemical potential (i.e. until the electrochemical potential of the dye is 0).

(8) 'In the 70 years since Hodgkin and Huxley first presented their model, a huge number of similar models have been proposed to describe cellular electrophysiology. We are not being hyperbolic when we state that the HH models for excitable cells are like the Schrödinger<br /> equation for molecules. We carefully adapted our HH model to reflect the currently understood electrophysiology of E. coli.'

I gave a very concrete comment on the fact that in the HH model conductivity and leakage are as they are because this was explicitly measured. The authors state that they have carefully adopted their model based on what is currently understood for E.coli electrophysiology. It is not clear how. HH uses gKn^4 based on Figure2 here https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1392413/pdf/jphysiol01442-0106.pdf, i.e. measured rise and fall of potassium conductance on msec time scales. I looked at the citation the authors have given and found a resistance of an entire biofilm of a given strain at 3 applied voltages. So why n^4 based on that? Why does unknown current have gqz^4 form? Sodium conductance in HH is described by m^3hgNa (again based on detailed conductance measurements), so why unknown current in E.coli by gQz^4? Why leakage is in the form that it is, based on what measurement?

Throughout their responses, the authors seem to think that collapsing the electrochemical gradient of protons is all about protons, and this is not the case. At near neutral inside and outside pH, the electrochemical potential of protons is simply membrane voltage. And membrane voltage acts on all ions in the cell.

Authors have started their response to concrete comments on the usage of ThT dye with comments on papers from my group that are not all directly relevant to this publication. I understand that their intention is to discredit a reviewer but given that my role here is to review this manuscript, I will only address their comments to the publications/part of publications that are relevant to this manuscript and mention what is not relevant.

Publications in the order these were commented on.

(1) In a comment on the paper that describes the usage of ThT dye as a Nernstian dye authors seem to talk about a model of an entire active cell.<br /> 'Huge oscillations occur in the membrane potentials of E. coli that cannot be described by the SNB model.' The two have nothing to do with each other. Nernstian dye equilibrates according to its electrochemical potential. Once that happens it can measure the potential (under the assumption that not too much dye has entered and thus lowered too much the membrane potential under measurement). The time scale of that is important, and the dye can only measure processes that are slower than that equilibration. If one wants to use a dye that acts under a different model, first that needs to be developed, and then coupled to any other active cell model.

(2) The part of this paper that is relevant is simply the usage of TMRM dye. It is used as Nernstian dye, so all the above said applies. The rest is a study of flagellar motor.

(3) The authors seem to not understand that the electrochemical potential of protons is coupled to the electrochemical potentials of all other ions, via the membrane potential. In the manuscript authors talk about, PMF~Vm, as DeltapH~0. Other than that this publication is not relevant to their current manuscript.

(4) The manuscript in fact states precisely that PMF cannot be generated by protons only and some other ions need to be moved out for the purpose. In near neutral environment it stated that these need to be cations (K+ e.g.). The model used in this manuscript is a pump-leak model. Neither is relevant for the usage of ThT dye.

Further comments include, along the lines of:

'The editors stress the main issue raised was a single referee questioning the use of ThT as an indicator of membrane potential. We are well aware of the articles by the Pilizota group and we believe them to be scientifically flawed. The authors assume there are no voltage-gated ion channels in E. coli and then attempt to explain motility data based on a simple Nernstian battery model (they assume E. coli are unexcitable<br /> matter). This in turn leads them to conclude the membrane dye ThT is faulty, when in fact it is a problem with their simple battery model.'

The only assumption made when using a cationic Nernstian dye is that it equilibrates passively across the membrane according to its electrochemical potential. As it does that, it does lower the membrane potential, which is why as little as possible is added so that this is negligible. The equilibration should be as fast as possible, but at the very least it should be known, as no change in membrane potential can be measured that is faster than that.

This behaviour should be orthogonal to what the cell is doing, it is a probe after all. If the cell is excitable, a Nernstian dye can be used, as long as it's still passively equilibrating and doing so faster than any changes in membrane potential due to excitations of the cells. There are absolutely no assumptions made on the active system that is about to be measured by this expected behaviour of a Nernstian dye. And there shouldn't be, it is a probe. If one wants to use a dye that is not purely Nernstian that behaviour needs to be described and a model proposed. As far as I can find, authors do no such thing.

There is a comment on the use of a flagellar motor as a readout of PMF, stating that the motor can be stopped by YcgR citing the work from 2023. Indeed, there is a range of references such as https://doi.org/10.1016/j.molcel.2010.03.001 that demonstrate this (from around 2000-2010 as far as I am aware). The timescale of such slowdown is hours (see here Figure 5 https://www.cell.com/cell/pdf/S0092-8674(10)00019-X.pdf). Needless to say, the flagellar motor when used as a probe, needs to stay that in the conditions used. Thus one should always be on the lookout at any other such proteins that could slow it down and we are not aware of yet or make the speed no longer proportional to the PMF. In the papers my group uses the motor the changes are fast, often reversible, and in the observation window of 30min. They are also the same with DeltaYcgR strain, which we have not included as it seemed given the time scales it's obvious, but certainly can in the future (as well as stay vigilant on any conditions that would render the motor a no longer suitable probe for PMF).

Reviewer #3 (Public Review):

This manuscript by Akabuogu et al. investigates membrane potential dynamics in E. coli. Membrane potential fluctuations have been observed in bacteria by several research groups in recent years, including in the context of bacterial biofilms where they have been proposed to play a role in cellular communication. Here, these authors investigate membrane potential in E. coli, in both single cells and biofilms. I have reviewed the revised manuscript provided by the authors, as well as their responses to the initial reviews; my opinion about the manuscript is largely unchanged. I have focused my public review on those issues that I believe to be most pressing, with additional comments included in the review to authors. Although these authors are working in an exciting research area, the evidence they provide for their claims is inadequate, and several key control experiments are still missing. In some cases, the authors allude to potentially relevant data in their responses to the initial reviews, but unfortunately these data are not shown. Furthermore, I cannot identify any traveling wavefronts in the data included in this manuscript. In addition to the challenges associated with the use of Thioflavin-T (ThT) raised by the second reviewer, these caveats make the work presented in this manuscript difficult to interpret.

First, some of the key experiments presented in the paper lack required controls:

(1) This paper asserts that the observed ThT fluorescence dynamics are induced by blue light. This is a fundamental claim in the paper, since the authors go on to argue that these dynamics are part of a blue light response. This claim must be supported by the appropriate negative control experiment measuring ThT fluorescence dynamics in the absence of blue light- if this idea is correct, these dynamics should not be observed in the absence of blue light exposure. If this experiment cannot be performed with ThT since blue light is used for its excitation, TMRM can be used instead.

In response to this, the authors wrote that "the fluorescent baseline is too weak to measure cleanly in this experiment." If they observe no ThT signal above noise in their time lapse data in the absence of blue light, this should be reported in the manuscript- this would be a satisfactory negative control. They then wrote that "It appears the collective response of all the bacteria hyperpolarization at the same time appears to dominate the signal." I am not sure what they mean by this- perhaps that ThT fluorescence changes strongly only in response to blue light? This is a fundamental control for this experiment that ought to be presented to the reader.

(2) The authors claim that a ∆kch mutant is more susceptible to blue light stress, as evidenced by PI staining. The premise that the cells are mounting a protective response to blue light via these channels rests on this claim. However, they do not perform the negative control experiment, conducting PI staining for WT the ∆kch mutant in the absence of blue light. In the absence of this control it is not possible to rule out effects of the ∆kch mutation on overall viability and/or PI uptake. The authors do include a growth curve for comparison, but planktonic growth is a very different context than surface-attached biofilm growth. Additionally, the ∆kch mutation may have impacts on PI permeability specifically that are not addressed by a growth curve. The negative control experiment is of key importance here.

Second, the ideas presented in this manuscript rely entirely on analysis of ThT fluorescence data, specifically a time course of cellular fluorescence following blue light treatment. However, alternate explanations for and potential confounders of the observed dynamics are not sufficiently addressed:

(1) Bacterial cells are autofluorescent, and this fluorescence can change significantly in response to stress (e.g. blue light exposure). To characterize and/or rule out autofluorescence contributions to the measurement, the authors should present time lapse fluorescence traces of unstained cells for comparison, acquired under the same imaging conditions in both wild type and ∆kch mutant cells. In their response to reviewers the authors suggested that they have conducted this experiment and found that the autofluorescence contribution is negligible, which is good, but these data should be included in the manuscript along with a description of how these controls were conducted.

(2) Similarly, in my initial review I raised a concern about the possible contributions of photobleaching to the observed fluorescence dynamics. This is particularly relevant for the interpretation of the experiment in which catalase appears to attenuate the decay of the ThT signal; this attenuation could alternatively be due to catalase decreasing ThT photobleaching. In their response, the authors indicated that photobleaching is negligible, which would be good, but they do not share any evidence to support this claim. Photobleaching can be assessed in this experiment by varying the light dosage (illumination power, frequency, and/or duration) and confirming that the observed fluorescence dynamics are unaffected.

Third, the paper claims in two instances that there are propagating waves of ThT fluorescence that move through biofilms, but I do not observe these waves in any case:

(1) The first wavefront claim relates to small cell clusters, in Fig. 2A and Video S2 and S3 (with Fig. 2A and Video S2 showing the same biofilm.) I simply do not see any evidence of propagation in either case- rather, all cells get brighter and dimmer in tandem. I downloaded and analyzed Video S3 in several ways (plotting intensity profiles for different regions at different distances from the cluster center, drawing a kymograph across the cluster, etc.) and in no case did I see any evidence of a propagating wavefront. (I attempted this same analysis on the biofilm shown in Fig. 2A and Video S2 with similar results, but the images shown in the figure panels and especially the video are still both so saturated that the quantification is difficult to interpret.) If there is evidence for wavefronts, it should be demonstrated explicitly by analysis of several clusters. For example, a figure of time-to-peak vs. position in the cluster demonstrating a propagating wave would satisfy this. Currently, I do not see any wavefronts in this data.

(2) The other wavefront claim relates to biofilms, and the relevant data is presented in Fig. S4 (and I believe also in what is now Video S8, but no supplemental video legends are provided, and this video is not cited in text.) As before, I cannot discern any wavefronts in the image and video provided; Reviewer 1 was also not able to detect wave propagation in this video by kymograph. Some mean squared displacements are shown in Fig. 7. As before, the methods for how these were obtained are not clearly documented either in this manuscript or in the BioRXiv preprint linked in the initial response to reviewers, and since wavefronts are not evident in the video it is hard to understand what is being measured here- radial distance from where? (The methods section mentions radial distance from the substrate, this should mean Z position above the imaging surface, and no wavefronts are evident in Z in the figure panels or movie.) Thus, clear demonstration of these wavefronts is still missing here as well.

Fourth, I have some specific questions about the study of blue light stress and the use of PI as a cell viability indicator:

(1) The logic of this paper includes the premise that blue light exposure is a stressor under the experimental conditions employed in the paper. Although it is of course generally true that blue light can be damaging to bacteria, this is dependent on light power and dosage. The control I recommended above, staining cells with PI in the presence and absence of blue light, will also allow the authors to confirm that this blue light treatment is indeed a stressor- the PI staining would be expected to increase in the presence of blue light if this is so.

(2) The presence of ThT may complicate the study of the blue light stress response, since ThT enhances the photodynamic effects of blue light in E. coli (Bondia et al. 2021 Chemical Communications). The authors could investigate ThT toxicity under these conditions by staining cells with PI after exposing them to blue light with or without ThT staining.

(3) In my initial review, I wrote the following: "In Figures 4D - E, the interpretation of this experiment can be confounded by the fact that PI uptake can sometimes be seen in bacterial cells with high membrane potential (Kirchhoff & Cypionka 2017 J Microbial Methods); the interpretation is that high membrane potential can lead to increased PI permeability. Because the membrane potential is largely higher throughout blue light treatment in the ∆kch mutant (Fig. 3[BC]), this complicates the interpretation of this experiment." In their response, the authors suggested that these results are not relevant in this case because "In our experiment methodology, cell death was not forced on the cells by introducing an extra burden or via anoxia." However, the logic of the paper is that the cells are in fact dying due to an imposed external stressor, which presumably also confers an increased burden as the cells try to deal with the stress. Instead, the authors should simply use a parallel method to confirm the results of PI staining. For example, the experiment could be repeated with other stains, or the viability of blue light-treated cells could be addressed more directly by outgrowth or colony-forming unit assays.

The CFU assay suggested above has the additional advantage that it can also be performed on planktonic cells in liquid culture that are exposed to blue light. If, as the paper suggests, a protective response to blue light is being coordinated at the biofilm level by these membrane potential fluctuations, the WT strain might be expected to lose its survival advantage vs. the ∆kch mutant in the absence of a biofilm.

Fifth, in several cases the data are presented in a way that are difficult to interpret, or the paper makes claims that are different to observe in the data:

(1) The authors suggest that the ThT and TMRM traces presented in Fig. S1D have similar shapes, but this is not obvious to me- the TMRM curve has very little decrease after the initial peak and only a modest, gradual rise thereafter. The authors suggest that this is due to increased TMRM photobleaching, but I would expect that photobleaching should exacerbate the signal decrease after the initial peak. Since this figure is used to support the use of ThT as a membrane potential indicator, and since this is the only alternative measurement of membrane potential presented in text, the authors should discuss this discrepancy in more detail.

(2) The comparison of single cells to microcolonies presented in figures 1B and D still needs revision:

First, both reviewer 1 and I commented in our initial reviews that the ThT traces, here and elsewhere, should not be normalized- this will help with the interpretation of some of the claims throughout the manuscript.

Second, the way these figures are shown with all traces overlaid at full opacity makes it very difficult to see what is being compared. Since the point of the comparison is the time to first peak (and the standard deviation thereof), histograms of the distributions of time to first peak in both cases should be plotted as a separate figure panel.<br /> Third, statistical significance tests ought to be used to evaluate the statistical strength of the comparisons between these curves. The authors compare both means and standard deviations of the time to first peak, and there are appropriate statistical tests for both types of comparisons.

(3) The authors claim that the curve shown in Fig. S4B is similar to the simulation result shown in Fig. 7B. I remain unconvinced that this is so, particularly with respect to the kinetics of the second peak- at least it seems to me that the differences should be acknowledged and discussed. In any case, the best thing to do would be to move Fig. S4B to the main text alongside Fig. 7B so that the readers can make the comparison more easily.

(4) As I wrote in my first review, in the discussion of voltage-gated calcium channels, the authors refer to "spiking events", but these are not obvious in Figure S3E. Although the fluorescence intensity changes over time, these fluctuations cannot be distinguished from measurement noise. A no-light control could help clarify this.

(5) In the lower irradiance conditions in Fig. 4A, the ThT dynamics are slower overall, and it looks like the ThT intensity is beginning to rise at the end of the measurement. The authors write that no second peak is observed below an irradiance threshold of 15.99 µW/mm2. However, could a more prominent second peak be observed in these cases if the measurement time was extended? Additionally, the end of these curves looks similar to the curve in Fig. S4B, in which the authors write that the slow rise is evidence of the presence of a second peak, in contrast to their interpretation here.

Additional considerations:

(1) The analysis and interpretation of the first peak, and particularly of the time-to-fire data is challenging throughout the manuscript the time resolution of the data set is quite limited. It seems that a large proportion of cells have already fired after a single acquisition frame. It would be ideal to increase the time resolution on this measurement to improve precision. This could be done by imaging more quickly, but that would perhaps necessitate more blue light exposure; an alternative is to do this experiment under lower blue light irradiance where the first spike time is increased (Figure 4A).

(2) The authors suggest in the manuscript that "E. coli biofilms use electrical signalling to coordinate long-range responses to light stress." In addition to the technical caveats discussed above, I am missing a discussion about what these responses might be. What constitutes a long-range response to light stress, and are there known examples of such responses in bacteria?

(3) The presence of long-range blue light responses can also be interrogated experimentally, for example, by repeating the Live/Dead experiment in planktonic culture or the single-cell condition. If the protection from blue light specifically emerges due to coordinated activity of the biofilm, the ∆kch mutant would not be expected to show a change in Live/Dead staining in non-biofilm conditions. The CFU experiment I mentioned above could also implicate coordinated long-range responses specifically, if biofilms and liquid culture experiments can be compared (although I know that recovering cells from biofilms is challenging.)

4. At the end of the results section, the authors suggest a critical biofilm size of only 4 μm for wavefront propagation (not much larger than a single cell!) The authors show responses for various biofilm sizes in Fig. 2C, but these are all substantially larger (and this figure also does not contain wavefront information.) Are there data for cell clusters above and below this size that could support this claim more directly?

(5) In Fig. 4C, the overall trajectories of extracellular potassium are indeed similar, but the kinetics of the second peak of potassium are different than those observed by ThT (it rises minutes earlier)- is this consistent with the idea that Kch is responsible for that peak? Additionally, the potassium dynamics also include the first ThT peak- is this surprising given that the Kch channel has no effect on this peak according to the model?

Detailed comments:

Why are Fig. 2A and Video S2 called a microcluster, whereas Video S3, which is smaller, is called a biofilm?

"We observed a spontaneous rapid rise in spikes within cells in the center of the biofilm" (Line 140): What does "spontaneous" mean here?

"This demonstrates that the ion-channel mediated membrane potential dynamics is a light stress relief process.", "E. coli cells employ ion-channel mediated dynamics to manage ROS-induced stress linked to light irradiation." (Line 268 and the second sentence of the Fig. 4F legend): This claim is not well-supported. There are several possible interpretations of the catalase experiment (which should be discussed); this experiment perhaps suggests that ROS impacts membrane potential but does not indicate that these membrane potential fluctuations help the cells respond to blue light stress. The loss of viability in the ∆kch mutant might indicate a link between these membrane potential experiments and viability, but it is hard to interpret without the no light controls I mention above.

"The model also predicts... the external light stress" (Lines 338-341): Please clarify this section. Where does this prediction arise from in the modeling work? Second, I am not sure what is meant by "modulates the light stress" or "keeps the cell dynamics robust to the intensity of external light stress" (especially since the dynamics clearly vary with irradiance, as seen in Figure 4A).

"We hypothesized that E. coli not only modulates the light-induced stress but also handles the increase of the ROS by adjusting the profile of the membrane potential dynamics" (Line 347): I am not sure what "handles the ROS by adjusting the profile of the membrane potential dynamics" means. What is meant by "handling" ROS? Is the hypothesis that membrane potential dynamics themselves are protective against ROS, or that they induce a ROS-protective response downstream, or something else? Later the authors write that changes in the response to ROS in the model agree with the hypothesis, but just showing that ROS impacts the membrane potential does not seem to demonstrate that this has a protective effect against ROS.

"Mechanosensitive ion channels (MS) are vital for the first hyperpolarization event in E. coli." (Line 391): This is misleading- mechanosensitive ion channels totally ablate membrane potential dynamics, they don't have a specific effect on the first hyperpolarization event. The claim that mechanonsensitive ion channels are specifically involved in the first event also appears in the abstract.

Also, the apparent membrane potential is much lower even at the start of the experiment in these mutants (Fig. 6C-D)- is this expected? This seems to imply that these ion channels also have a blue light-independent effect.

Throughout the paper, there are claims that the initial ThT spike is involved in "registering the presence of the light stress" and similar. What is the evidence for this claim?

"We have presented much better quantitative agreement of our model with the propagating wavefronts in E. coli biofilms..." (Line 619): It is not evident to me that the agreement between model and prediction is "much better" in this work than in the cited work (reference 57, Hennes et al. 2023). The model in Figure 4 of ref. 57 seems to capture the key features of their data.

In methods, "Only cells that are hyperpolarized were counted in the experiment as live" (Line 745): what percentage of cells did not hyperpolarize in these experiments?

Some indication of standard deviation (error bars or shading) should be added to all figures where mean traces are plotted.

Video S8 is very confusing- why does the video play first forwards and then backwards? It is easy to misinterpret this as a rise in the intensity at the end of the experiment.

Author response:

The issue of a control without blue light illumination was raised. Clearly without the light we will not obtain any signal in the fluorescence microscopy experiments, which would not be very informative. Instead, we changed the level of blue light illumination in the fluorescence microscopy experiments (figure 4A) and the response of the bacteria scales with dosage. It is very hard to find an alternative explanation, beyond that the blue light is stressing the bacteria and modulating their membrane potentials.

One of the referees refuses to see wavefronts in our microscopy data. We struggle to understand whether it is an issue with definitions (Waigh has published a tutorial on the subject in Chapter 5 of his book ‘The physics of bacteria: from cells to biofilms’, T.A.Waigh, CUP, 2024 – figure 5.1 shows a sketch) or something subtler on diffusion in excitable systems. We stand by our claim that we observe wavefronts, similar to those observed by Prindle et al¹ and Blee et al² for B. subtilis biofilms.

The referee is questioning our use of ThT to probe the membrane potential. We believe the Pilizota and Strahl groups are treating the E. coli as unexcitable cells, leading to their problems. Instead, we believe E. coli cells are excitable (containing the voltage-gated ion channel Kch) and we now clearly state this in the manuscript. Furthermore, we include a section here discussing some of the issues with ThT.

Use of ThT as a voltage sensor in cells

ThT is now used reasonably widely in the microbiology community as a voltage sensor in both bacterial [Prindle et al]1 and fungal cells [Pena et al]12. ThT is a small cationic fluorophore that loads into the cells in proportion to their membrane potential, thus allowing the membrane potential to be measured from fluorescence microscopy measurements.

Previously ThT was widely used to quantify the growth of amyloids in molecular biology experiments (standardized protocols exist and dedicated software has been created)13 and there is a long history of its use14. ThT fluorescence is bright, stable and slow to photobleach.

Author response image 1 shows a schematic diagram of the ThT loading in E. coli in our experiments in response to illumination with blue light. Similar results were previously presented by Mancini et al15, but regimes 2 and 3 were mistakenly labelled as artefacts.

Author response image 1.

Schematic diagram of ThT loading during an experiment with E. coli cells under blue light illumination i.e. ThT fluorescence as a function of time. Three empirical regimes for the fluorescence are shown (1, 2 and 3).

The classic study of Prindle et al on bacterial biofilm electrophysiology established the use of ThT in B. subtilis biofilms by showing similar results occurred with DiSc3 which is widely used as a Nernstian voltage sensor in cellular biology1 e.g. with mitochondrial membrane potentials in eukaryotic organisms where there is a large literature. We repeated such a comparative calibration of ThT with DiSc3 in a previous publication with both B. subtilis and P. aeruginosa cells2. ThT thus functioned well in our previous publications with Gram positive and Gram negative cells.

However, to our knowledge, there are now two groups questioning the use of ThT and DiSc3 as voltage sensors with E. coli cells15-16. The first by the Pilizota group claims ThT only works as a voltage sensor in regime 1 of Author response image 1 using a method based on the rate of rotation of flagellar motors. Another slightly contradictory study by the Strahl group claims DiSc316 only acts as a voltage sensor with the addition of an ionophore for potassium which allows free movement of potassium through the E. coli membranes.

Our resolution to this contradiction is that ThT does indeed work reasonably well with E. coli. The Pilizota group’s model for rotating flagellar motors assumes the membrane voltage is not varying due to excitability of the membrane voltage (otherwise a non-linear Hodgkin Huxley type model would be needed to quantify their results) i.e. E. coli cells are unexcitable. We show clearly in our study that ThT loading in E. coli is a function of irradiation with blue light and is a stress response of the excitable cells. This is in contradiction to the Pilizota group’s model. The Pilizota group’s model also requires the awkward fiction of why cells decide to unload and then reload ThT in regimes 2 and 3 of Author response image 1 due to variable membrane partitioning of the ThT. Our simple explanation is that it is just due to the membrane voltage changing and no membrane permeability switch needs to be invoked. The Strahl group’s16 results with DiSc3 are also explained by a neglect of the excitable nature of E. coli cells that are reacting to blue light irradiation. Adding ionophores to the E. coli membranes makes the cells unexcitable, reduces their response to blue light and thus leads to simple loading of DiSc3 (the physiological control of K+ in the cells by voltage-gated ion channels has been short circuited by the addition of the ionophore).

Further evidence of our model that ThT functions as a voltage sensor with E. coli include:

1) The 3 regimes in Author response image 1 from ThT correlate well with measurements of extracellular potassium ion concentration using TMRM i.e. all 3 regimes in Author response image 1 are visible with this separate dye (figure 1d).

2) We are able to switch regime 3 in Author response image 1, off and then on again by using knock downs of the potassium ion channel Kch in the membranes of the E. coli and then reinserting the gene back into the knock downs. This cannot be explained by the Pilizota model.

We conclude that ThT works reasonably well as a sensor of membrane voltage in E. coli and the previous contradictory studies15-16 are because they neglect the excitable nature of the membrane voltage of E. coli cells in response to the light used to make the ThT fluoresce.

Three further criticisms of the Mancini et al method15 for calibrating membrane voltages include:

1) E. coli cells have clutches that are not included in their models. Otherwise the rotation of the flagella would be entirely enslaved to the membrane voltage allowing the bacteria no freedom to modulate their speed of motility.

2) Ripping off the flagella may perturb the integrity of the cell membrane and lead to different loading of the ThT in the E. coli cells.

3) Most seriously, the method ignores the activity of many other ion channels (beyond H+) on the membrane voltage that are known to exist with E. coli cells e.g. Kch for K+ ions. The Pilizota groups uses a simple Nernstian battery model developed for mitochondria in the 1960s. It is not adequate to explain our results.

An additional criticism of the Winkel et al study17 from the Strahl group is that it indiscriminately switches between discussion of mitochondria and bacteria e.g. on page 8 ‘As a consequence the membrane potential is dominated by H+’. Mitochondria are slightly alkaline intracellular organelles with external ion concentrations in the cytoplasm that are carefully controlled by the eukaryotic cells. E. coli are not i.e. they have neutral internal pHs, with widely varying extracellular ionic concentrations and have reinforced outer membranes to resist osmotic shocks (in contrast mitochondria can easily swell in response to moderate changes in osmotic pressure).

A quick calculation of the equilibrium membrane voltage of E. coli can be easily done using the Nernst equation dependent on the extracellular ion concentrations defined by the growth media (the intracellular ion concentrations in E. coli are 0.2 M K+ and 10-7 M H+ i.e. there is a factor of a million fewer H+ ions). Thus in contradiction to the claims of the groups of Pilizota15 and Strahl17, H+ is a minority determinant to the membrane voltage of E. coli. The main determinant is K+. For a textbook version of this point the authors can refer to Chapter 4 of D. White, et al’s ‘The physiology and biochemistry of prokaryotes’, OUP, 2012, 4th edition.

Even in mitochondria the assumption that H+ dominates the membrane potential and the cells are unexcitable can be questioned e.g. people have observed pulsatile depolarization phenomena with mitochondria18-19. A large number of K+ channels are now known to occur in mitochondrial membranes (not to mention Ca2+ channels; mitochondria have extensive stores of Ca2+) and they are implicated in mitochondrial membrane potentials. In this respect the seminal Nobel prize winning research of Peter Mitchell (1961) on mitochondria needs to be amended20. Furthermore, the mitochondrial work is clearly inapplicable to bacteria (the proton motive force, PMF, will instead subtly depend on non-linear Hodgkin-Huxley equations for the excitable membrane potential, similar to those presented in the current article). A much more sophisticated framework has been developed to describe electrophysiology by the mathematical biology community to describe the activity of electrically excitable cells (e.g. with neurons, sensory cells and cardiac cells), beyond Mitchell’s use of the simple stationary equilibrium thermodynamics to define the Proton Motive Force via the electrochemical potential of a proton (the use of the word ‘force’ is unfortunate, since it is a potential). The tools developed in the field of mathematical electrophysiology8 should be more extensively applied to bacteria, fungi, mitochondria and chloroplasts if real progress is to be made.

Related to the previous point, we now cite articles from the Pilizota and Strahl groups in the main text (one from each group). Unfortunately, the space constraints of eLife mean we cannot make a more detailed discussion in the main article.

In terms of modelling the ion channels, the Hodgkin-Huxley type model proposes that the Kch ion channel can be modelled as a typical voltage-gated potassium ion channel i.e. with a 𝑛⁴ term in its conductivity. The literature agrees that Kch is a voltage-gated potassium ion channel based on its primary sequence³. The protein has the typical 6 transmembrane helix motif for a voltage-gated ion channel. The agent-based model assumes little about the structure of ion channels in E. coli, other than they release potassium in response to a threshold potassium concentration in their environment. The agent based model is thus robust to the exact molecular details chosen and predicts the anomalous transport of the potassium wavefronts reasonably well (the modelling was extended in a recent Physical Review E article(⁴). Such a description of reaction-anomalous diffusion phenomena has not to our knowledge been previously achieved in the literature⁵ and in general could be used to describe other signaling molecules.

1. Prindle, A.; Liu, J.; Asally, M.; Ly, S.; Garcia-Ojalvo, J.; Sudel, G. M., Ion channels enable electrical communication in bacterial communities. Nature 2015, 527, 59.

2. Blee, J. A.; Roberts, I. S.; Waigh, T. A., Membrane potentials, oxidative stress and the dispersal response of bacterial biofilms to 405 nm light. Physical Biology 2020, 17, 036001.

3. Milkman, R., An E. col_i homologue of eukaryotic potassium channel proteins. _PNAS 1994, 91, 3510-3514.

4. Martorelli, V.; Akabuogu, E. U.; Krasovec, R.; Roberts, I. S.; Waigh, T. A., Electrical signaling in three-dimensional bacterial biofilms using an agent-based fire-diffuse-fire model. Physical Review E 2024, 109, 054402.

5. Waigh, T. A.; Korabel, N., Heterogeneous anomalous transport in cellular and molecular biology. Reports on Progress in Physics 2023, 86, 126601.

6. Hodgkin, A. L.; Huxley, A. F., A quantitative description of membrane current and its application to conduction and excitation in nerve. Journal of Physiology 1952, 117, 500.

7. Dawson, S. P.; Keizer, J.; Pearson, J. E., Fire-diffuse-fire model of dynamics of intracellular calcium waves. PNAS 1999, 96, 606.

8. Keener, J.; Sneyd, J., Mathematical Physiology. Springer: 2009.

9. Coombes, S., The effect of ion pumps on the speed of travelling waves in the fire-diffuse-fire model of Ca2+ release. Bulletin of Mathematical Biology 2001, 63, 1.

10. Blee, J. A.; Roberts, I. S.; Waigh, T. A., Spatial propagation of electrical signals in circular biofilms. Physical Review E 2019, 100, 052401.

11. Gorochowski, T. E.; Matyjaszkiewicz, A.; Todd, T.; Oak, N.; Kowalska, K., BSim: an agent-based tool for modelling bacterial populations in systems and synthetic biology. PloS One 2012, 7, 1.

12. Pena, A.; Sanchez, N. S.; Padilla-Garfias, F.; Ramiro-Cortes, Y.; Araiza-Villaneuva, M.; Calahorra, M., The use of thioflavin T for the estimation and measurement of the plasma membrane electric potential difference in different yeast strains. Journal of Fungi 2023, 9 (9), 948.

13. Xue, C.; Lin, T. Y.; Chang, D.; Guo, Z., Thioflavin T as an amyloid dye: fibril quantification, optimal concentration and effect on aggregation. Royal Society Open Science 2017, 4, 160696.

14. Meisl, G.; Kirkegaard, J. B.; Arosio, P.; Michaels, T. C. T.; Vendruscolo, M.; Dobson, C. M.; Linse, S.; Knowles, T. P. J., Molecular mechanisms of protein aggregation from global fitting of kinetic models. Nature Protocols 2016, 11 (2), 252-272.

15. Mancini, L.; Tian, T.; Guillaume, T.; Pu, Y.; Li, Y.; Lo, C. J.; Bai, F.; Pilizota, T., A general workflow for characterization of Nernstian dyes and their effects on bacterial physiology. Biophysical Journal 2020, 118 (1), 4-14.

16. Buttress, J. A.; Halte, M.; Winkel, J. D. t.; Erhardt, M.; Popp, P. F.; Strahl, H., A guide for membrane potential measurements in Gram-negative bacteria using voltage-sensitive dyes. Microbiology 2022, 168, 001227.

17. Derk te Winkel, J.; Gray, D. A.; Seistrup, K. H.; Hamoen, L. W.; Strahl, H., Analysis of antimicrobial-triggered membrane depolarization using voltage sensitive dyes. Frontiers in Cell and Developmental Biology 2016, 4, 29.

18. Schawarzlander, M.; Logan, D. C.; Johnston, I. G.; Jones, N. S.; Meyer, A. J.; Fricker, M. D.; Sweetlove, L. J., Pulsing of membrane potential in individual mitochondria. The Plant Cell 2012, 24, 1188-1201.

19. Huser, J.; Blatter, L. A., Fluctuations in mitochondrial membrane potential caused by repetitive gating of the permeability transition pore. Biochemistry Journal 1999, 343, 311-317.

20. Mitchell, P., Coupling of phosphorylation to electron and hydrogen transfer by a chemi-osmotic type of mechanism. Nature 1961, 191 (4784), 144-148.

21. Baba, T.; Ara, M.; Hasegawa, Y.; Takai, Y.; Okumura, Y.; Baba, M.; Datsenko, K. A.; Tomita, M.; Wanner, B. L.; Mori, H., Construction of Escherichia Coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Molecular Systems Biology 2006, 2, 1.

22. Schinedlin, J.; al, e., Fiji: an open-source platform for biological-image analysis. Nature Methods 2012, 9, 676.

23. Hartmann, R.; al, e., Quantitative image analysis of microbial communities with BiofilmQ. Nature Microbiology 2021, 6 (2), 151.

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

Critical synopsis of the articles cited by referee 2:

(1) ‘Generalized workflow for characterization of Nernstian dyes and their effects on bacterial physiology’, L.Mancini et al, Biophysical Journal, 2020, 118, 1, 4-14.

This is the central article used by referee 2 to argue that there are issues with the calibration of ThT for the measurement of membrane potentials. The authors use a simple Nernstian battery (SNB) model and unfortunately it is wrong when voltage-gated ion channels occur. Huge oscillations occur in the membrane potentials of E. coli that cannot be described by the SNB model. Instead a Hodgkin Huxley model is needed, as shown in our eLife manuscript and multiple other studies (see above). Arrhenius kinetics are assumed in the SNB model for pumping with no real evidence and the generalized workflow involves ripping the flagella off the bacteria! The authors construct an elaborate ‘work flow’ to insure their ThT results can be interpreted using their erroneous SNB model over a limited range of parameters.

(2) ‘Non-equivalence of membrane voltage and ion-gradient as driving forces for the bacterial flagellar motor at low load’, C.J.Lo, et al, Biophysical Journal, 2007, 93, 1, 294.

An odd de novo chimeric species is developed using an E. coli  chassis which uses Na+ instead of H+ for the motility of its flagellar motor. It is not clear the relevance to wild type E. coli, due to the massive physiological perturbations involved. A SNB model is using to fit the data over a very limited parameter range with all the concomitant errors.

(3) Single-cell bacterial electrophysiology reveals mechanisms of stress-induced damage’, E.Krasnopeeva, et al, Biophysical Journal, 2019, 116, 2390.

The abstract says ‘PMF defines the physiological state of the cell’. This statement is hyperbolic. An extremely wide range of molecules contribute to the physiological state of a cell. PMF does not even define the electrophysiology of the cell e.g. via the membrane potential. There are 0.2 M of K+ compared with 0.0000001 M of H+ in E. coli, so K+ is arguably a million times more important for the membrane potential than H+ and thus the electrophysiology!

Equation (1) in the manuscript assumes no other ions are exchanged during the experiments other than H+. This is a very bad approximation when voltage-gated potassium ion channels move the majority ion (K+) around!

In our model Figure 4A is better explained by depolarisation due to K+ channels closing than direct irreversible photodamage. Why does the THT fluorescence increase again for the second hyperpolarization event if the THT is supposed to be damaged? It does not make sense.

(4) ‘The proton motive force determines E. coli robustness to extracellular pH’, G.Terradot et al, 2024, preprint.

This article expounds the SNB model once more. It still ignores the voltage-gated ion channels. Furthermore, it ignores the effect of the dominant ion in E. coli, K+. The manuscript is incorrect as a result and I would not recommend publication.

In general, an important problem is being researched i.e. how the membrane potential of E. coli is related to motility, but there are serious flaws in the SNB approach and the experimental methodology appears tenuous.

Answers to specific questions raised by the referees

Reviewer #1 (Public Review):

Summary:

Cell-to-cell communication is essential for higher functions in bacterial biofilms. Electrical signals have proven effective in transmitting signals across biofilms. These signals are then used to coordinate cellular metabolisms or to increase antibiotic tolerance. Here, the authors have reported for the first time coordinated oscillation of membrane potential in E. coli biofilms that may have a functional role in photoprotection.

Strengths:

- The authors report original data.

- For the first time, they showed that coordinated oscillations in membrane potential occur in E. Coli biofilms.

- The authors revealed a complex two-phase dynamic involving distinct molecular response mechanisms.

- The authors developed two rigorous models inspired by 1) Hodgkin-Huxley model for the temporal dynamics of membrane potential and 2) Fire-Diffuse-Fire model for the propagation of the electric signal.

- Since its discovery by comparative genomics, the Kch ion channel has not been associated with any specific phenotype in E. coli. Here, the authors proposed a functional role for the putative K+ Kch channel : enhancing survival under photo-toxic conditions.

We thank the referee for their positive evaluations and agree with these statements.

Weaknesses:

- Since the flow of fresh medium is stopped at the beginning of the acquisition, environmental parameters such as pH and RedOx potential are likely to vary significantly during the experiment. It is therefore important to exclude the contributions of these variations to ensure that the electrical response is only induced by light stimulation. Unfortunately, no control experiments were carried out to address this issue.

The electrical responses occur almost instantaneously when the stimulation with blue light begins i.e. it is too fast to be a build of pH. We are not sure what the referee means by Redox potential since it is an attribute of all chemicals that are able to donate/receive electrons. The electrical response to stress appears to be caused by ROS, since when ROS scavengers are added the electrical response is removed i.e. pH plays a very small minority role if any.

- Furthermore, the control parameter of the experiment (light stimulation) is the same as that used to measure the electrical response, i.e. through fluorescence excitation. The use of the PROPS system could solve this problem.

>>We were enthusiastic at the start of the project to use the PROPs system in E. coli as presented by J.M.Krajl et al, ‘Electrical spiking in E. coli probed with a fluorescent voltage-indicating protein’, Science, 2011, 333, 6040, 345. However, the people we contacted in the microbiology community said that it had some technical issues and there have been no subsequent studies using PROPs in bacteria after the initial promising study. The fluorescent protein system recently presented in PNAS seems more promising, ‘Sensitive bacterial Vm sensors revealed the excitability of bacterial Vm and its role in antibiotic tolerance’, X.Jin et al, PNAS, 120, 3, e2208348120.

- Electrical signal propagation is an important aspect of the manuscript. However, a detailed quantitative analysis of the spatial dynamics within the biofilm is lacking. In addition, it is unclear if the electrical signal propagates within the biofilm during the second peak regime, which is mediated by the Kch channel. This is an important question, given that the fire-diffuse-fire model is presented with emphasis on the role of K+ ions.

We have presented a more detailed account of the electrical wavefront modelling work and it is currently under review in a physical journal, ‘Electrical signalling in three dimensional bacterial biofilms using an agent based fire-diffuse-fire model’, V.Martorelli, et al, 2024 https://www.biorxiv.org/content/10.1101/2023.11.17.567515v1

- Since deletion of the kch gene inhibits the long-term electrical response to light stimulation (regime II), the authors concluded that K+ ions play a role in the habituation response. However, Kch is a putative K+ ion channel. The use of specific drugs could help to clarify the role of K+ ions.

Our recent electrical impedance spectroscopy publication provides further evidence that Kch is associated with large changes in conductivity as expected for a voltage-gated ion channel (https://pubs.acs.org/doi/10.1021/acs.nanolett.3c04446, 'Electrical impedance spectroscopy with bacterial biofilms: neuronal-like behavior', E.Akabuogu et al, ACS Nanoletters, 2024, in print.

- The manuscript as such does not allow us to properly conclude on the photo-protective role of the Kch ion channel.

That Kch has a photoprotective role is our current working hypothesis. The hypothesis fits with the data, but we are not saying we have proven it beyond all possible doubt.

- The link between membrane potential dynamics and mechanosensitivity is not captured in the equation for the Q-channel opening dynamics in the Hodgkin-Huxley model (Supp Eq 2).

Our model is agnostic with respect to the mechanosensitivity of the ion channels, although we deduce that mechanosensitive ion channels contribute to ion channel Q.

- Given the large number of parameters used in the models, it is hard to distinguish between prediction and fitting.

This is always an issue with electrophysiological modelling (compared with most heart and brain modelling studies we are very conservative in the choice of parameters for the bacteria). In terms of predicting the different phenomena observed, we believe the model is very successful.

Reviewer #2 (Public Review):

Summary of what the authors were trying to achieve:

The authors thought they studied membrane potential dynamics in E.coli biofilms. They thought so because they were unaware that the dye they used to report that membrane potential in E.coli, has been previously shown not to report it. Because of this, the interpretation of the authors' results is not accurate.

We believe the Pilizota work is scientifically flawed.

Major strengths and weaknesses of the methods and results:

The strength of this work is that all the data is presented clearly, and accurately, as far as I can tell.

The major critical weakness of this paper is the use of ThT dye as a membrane potential dye in E.coli. The work is unaware of a publication from 2020 https://www.sciencedirect.com/science/article/pii/S0006349519308793 [sciencedirect.com] that demonstrates that ThT is not a membrane potential dye in E. coli. Therefore I think the results of this paper are misinterpreted. The same publication I reference above presents a protocol on how to carefully calibrate any candidate membrane potential dye in any given condition.

We are aware of this study, but believe it to be scientifically flawed. We do not cite the article because we do not think it is a particularly useful contribution to the literature.

I now go over each results section in the manuscript.

Result section 1: Blue light triggers electrical spiking in single E. coli cells

I do not think the title of the result section is correct for the following reasons. The above-referenced work demonstrates the loading profile one should expect from a Nernstian dye (Figure 1). It also demonstrates that ThT does not show that profile and explains why is this so. ThT only permeates the membrane under light exposure (Figure 5). This finding is consistent with blue light peroxidising the membrane (see also following work Figure 4 https://www.sciencedirect.com/science/article/pii/S0006349519303923 [sciencedirect.com] on light-induced damage to the electrochemical gradient of protons-I am sure there are more references for this).

The Pilizota group invokes some elaborate artefacts to explain the lack of agreement with a simple Nernstian battery model. The model is incorrect not the fluorophore.

Please note that the loading profile (only observed under light) in the current manuscript in Figure 1B as well as in the video S1 is identical to that in Figure 3 from the above-referenced paper (i.e. https://www.sciencedirect.com/science/article/pii/S0006349519308793 [sciencedirect.com]), and corresponding videos S3 and S4. This kind of profile is exactly what one would expect theoretically if the light is simultaneously lowering the membrane potential as the ThT is equilibrating, see Figure S12 of that previous work. There, it is also demonstrated by the means of monitoring the speed of bacterial flagellar motor that the electrochemical gradient of protons is being lowered by the light. The authors state that applying the blue light for different time periods and over different time scales did not change the peak profile. This is expected if the light is lowering the electrochemical gradient of protons. But, in Figure S1, it is clear that it affected the timing of the peak, which is again expected, because the light affects the timing of the decay, and thus of the decay profile of the electrochemical gradient of protons (Figure 4 https://www.sciencedirect.com/science/article/pii/S0006349519303923 [sciencedirect.com]).

We think the proton effect is a million times weaker than that due to potasium i.e. 0.2 M K+ versus 10-7 M H+. We can comfortably neglect the influx of H+ in our experiments.

If find Figure S1D interesting. There authors load TMRM, which is a membrane voltage dye that has been used extensively (as far as I am aware this is the first reference for that and it has not been cited https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1914430 [ncbi.nlm.nih.gov]/). As visible from the last TMRM reference I give, TMRM will only load the cells in Potassium Phosphate buffer with NaCl (and often we used EDTA to permeabilise the membrane). It is not fully clear (to me) whether here TMRM was prepared in rich media (it explicitly says so for ThT in Methods but not for TMRM), but it seems so. If this is the case, it likely also loads because of the damage to the membrane done with light, and therefore I am not surprised that the profiles are similar.

The vast majority of cells continue to be viable. We do not think membrane damage is dominating.

The authors then use CCCP. First, a small correction, as the authors state that it quenches membrane potential. CCCP is a protonophore (https://pubmed.ncbi.nlm.nih.gov/4962086 [pubmed.ncbi.nlm.nih.gov]/), so it collapses electrochemical gradient of protons. This means that it is possible, and this will depend on the type of pumps present in the cell, that CCCP collapses electrochemical gradient of protons, but the membrane potential is equal and opposite in sign to the DeltapH. So using CCCP does not automatically mean membrane potential will collapse (e.g. in some mammalian cells it does not need to be the case, but in E.coli it is https://www.biorxiv.org/content/10.1101/2021.11.19.469321v2 [biorxiv.org]). CCCP has also been recently found to be a substrate for TolC (https://journals.asm.org/doi/10.1128/mbio.00676-21 [journals.asm.org]), but at the concentrations the authors are using CCCP (100uM) that should not affect the results. However, the authors then state because they observed, in Figure S1E, a fast efflux of ions in all cells and no spiking dynamics this confirms that observed dynamics are membrane potential related. I do not agree that it does. First, Figure S1E, does not appear to show transients, instead, it is visible that after 50min treatment with 100uM CCCP, ThT dye shows no dynamics. The action of a Nernstian dye is defined. It is not sufficient that a charged molecule is affected in some way by electrical potential, this needs to be in a very specific way to be a Nernstian dye. Part of the profile of ThT loading observed in https://www.sciencedirect.com/science/article/pii/S0006349519308793 [sciencedirect.com] is membrane potential related, but not in a way that is characteristic of Nernstian dye.

Our understanding of the literature is CCCP poisons the whole metabolism of the bacterial cells. The ATP driven K+ channels will stop functioning and this is the dominant contributor to membrane potential.

Result section 2: Membrane potential dynamics depend on the intercellular distance

In this chapter, the authors report that the time to reach the first intensity peak during ThT loading is different when cells are in microclusters. They interpret this as electrical signalling in clusters because the peak is reached faster in microclusters (as opposed to slower because intuitively in these clusters cells could be shielded from light). However, shielding is one possibility. The other is that the membrane has changed in composition and/or the effective light power the cells can tolerate (with mechanisms to handle light-induced damage, some of which authors mention later in the paper) is lower. Given that these cells were left in a microfluidic chamber for 2h hours to attach in growth media according to Methods, there is sufficient time for that to happen. In Figure S12 C and D of that same paper from my group (https://ars.els-cdn.com/content/image/1-s2.0-S0006349519308793-mmc6.pdf [ars.els-cdn.com]) one can see the effects of peak intensity and timing of the peak on the permeability of the membrane. Therefore I do not think the distance is the explanation for what authors observe.

Shielding would provide the reverse effect, since hyperpolarization begins in the dense centres of the biofilms. For the initial 2 hours the cells receive negligible blue light. Neither of the referee’s comments thus seem tenable.

Result section 3: Emergence of synchronized global wavefronts in E. coli biofilms

In this section, the authors exposed a mature biofilm to blue light. They observe that the intensity peak is reached faster in the cells in the middle. They interpret this as the ion-channel-mediated wavefronts moved from the center of the biofilm. As above, cells in the middle can have different membrane permeability to those at the periphery, and probably even more importantly, there is no light profile shown anywhere in SI/Methods. I could be wrong, but the SI3 A profile is consistent with a potential Gaussian beam profile visible in the field of view. In Methods, I find the light source for the blue light and the type of microscope but no comments on how 'flat' the illumination is across their field of view. This is critical to assess what they are observing in this result section. I do find it interesting that the ThT intensity collapsed from the edges of the biofilms. In the publication I mentioned https://www.sciencedirect.com/science/article/pii/S0006349519308793#app2 [sciencedirect.com], the collapse of fluorescence was not understood (other than it is not membrane potential related). It was observed in Figure 5A, C, and F, that at the point of peak, electrochemical gradient of protons is already collapsed, and that at the point of peak cell expands and cytoplasmic content leaks out. This means that this part of the ThT curve is not membrane potential related. The authors see that after the first peak collapsed there is a period of time where ThT does not stain the cells and then it starts again. If after the first peak the cellular content leaks, as we have observed, then staining that occurs much later could be simply staining of cytoplasmic positively charged content, and the timing of that depends on the dynamics of cytoplasmic content leakage (we observed this to be happening over 2h in individual cells). ThT is also a non-specific amyloid dye, and in starving E. coli cells formation of protein clusters has been observed (https://pubmed.ncbi.nlm.nih.gov/30472191 [pubmed.ncbi.nlm.nih.gov]/), so such cytoplasmic staining seems possible.

>>It is very easy to see if the illumination is flat (Köhler illumination) by comparing the intensity of background pixels on the detector. It was flat in our case. Protons have little to do with our work for reasons highlighted before. Differential membrane permittivity is a speculative phenomenon not well supported by any evidence and with no clear molecular mechanism.

Finally, I note that authors observe biofilms of different shapes and sizes and state that they observe similar intensity profiles, which could mean that my comment on 'flatness' of the field of view above is not a concern. However, the scale bar in Figure 2A is not legible, so I can't compare it to the variation of sizes of the biofilms in Figure 2C (67 to 280um). Based on this, I think that the illumination profile is still a concern.

The referee now contradicts themselves and wants a scale bar to be more visible. We have changed the scale bar.

Result section 4: Voltage-gated Kch potassium channels mediate ion-channel electrical oscillations in E. coli

First I note at this point, given that I disagree that the data presented thus 'suggest that E. coli biofilms use electrical signaling to coordinate long-range responses to light stress' as the authors state, it gets harder to comment on the rest of the results.

In this result section the authors look at the effect of Kch, a putative voltage-gated potassium channel, on ThT profile in E. coli cells. And they see a difference. It is worth noting that in the publication https://www.sciencedirect.com/science/article/pii/S0006349519308793 [sciencedirect.com] it is found that ThT is also likely a substrate for TolC (Figure 4), but that scenario could not be distinguished from the one where TolC mutant has a different membrane permeability (and there is a publication that suggests the latter is happening https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2958.2010.07245.x [onlinelibrary.wiley.com]). Given this, it is also possible that Kch deletion affects the membrane permeability. I do note that in video S4 I seem to see more of, what appear to be, plasmolysed cells. The authors do not see the ThT intensity with this mutant that appears long after the initial peak has disappeared, as they see in WT. It is not clear how long they waited for this, as from Figure S3C it could simply be that the dynamics of this is a lot slower, e.g. Kch deletion changes membrane permeability.

The work that TolC provides a possible passive pathway for ThT to leave cells seems slightly niche. It just demonstrates another mechanism for the cells to equilibriate the concentrations of ThT in a Nernstian manner i.e. driven by the membrane voltage.

The authors themselves state that the evidence for Kch being a voltage-gated channel is indirect (line 54). I do not think there is a need to claim function from a ThT profile of E. coli mutants (nor do I believe it's good practice), given how accurate single-channel recordings are currently. To know the exact dependency on the membrane potential, ion channel recordings on this protein are needed first.

We have good evidence form electrical impedance spectroscopy experiments that Kch increases the conductivity of biofilms  (https://pubs.acs.org/doi/10.1021/acs.nanolett.3c04446, 'Electrical impedance spectroscopy with bacterial biofilms: neuronal-like behavior', E.Akabuogu et al, ACS Nanoletters, 2024, in print.

Result section 5: Blue light influences ion-channel mediated membrane potential events in E. coli

In this chapter the authors vary the light intensity and stain the cells with PI (this dye gets into the cells when the membrane becomes very permeable), and the extracellular environment with K+ dye (I have not yet worked carefully with this dye). They find that different amounts of light influence ThT dynamics. This is in line with previous literature (both papers I have been mentioning: Figure 4 https://www.sciencedirect.com/science/article/pii/S0006349519303923 [sciencedirect.com] and https://ars.els-cdn.com/content/image/1-s2.0-S0006349519308793-mmc6.pdf [ars.els-cdn.com] especially SI12), but does not add anything new. I think the results presented here can be explained with previously published theory and do not indicate that the ion-channel mediated membrane potential dynamics is a light stress relief process.

The simple Nernstian battery model proposed by Pilizota et al is erroneous in our opinion for reasons outlined above. We believe it will prove to be a dead end for bacterial electrophysiology studies.

Result section 6: Development of a Hodgkin-Huxley model for the observed membrane potential dynamics

This results section starts with the authors stating: 'our data provide evidence that E. coli manages light stress through well-controlled modulation of its membrane potential dynamics'. As stated above, I think they are instead observing the process of ThT loading while the light is damaging the membrane and thus simultaneously collapsing the electrochemical gradient of protons. As stated above, this has been modelled before. And then, they observe a ThT staining that is independent from membrane potential.

This is an erroneous niche opinion. Protons have little say in the membrane potential since there are so few of them. The membrane potential is mostly determined by K+.

I will briefly comment on the Hodgkin Huxley (HH) based model. First, I think there is no evidence for two channels with different activation profiles as authors propose. But also, the HH model has been developed for neurons. There, the leakage and the pumping fluxes are both described by a constant representing conductivity, times the difference between the membrane potential and Nernst potential for the given ion. The conductivity in the model is given as gK*n^4 for potassium, gNa*m^3*h sodium, and gL for leakage, where gK, gNa and gL were measured experimentally for neurons. And, n, m, and h are variables that describe the experimentally observed voltage-gated mechanism of neuronal sodium and potassium channels. (Please see Hodgkin AL, Huxley AF. 1952. Currents carried by sodium and potassium ions through the membrane of the giant axon of Loligo. J. Physiol. 116:449-72 and Hodgkin AL, Huxley AF. 1952. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. 117:500-44).

In the 70 years since Hodgkin and Huxley first presented their model, a huge number of similar models have been proposed to describe cellular electrophysiology. We are not being hyperbolic when we state that the HH models for excitable cells are like the Schrödinger equation for molecules. We carefully adapted our HH model to reflect the currently understood electrophysiology of E. coli.

Thus, in applying the model to describe bacterial electrophysiology one should ensure near equilibrium requirement holds (so that (V-VQ) etc terms in authors' equation Figure 5 B hold), and potassium and other channels in a given bacterium have similar gating properties to those found in neurons. I am not aware of such measurements in any bacteria, and therefore think the pump leak model of the electrophysiology of bacteria needs to start with fluxes that are more general (for example Keener JP, Sneyd J. 2009. Mathematical physiology: I: Cellular physiology. New York: Springer or https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0000144 [journals.plos.org])

The reference is to a slightly more modern version of a simple Nernstian battery model. The model will not oscillate and thus will not help modelling membrane potentials in bacteria. We are unsure where the equilibrium requirement comes from (inadequate modelling of the dynamics?)

Result section 7: Mechanosensitive ion channels (MS) are vital for the first hyperpolarization event in E. coli.

The results that Mcs channels affect the profile of ThT dye are interesting. It is again possible that the membrane permeability of these mutants has changed and therefore the dynamics have changed, so this needs to be checked first. I also note that our results show that the peak of ThT coincides with cell expansion. For this to be understood a model is needed that also takes into account the link between maintenance of electrochemical gradients of ions in the cell and osmotic pressure.

The evidence for permeability changes in the membranes seems to be tenuous.

A side note is that the authors state that the Msc responds to stress-related voltage changes. I think this is an overstatement. Mscs respond to predominantly membrane tension and are mostly nonspecific (see how their action recovers cellular volume in this publication https://www.pnas.org/doi/full/10.1073/pnas.1522185113 [pnas.org]). Authors cite references 35-39 to support this statement. These publications still state that these channels are predominantly membrane tension-gated. Some of the references state that the presence of external ions is important for tension-related gating but sometimes they gate spontaneously in the presence of certain ions. Other publications cited don't really look at gating with respect to ions (39 is on clustering). This is why I think the statement is somewhat misleading.

We have reworded the discussion of Mscs since the literature appears to be ambiguous. We will try to run some electrical impedance spectroscopy experiments on the Msc mutants in the future to attempt to remove the ambiguity.

Result section 8: Anomalous ion-channel-mediated wavefronts propagate light stress signals in 3D E. coli biofilms.

I am not commenting on this result section, as it would only be applicable if ThT was membrane potential dye in E. coli.

Ok, but we disagree on the use of ThT.

Aims achieved/results support their conclusions:

The authors clearly present their data. I am convinced that they have accurately presented everything they observed. However, I think their interpretation of the data and conclusions is inaccurate in line with the discussion I provided above.

Likely impact of the work on the field, and the utility of the methods and data to the community:

I do not think this publication should be published in its current format. It should be revised in light of the previous literature as discussed in detail above. I believe presenting it in it's current form on eLife pages would create unnecessary confusion.

We believe many of the Pilizota group articles are scientifically flawed and are causing the confusion in the literature.

Any other comments:

I note, that while this work studies E. coli, it references papers in other bacteria using ThT. For example, in lines 35-36 authors state that bacteria (Bacillus subtilis in this case) in biofilms have been recently found to modulate membrane potential citing the relevant literature from 2015. It is worth noting that the most recent paper https://journals.asm.org/doi/10.1128/mbio.02220-23 [journals.asm.org] found that ThT binds to one or more proteins in the spore coat, suggesting that it does not act as a membrane potential in Bacillus spores. It is possible that it still reports membrane potential in Bacillus cells and the recent results are strictly spore-specific, but these should be kept in mind when using ThT with Bacillus.

>>ThT was used successfully in previous studies of normal B. subtilis cells (by our own group and A.Prindle, ‘Spatial propagation of electrical signal in circular biofilms’, J.A.Blee et al, Physical Review E, 2019, 100, 052401, J.A.Blee et al, ‘Membrane potentials, oxidative stress and the dispersal response of bacterial biofilms to 405 nm light’, Physical Biology, 2020, 17, 2, 036001, A.Prindle et al, ‘Ion channels enable electrical communication in bacterial communities’, Nature, 2015, 527, 59-63). The connection to low metabolism pore research seems speculative.

Reviewer #3 (Public Review):

It has recently been demonstrated that bacteria in biofilms show changes in membrane potential in response to changes in their environment, and that these can propagate signals through the biofilm to coordinate bacterial behavior. Akabuogu et al. contribute to this exciting research area with a study of blue light-induced membrane potential dynamics in E. coli biofilms. They demonstrate that Thioflavin-T (ThT) intensity (a proxy for membrane potential) displays multiphasic dynamics in response to blue light treatment. They additionally use genetic manipulations to implicate the potassium channel Kch in the latter part of these dynamics. Mechanosensitive ion channels may also be involved, although these channels seem to have blue light-independent effects on membrane potential as well. In addition, there are challenges to the quantitative interpretation of ThT microscopy data which require consideration. The authors then explore whether these dynamics are involved in signaling at the community level. The authors suggest that cell firing is both more coordinated when cells are clustered and happens in waves in larger, 3D biofilms; however, in both cases evidence for these claims is incomplete. The authors present two simulations to describe the ThT data. The first of these simulations, a Hodgkin-Huxley model, indicates that the data are consistent with the activity of two ion channels with different kinetics; the Kch channel mutant, which ablates a specific portion of the response curve, is consistent with this. The second model is a fire-diffuse-fire model to describe wavefront propagation of membrane potential changes in a 3D biofilm; because the wavefront data are not presented clearly, the results of this model are difficult to interpret. Finally, the authors discuss whether these membrane potential changes could be involved in generating a protective response to blue light exposure; increased death in a Kch ion channel mutant upon blue light exposure suggests that this may be the case, but a no-light control is needed to clarify this.

In a few instances, the paper is missing key control experiments that are important to the interpretation of the data. This makes it difficult to judge the meaning of some of the presented experiments.

(1) An additional control for the effects of autofluorescence is very important. The authors conduct an experiment where they treat cells with CCCP and see that Thioflavin-T (ThT) dynamics do not change over the course of the experiment. They suggest that this demonstrates that autofluorescence does not impact their measurements. However, cellular autofluorescence depends on the physiological state of the cell, which is impacted by CCCP treatment. A much simpler and more direct experiment would be to repeat the measurement in the absence of ThT or any other stain. This experiment should be performed both in the wild-type strain and in the ∆kch mutant.

ThT is a very bright fluorophore (much brighter than a GFP). It is clear from the images of non-stained samples that autofluorescence provides a negligible contribution to the fluorescence intensity in an image.

(2) The effects of photobleaching should be considered. Of course, the intensity varies a lot over the course of the experiment in a way that photobleaching alone cannot explain. However, photobleaching can still contribute to the kinetics observed. Photobleaching can be assessed by changing the intensity, duration, or frequency of exposure to excitation light during the experiment. Considerations about photobleaching become particularly important when considering the effect of catalase on ThT intensity. The authors find that the decrease in ThT signal after the initial "spike" is attenuated by the addition of catalase; this is what would be predicted by catalase protecting ThT from photobleaching (indeed, catalase can be used to reduce photobleaching in time lapse imaging).

Photobleaching was negligible over the course of the experiments. We employed techniques such as reducing sample exposure time and using the appropriate light intensity to minimize photobleaching.

(3) It would be helpful to have a baseline of membrane potential fluctuations in the absence of the proposed stimulus (in this case, blue light). Including traces of membrane potential recorded without light present would help support the claim that these changes in membrane potential represent a blue light-specific stress response, as the authors suggest. Of course, ThT is blue, so if the excitation light for ThT is problematic for this experiment the alternative dye tetramethylrhodamine methyl ester perchlorate (TMRM) can be used instead.

Unfortunately the fluorescent baseline is too weak to measure cleanly in this experiment. It appears the collective response of all the bacteria hyperpolarization at the same time appears to dominate the signal (measurements in the eLife article and new potentiometry measurements).

(4) The effects of ThT in combination with blue light should be more carefully considered. In mitochondria, a combination of high concentrations of blue light and ThT leads to disruption of the PMF (Skates et al. 2021 BioRXiv), and similarly, ThT treatment enhances the photodynamic effects of blue light in E. coli (Bondia et al. 2021 Chemical Communications). If present in this experiment, this effect could confound the interpretation of the PMF dynamics reported in the paper.

We think the PMF plays a minority role in determining the membrane potential in E. coli. For reasons outlined before (H+ is a minority ion in E. coli compared with K+).

(5) Figures 4D - E indicate that a ∆kch mutant has increased propidium iodide (PI) staining in the presence of blue light; this is interpreted to mean that Kch-mediated membrane potential dynamics help protect cells from blue light. However, Live/Dead staining results in these strains in the absence of blue light are not reported. This means that the possibility that the ∆kch mutant has a general decrease in survival (independent of any effects of blue light) cannot be ruled out.

>>Both strains of bacterial has similar growth curve and also engaged in membrane potential dynamics for the duration of the experiment. We were interested in bacterial cells that observed membrane potential dynamics in the presence of the stress. Bacterial cells need to be alive to engage in membrane potential  dynamics (hyperpolarize) under stress conditions. Cells that engaged in membrane potential dynamics and later stained red were only counted after the entire duration. We believe that the wildtype handles the light stress better than the ∆kch mutant as measured with the PI.

(6) Additionally in Figures 4D - E, the interpretation of this experiment can be confounded by the fact that PI uptake can sometimes be seen in bacterial cells with high membrane potential (Kirchhoff & Cypionka 2017 J Microbial Methods); the interpretation is that high membrane potential can lead to increased PI permeability. Because the membrane potential is largely higher throughout blue light treatment in the ∆kch mutant (Fig. 3AB), this complicates the interpretation of this experiment.

Kirchhoff & Cypionka 2017 J Microbial Methods, using fluorescence microscopy, suggested that changes in membrane potential dynamics can introduce experimental bias when propidium iodide is used to confirm the viability of tge bacterial strains, B subtilis (DSM-10) and Dinoroseobacter shibae, that are starved of oxygen (via N2 gassing) for 2 hours. They attempted to support their findings by using CCCP in stopping the membrane potential dynamics (but never showed any pictoral or plotted data for this confirmatory experiment). In our experiment methodology, cell death was not forced on the cells by introducing an extra burden or via anoxia. We believe that the accumulation of PI in ∆kch mutant is not due to high membrane potential dynamics but is attributed to the PI, unbiasedly showing damaged/dead cells. We think that propidium iodide is good for this experiment. Propidium iodide is a dye that is extensively used in life sciences. PI has also been used in the study of bacterial electrophysiology (https://pubmed.ncbi.nlm.nih.gov/32343961/, ) and no membrane potential related bias was reported.

Throughout the paper, many ThT intensity traces are compared, and described as "similar" or "dissimilar", without detailed discussion or a clear standard for comparison. For example, the two membrane potential curves in Fig. S1C are described as "similar" although they have very different shapes, whereas the curves in Fig. 1B and 1D are discussed in terms of their differences although they are evidently much more similar to one another. Without metrics or statistics to compare these curves, it is hard to interpret these claims. These comparative interpretations are additionally challenging because many of the figures in which average trace data are presented do not indicate standard deviation.

Comparison of small changes in the absolute intensities is problematic in such fluorescence experiments. We mean the shape of the traces is similar and they can be modelled using a HH model with similar parameters.

The differences between the TMRM and ThT curves that the authors show in Fig. S1C warrant further consideration. Some of the key features of the response in the ThT curve (on which much of the modeling work in the paper relies) are not very apparent in the TMRM data. It is not obvious to me which of these traces will be more representative of the actual underlying membrane potential dynamics.

In our experiment, TMRM was used to confirm the dynamics observed using ThT. However, ThT appear to be more photostable than TMRM (especially towars the 2nd peak). The most interesting observation is that with both dyes, all phases of the membrane potential dynamics were conspicuous (the first peak, the quiescent period and the second peak). The time periods for these three episodes were also similar.

A key claim in this paper (that dynamics of firing differ depending on whether cells are alone or in a colony) is underpinned by "time-to-first peak" analysis, but there are some challenges in interpreting these results. The authors report an average time-to-first peak of 7.34 min for the data in Figure 1B, but the average curve in Figure 1B peaks earlier than this. In Figure 1E, it appears that there are a handful of outliers in the "sparse cell" condition that likely explain this discrepancy. Either an outlier analysis should be done and the mean recomputed accordingly, or a more outlier-robust method like the median should be used instead. Then, a statistical comparison of these results will indicate whether there is a significant difference between them.

The key point is the comparison of standard errors on the standard deviation.

In two different 3D biofilm experiments, the authors report the propagation of wavefronts of membrane potential; I am unable to discern these wavefronts in the imaging data, and they are not clearly demonstrated by analysis.

The first data set is presented in Figures 2A, 2B, and Video S3. The images and video are very difficult to interpret because of how the images have been scaled: the center of the biofilm is highly saturated, and the zero value has also been set too high to consistently observe the single cells surrounding the biofilm. With the images scaled this way, it is very difficult to assess dynamics. The time stamps in Video S3 and on the panels in Figure 2A also do not correspond to one another although the same biofilm is shown (and the time course in 2B is also different from what is indicated in 2B). In either case, it appears that the center of the biofilm is consistently brighter than the edges, and the intensity of all cells in the biofilm increases in tandem; by eye, propagating wavefronts (either directed toward the edge or the center) are not evident to me. Increased brightness at the center of the biofilm could be explained by increased cell thickness there (as is typical in this type of biofilm). From the image legend, it is not clear whether the image presented is a single confocal slice or a projection. Even if this is a single confocal slice, in both Video S3 and Figure 2A there are regions of "haze" from out-of-focus light evident, suggesting that light from other focal planes is nonetheless present. This seems to me to be a simpler explanation for the fluorescence dynamics observed in this experiment: cells are all following the same trajectory that corresponds to that seen for single cells, and the center is brighter because of increased biofilm thickness.

We appreciate the reviewer for this important observation. We have made changes to the figures to address this confusion. The cell cover has no influence on the observed membrane potential dynamics. The entire biofilm was exposed to the same blue light at each time. Therefore all parts of the biofilm received equal amounts of the blue light intensity. The membrane potential dynamics was not influenced by cell density (see Fig 2C). 

The second data set is presented in Video S6B; I am similarly unable to see any wave propagation in this video. I observe only a consistent decrease in fluorescence intensity throughout the experiment that is spatially uniform (except for the bright, dynamic cells near the top; these presumably represent cells that are floating in the microfluidic and have newly arrived to the imaging region).

A visual inspection of Video S6B shows a fast rise, a decrease in fluorescence and a second rise (supplementary figure 4B). The data for the fluorescence was carefully obtained using the imaris software. We created a curved geometry on each slice of the confocal stack. We analyzed the surfaces of this curved plane along the z-axis. This was carried out in imaris.

3D imaging data can be difficult to interpret by eye, so it would perhaps be more helpful to demonstrate these propagating wavefronts by analysis; however, such analysis is not presented in a clear way. The legend in Figure 2B mentions a "wavefront trace", but there is no position information included - this trace instead seems to represent the average intensity trace of all cells. To demonstrate the propagation of a wavefront, this analysis should be shown for different subpopulations of cells at different positions from the center of the biofilm. Data is shown in Figure 8 that reflects the velocity of the wavefront as a function of biofilm position; however, because the wavefronts themselves are not evident in the data, it is difficult to interpret this analysis. The methods section additionally does not contain sufficient information about what these velocities represent and how they are calculated. Because of this, it is difficult for me to evaluate the section of the paper pertaining to wave propagation and the predicted biofilm critical size.

The analysis is considered in more detail in a more expansive modelling article, currently under peer review in a physics journal, ‘Electrical signalling in three dimensional bacterial biofilms using an agent based fire-diffuse-fire model’, V.Martorelli, et al, 2024 https://www.biorxiv.org/content/10.1101/2023.11.17.567515v1

There are some instances in the paper where claims are made that do not have data shown or are not evident in the cited data:

(1) In the first results section, "When CCCP was added, we observed a fast efflux of ions in all cells"- the data figure pertaining to this experiment is in Fig. S1E, which does not show any ion efflux. The methods section does not mention how ion efflux was measured during CCCP treatment.

We have worded this differently to properly convey our results.

(2) In the discussion of voltage-gated calcium channels, the authors refer to "spiking events", but these are not obvious in Figure S3E. Although the fluorescence intensity changes over time, it's hard to distinguish these fluctuations from measurement noise; a no-light control could help clarify this.

The calcium transients observed were not due to noise or artefacts.

(3) The authors state that the membrane potential dynamics simulated in Figure 7B are similar to those observed in 3D biofilms in Fig. S4B; however, the second peak is not clearly evident in Fig. S4B and it looks very different for the mature biofilm data reported in Fig. 2. I have some additional confusion about this data specifically: in the intensity trace shown in Fig. S4B, the intensity in the second frame is much higher than the first; this is not evident in Video S6B, in which the highest intensity is in the first frame at time 0. Similarly, the graph indicates that the intensity at 60 minutes is higher than the intensity at 4 minutes, but this is not the case in Fig. S4A or Video S6B.

The confusion stated here has now been addressed. Also it should be noted that while Fig 2.1 was obtained with LED light source, Fig S4A was obtained using a laser light source. While obtaining the confocal images (for Fig S4A ), the light intensity was controlled to further minimize photobleaching. Most importantly, there is an evidence of slow rise to the 2nd peak in Fig S4B. The first peak, quiescence and slow rise to second peak are evident.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Scientific recommendations:

- Although Fig 4A clearly shows that light stimulation has an influence on the dynamics of cell membrane potential in the biofilm, it is important to rule out the contribution of variations in environmental parameters. I understand that for technical reasons, the flow of fresh medium must be stopped during image acquisition. Therefore, I suggest performing control experiments, where the flow is stopped before image acquisition (15min, 30min, 45min, and 1h before). If there is no significant contribution from environmental variations (pH, RedOx), the dynamics of the electrical response should be superimposed whatever the delay between stopping the flow stop and switching on the light.

In this current research study, we were focused on studying how E. coli cells and biofilms react to blue light stress via their membrane potential dynamics. This involved growing the cells and biofilms, stopping the media flow and obtaining data immediately. We believe that stopping the flow not only helped us to manage data acquisition, it also helped us reduce the effect of environmental factors. In our future study we will expand the work to include how the membrane potential dynamics evolve in the presence of changing environmental factors for example such induced by stopping the flow at varied times.

- Since TMRM signal exhibits a linear increase after the first response peak (Supplementary Figure 1D), I recommend mitigating the statement at line 78.

- To improve the spatial analysis of the electrical response, I suggest plotting kymographs of the intensity profiles across the biofilm. I have plotted this kymograph for Video S3 and it appears that there is no electrical propagation for the second peak. In addition, the authors should provide technical details of how R^2(t) is measured in the first regime (Figure 7E).

See the dedicated simulation article for more details. https://www.biorxiv.org/content/10.1101/2023.11.17.567515v1

- Line 152: To assess the variability of the latency, the authors should consider measuring the variance divided by the mean instead of SD, which may depend on the average value.

We are happy with our current use of standard error on the standard deviation. It shows what we claim to be true.

- Line 154-155: To truly determine whether the amplitude of the "action potential" is independent of biofilm size, the authors should not normalise the signals.

Good point. We qualitatively compared both normalized and unnormalized data. Recent electrical impedance spectroscopy measurements (unpublished) indicate that the electrical activity is an extensive quantity i.e. it scales with the size of the biofilms.

- To precise the role of K+ in the habituation response, I suggest using valinomycin at sub-inhibitory concentrations (10µM). Besides, the high concentration of CCCP used in this study completely inhibits cell activity. Not surprisingly, no electrical response to light stimulation was observed in the presence of CCCP. Finally, the Kch complementation experiment exhibits a "drop after the first peak" on a single point. It would be more convincing to increase the temporal resolution (1min->10s) to show that there is indeed a first and a second peak.

An interesting experiment for the future.

- Line 237-238: There are only two points suggesting that the dynamics of hyperpolarization are faster at higher irradiance(Fig 4A). The authors should consider adding a third intermediate point at 17µW/mm^2 to confirm the statement made in this sentence.

Multiple repeats were performed. We are confident of the robustness of our data.

- Line 249 + Fig 4E: It seems that the data reported on Fig 4E are extracted from Fig 4D. If this is indeed the case, the data should be normalised by the total population size to compare survival probabilities under the two conditions. It would also be great to measure these probabilities (for WT and ∆kch) in the presence of ROS scavengers.

- To distinguish between model fitting and model predictions, the authors should clearly state which parameters are taken from the literature and which parameters are adjusted to fit the experimental data.

- Supplementary Figure 4A: why can't we see any wavefront in this series of images?

For the experimental data, the wavefront was analyzed by employing the imaris software. We systematically created a ROI with a curved geometry within the confocal stack (the biofilm). The fluorescence of ThT was traced along the surface of the curved geometry was analyzed along the z-axis.

- Fig 7B: Could the authors explain why the plateau is higher in the simulations than in the biofilm experiments? Could they add noise on the firing activities?

See the dedicated Martorelli modelling article. In general we would need to approach stochastic Hodgkin-Huxley modelling and the fluorescence data (and electrical impedance spectroscopy data) presented does not have extensive noise (due to collective averaging over many bacteria cells).

- Supplementary Figure 4B: Why can't we see the second peak in confocal images?

The second peak is present although not as robust as in Fig 2B. The confocal images were obtained with a laser source. Therefore we tried to create a balance between applying sufficient light stress on the bacterial cells and mitigating photobleaching.

Editing recommendations:

The editing recommendations below has been applied where appropriate

- Many important technical details are missing (e.g. R^2, curvature, and 445nm irradiance measurements). Error bars are missing from most graphs. The captions should clearly indicate if these are single-cell or biofilm experiments, strain name, illumination conditions, number of experiments, SD, or SE. Please indicate on all panels of all figures in the main text and in the supplements, which are the conditions: single cell vs. biofilm, strains, medium, centrifugal vs centripetal etc..., where relevant. Please also draw error bars everywhere.

We have now made appropriate changes. We specifically use cells when we were dealing with single cells and biofilms when we worked on biofilms. We decided to describe the strain name either on the panel or the image description.

- Line 47-51: The way the paragraph is written suggests that no coordinated electrical oscillations have been observed in Gram-negative biofilms. However, Hennes et al (referenced as 57 in this manuscript) have shown that a wave of hyperpolarized cells propagates in Neisseria gonorrhoea colony, which is a Gram-negative bacterium.

We are now aware of this work. It was not published when we first submitted our work and the authors claim the waves of activity are due to ROS diffusion NOT propagating waves of ions (coordinated electrical wavefronts).

- Line 59: "stressor" -> "stress" or "perturbation".

The correction has been made.

- Line 153: Please indicate in the Material&Methods how the size of the biofilm is measured.

The biofilm size was obtained using BiofilmQ and the step by step guide for using BiofilmQ were stated..

- Figure 2A: Please provide associated brightfield images to locate bacteria.

- Line 186: Please remove "wavefront" from the caption. Fig2B only shows the average signal as a function of time.

This correction has been implemented.

- Fig 3B,C: Please indicate single cell and biofilm on the panels and also WT and ∆kch.

- Line 289: I suggest adding "in single cell experiments" to the title of this section.

- Fig 5A: blue light is always present at regular time intervals during regime I and II. The presence of blue light only in regime I could be misleading.

- Fig 5C: The curve in Fig 5D seems to correspond to the biofilm case. The curve given by the model, should be compared with the average curve presented in Fig 1D.

- Fig 6A, B, and C: These figures could be moved to supplements.

- Line 392: Replace "turgidity" with "turgor pressure".

- Fig 7C,E: Please use a log-log scale to represent these data and indicate the line of slope 1.

- Fig 7E: The x-axis has been cropped.

- Please provide a supplementary movie for the data presented in Fig 7E.

- Line 455: E. Coli biofilms do not express ThT.

- Line 466: "\gamma is the anomalous exponent". Please remove anomalous (\gamma can equal 1 at this stage).

- Line 475: Please replace "section" with "projection".

- Line 476: Please replace "spatiotemporal" with "temporal". There is no spatial dependency in either figure.

- Line 500: Please define Eikonal approximation.

- Fig 8 could be moved to supplements.

- Line 553: "predicted" -> "predict".

- Line 593: Could the authors explain why their model offers much better quantitative agreement?

- Line 669: What does "universal" mean in that context?

- Line 671: A volume can be pipetted but not a concentration.

- Line 676: Are triplicates technical or biological replicates?

- Sup Fig1: Please use minutes instead of seconds in panel A.

- Model for membrane dynamics: "The fraction of time the Q+ channel is open" -> "The dynamics of Q+ channel activity can be written". Ditto for K+ channel...

- Model for membrane dynamics: "the term ... is a threshold-linear". This function is not linear at all. Why is it called linear? Also, please describe what \sigma is.

- ABFDF model: "releasing a given concentration" -> "releasing a local concentration" or "a given number" but it's not \sigma anymore. Besides, this \sigma is unlikely related to the previous \sigma used in the model of membrane potential dynamics in single cells. Please consider renaming one or the other. Also, ions are referred to as C+ in the text and C in equation 8. Am I missing something?

Reviewer #2 (Recommendations For The Authors):

I have included all my comments as one review. I have done so, despite the fact that some minor comments could have gone into this section, because I decided to review each Result section. I thus felt that not writing it as one review might be harder to follow. I have however highlighted which comments are minor suggestions or where I felt corrections.

However, while I am happy with all my comments being public, given their nature I think they should be shown to authors first. Perhaps the authors want to go over them and think about it before deciding if they are happy for their manuscript to be published along with these comments, or not. I will highlight this in an email to the editor. I question whether in this case, given that I am raising major issues, publishing both the manuscript and the comments is the way to go as I think it might just generate confusion among the audience.

Reviewer #3 (Recommendations For The Authors):

I was unable to find any legends for any of the supplemental videos in my review materials, and I could not open supplemental video 5.

I made some comments in the public review about the analysis and interpretation of the time-to-fire data. One of the other challenges in this data set is that the time resolution is limited- it seems that a large proportion of cells have already fired after a single acquisition frame. It would be ideal to increase the time resolution on this measurement to improve precision. This could be done by imaging more quickly, but that would perhaps necessitate more blue light exposure; an alternative is to do this experiment under lower blue light irradiance where the first spike time is increased (Figure 4A).

In the public review, I mentioned the possible impact of high membrane potential on PI permeability. To address this, the experiment could be repeated with other stains, or the viability of blue light-treated cells could be addressed more directly by outgrowth or colony-forming unit assays.

In the public review, I mentioned the possible combined toxicity of ThT and blue light. Live/dead experiments after blue light exposure with and without ThT could be used to test for such effects, and/or the growth curve experiment in Figure 1F could be repeated with blue light exposure at a comparable irradiance used in the experiment.

Throughout the paper and figure legends, it would help to have more methodological details in the main text, especially those that are critical for the interpretation of the experiment. The experimental details in the methods section are nicely described, but the data analysis section should be expanded significantly.

At the end of the results section, the authors suggest a critical biofilm size of only 4 µm for wavefront propagation (not much larger than a single cell!). The authors show responses for various biofilm sizes in Fig. 2C, but these are all substantially larger. Are there data for cell clusters above and below this size that could support this claim more directly?

The authors mention image registration as part of their analysis pipeline, but the 3D data sets in Video S6B and Fig. S4A do not appear to be registered- were these registered prior to the velocity analysis reported in Fig. 8?

One of the most challenging claims to demonstrate in this paper is that these membrane potential wavefronts are involved in coordinating a large, biofilm-scale response to blue light. One possible way to test this might be to repeat the Live/Dead experiment in planktonic culture or the single-cell condition. If the protection from blue light specifically emerges due to coordinated activity of the biofilm, the Kch mutant would not be expected to show a change in Live/Dead staining in non-biofilm conditions.

Line 140: How is "mature biofilm" defined? Also on this same line, what does "spontaneous" mean here?

Line 151: "much smaller": Given that the reported time for 3D biofilms is 2.73 {plus minus} 0.85 min and in microclusters is 3.27 {plus minus} 1.77 min, this seems overly strong.

Line 155: How is "biofilm density" characterized? Additionally, the data in Figure 2C are presented in distance units (µm), but the text refers to "areal coverage"- please define the meaning of these distance units in the legend and/or here in the text (is this the average radius?).

Lines 161-162: These claims seem strong given the data presented before, and the logic is not very explicit. For example, in the second sentence, the idea that this signaling is used to "coordinate long-range responses to light stress" does not seem strongly evidenced at this point in the paper. What is meant by a long-range response to light stress- are there processes to respond to light that occur at long-length scales (rather than on the single-cell scale)? If so, is there evidence that these membrane potential changes could induce these responses? Please clarify the logic behind these conclusions.

Lines 235-236: In the lower irradiance conditions, the responses are slower overall, and it looks like the ThT intensity is beginning to rise at the end of the measurement. Could a more prominent second peak be observed in these cases if the measurement time was extended?

Line 242-243: The overall trajectories of extracellular potassium are indeed similar, but the kinetics of the second peak of potassium are different than those observed by ThT (it rises some minutes earlier)- is this consistent with the idea that Kch is responsible for that peak? Additionally, the potassium dynamics also reflect the first peak- is this surprising given that the Kch channel has no effect on this peak?

Line 255-256: Again, this seems like a very strong claim. There are several possible interpretations of the catalase experiment (which should be discussed); this experiment perhaps suggests that ROS impacts membrane potential, but does not obviously indicate that these membrane potential fluctuations mitigate ROS levels or help the cells respond to ROS stress. The loss of viability in the ∆kch mutant might indicate a link between these membrane potential experiments and viability, but it is hard to interpret without the no-light control I mention in the public review.

Lines 313-315: "The model predicts... the external light stress". Please clarify this section. Where this prediction arises from in the modeling work? Second, I am not sure what is meant by "modulates the light stress" or "keeps the cell dynamics robust to the intensity of external light stress" (especially since the dynamics clearly vary with irradiance, as seen in Figure 4A).

Line 322: I am not sure what "handles the ROS by adjusting the profile of the membrane potential dynamics" means. What is meant by "handling" ROS? Is the hypothesis that membrane potential dynamics themselves are protective against ROS, or that they induce a ROS-protective response downstream, or something else? Later in lines 327-8 the authors write that changes in the response to ROS in the model agree with the hypothesis, but just showing that ROS impacts the membrane potential does not seem to demonstrate that this has a protective effect against ROS.

Line 365-366: This section title seems confusing- mechanosensitive ion channels totally ablate membrane potential dynamics, they don't have a specific effect on the first hyperpolarization event. The claim that mechanonsensitive ion channels are specifically involved in the first event also appears in the abstract.

Also, the apparent membrane potential is much lower even at the start of the experiment in these mutants- is this expected? This seems to imply that these ion channels also have a blue light independent effect.

Lines 368, 371: Should be VGCCs rather than VGGCs.

Line 477: I believe the figure reference here should be to Figure 7B, not 6B.

Line 567-568: "The initial spike is key to registering the presence of the light stress." What is the evidence for this claim?

Line 592-594: "We have presented much better quantitative agreement..." This is a strong claim; it is not immediately evident to me that the agreement between model and prediction is "much better" in this work than in the cited work. The model in Figure 4 of reference 57 seems to capture the key features of their data. Clarification is needed about this claim.

Line 613: "...strains did not have any additional mutations." This seems to imply that whole genome sequencing was performed- is this the case?

Line 627: I believe this should refer to Figure S2A-B rather than S1.

Line 719: What percentage of cells did not hyperpolarize in these experiments?

Lines 751-754: As I mentioned above, significant detail is missing here about how these measurements were made. How is "radius" defined in 3D biofilms like the one shown in Video S6B, which looks very flat? What is meant by the distance from the substrate to the core, since usually in this biofilm geometry, the core is directly on the substrate? Most importantly, this only describes the process of sectioning the data- how were these sections used to compute the velocity of ThT signal propagation?

I also have some comments specifically on the figure presentation:

Normalization from 0 to 1 has been done in some of the ThT traces in the paper, but not all. The claims in the paper would be easiest to evaluate if the non-normalized data were shown- this is important for the interpretation of some of the claims.

Some indication of standard deviation (error bars or shading) should be added to all figures where mean traces are plotted.

Throughout the paper, I am a bit confused by the time axis; the data consistently starts at 1 minute. This is not intuitive to me, because it seems that the blue light being applied to the cells is also the excitation laser for ThT- in that case, shouldn't the first imaging frame be at time 0 (when the blue light is first applied)? Or is there an additional exposure of blue light 1 minute before imaging starts? This is consequential because it impacts the measured time to the first spike. (Additionally, all of the video time stamps start at 0).

Please increase the size of the scale bars and bar labels throughout, especially in Figure 2A and S4A.

In Figure 1B and D, it would help to decrease the opacity on the individual traces so that more of them can be discerned. It would also improve clarity to have data from the different experiments shown with different colored lines, so that variability between experiments can be clearly visualized.

Results in Figure 1E would be easier to interpret if the frequency were normalized to total N. It is hard to tell from this graph whether the edges and bin widths are the same between the data sets, but if not, they should be. Also, it would help to reduce the opacity of the sparse cell data set so that the full microcluster data set can be seen as well.

Biofilm images are shown in Figures 2A, S3A, and Video S3- these are all of the same biofilm. Why not take the opportunity to show different experimental replicates in these different figures? The same goes for Figure S4A and Video S6B, which again are of the same biofilm.

Figure 2C would be much easier to read if the curves were colored in order of their size; the same is true for Figure 4A and irradiance.

The complementation data in Figure S3D should be moved to the main text figure 3 alongside the data about the corresponding knockout to make it easier to compare the curves.

Fig.ure S3E: Is the Y-axis in this graph mislabeled? It is labeled as ThT fluorescence, but it seems that it is reporting fluorescence from the calcium indicator?

Video S6B is very confusing - why does the video play first forwards and then backwards? Unless I am looking very carefully at the time stamps it is easy to misinterpret this as a rise in the intensity at the end of the experiment. Without a video legend, it's hard to understand this, but I think it would be much more straightforward to interpret if it only played forward. (Also, why is this video labeled 6B when there is no video 6A?)

eLife Assessment

This is a fundamental study that provides a detailed single-cell transcriptomic and epigenomic map of the mouse trabecular meshwor, identifying three distinct trabecular meshwor subtypes with specific functional roles. It links the glaucoma-associated transcription factor LMX1B to mitochondrial regulation in TM3 cells and demonstrates that nicotinamide treatment prevents IOP elevation in Lmx1bV265D/+ mutant mice, highlighting a potential metabolic therapeutic strategy for glaucoma. This convincing work would be further supported by data that link the transcriptional data with mitochondrial functional assays.

Reviewer #1 (Public review):

Summary:

This study provides a comprehensive single-cell and multiomic characterization of trabecular meshwork (TM) cells in the mouse eye, a structure critical to intraocular pressure (IOP) regulation and glaucoma pathogenesis. Using scRNA-seq, snATAC-seq, immunofluorescence, and in situ hybridization, the authors identify three transcriptionally and spatially distinct TM cell subtypes. The study further demonstrates that mitochondrial dysfunction, specifically in one subtype (TM3), contributes to elevated IOP in a genetic mouse model of glaucoma carrying a mutation in the transcription factor Lmx1b. Importantly, treatment with nicotinamide (vitamin B3), known to support mitochondrial health, prevents IOP elevation in this model. The authors also link their findings to human datasets, suggesting the existence of analogous TM3-like cells with potential relevance to human glaucoma.

Strengths:

The study is methodologically rigorous, integrating single-cell transcriptomic and chromatin accessibility profiling with spatial validation and in vivo functional testing. The identification of TM subtypes is consistent across mouse strains and institutions, providing robust evidence of conserved TM cell heterogeneity. The use of a glaucoma model to show subtype-specific vulnerability, combined with a therapeutic intervention-gives the study strong mechanistic and translational significance. The inclusion of chromatin accessibility data adds further depth by implicating active transcription factors such as LMX1B, a gene known to be associated with glaucoma risk. The integration with human single-cell datasets enhances the potential relevance of the findings to human disease.

Weaknesses:

Although the LMX1B transcription factor is implicated as a key regulator in TM3 cells, its role in directly controlling mitochondrial gene expression is not fully explored. Additional analysis of motif accessibility or binding enrichment near relevant target genes could substantiate this mechanistic link. The therapeutic effect of vitamin B3 is clearly demonstrated phenotypically, but the underlying cellular and molecular mechanisms remain somewhat underdeveloped - for instance, changes in mitochondrial function, oxidative stress markers, or NAD+ levels are not directly measured. While the human relevance of TM3 cells is suggested through marker overlap, more quantitative approaches, such as cell identity mapping or gene signature scoring in human datasets, would strengthen the translational connection.

Overall, this is a compelling and carefully executed study that offers significant advances in our understanding of TM cell biology and its role in glaucoma. The integration of multimodal data, disease modeling, and therapeutic testing represents a valuable contribution to the field. With additional mechanistic depth, the study has the potential to become a foundational resource for future research into IOP regulation and glaucoma treatment.

Reviewer #2 (Public review):

Summary:

This elegant study by Tolman and colleagues provides fundamental findings that substantially advance our knowledge of the major cell types within the limbus of the mouse eye, focusing on the aqueous humor outflow pathway. The authors used single-cell and single-nuclei RNAseq to very clearly identify 3 subtypes of the trabecular meshwork (TM) cells in the mouse eye, with each subtype having unique markers and proposed functions. The U. Columbia results are strengthened by an independent replication in a different mouse strain at a separate laboratory (Duke). Bioinformatics analyses of these expression data were used to identify cellular compartments, molecular functions, and biological processes. Although there were some common pathways among the 3 subtypes of TM cells (e.g., ECM metabolism), there also were distinct functions. For example:

• TM1 cell expression supports heavy engagement in ECM metabolism and structure, as well as TGFβ2 signaling.

• TM2 cells were enriched in laminin and pathways involved in phagocytosis, lysosomal function, and antigen expression, as well as End3/VEGF/angiopoietin signaling.

• TM3 cells were enriched in actin binding and mitochondrial metabolism.

They used high-resolution immunostaining and in situ hybridization to show that these 3 TM subtypes express distinct markers and occupy distinct locations within the TM tissue. The authors compared their expression data with other published scRNAseq studies of the mouse as well as the human aqueous outflow pathway. They used ATAC-seq to map open chromatin regions in order to predict transcription factor binding sites. Their results were also evaluated in the context of human IOP and glaucoma risk alleles from published GWAS data, with interesting and meaningful correlations. Although not discussed in their manuscript, their expression data support other signaling pathways/ proteins/ genes that have been implicated in glaucoma, including: TGFβ2, BMP signaling (including involvement of ID proteins), MYOC, actin cytoskeleton (CLANs), WNT signaling, etc.

In addition to these very impressive data, the authors used scRNAseq to examine changes in TM cell gene expression in the mouse glaucoma model of mutant Lmxb1-induced ocular hypertension. In man, LMX1B is associated with Nail-Patella syndrome, which can include the development of glaucoma, demonstrating the clinical relevance of this mouse model. Among the gene expression changes detected, TM3 cells had altered expression of genes associated with mitochondrial metabolism. The authors used their previous experience using nicotinamide to metabolically protect DBA2/J mice from glaucomatous damage, and they hypothesized that nicotinamide supplementation of mutant Lmx1b mice would help restore normal mitochondrial metabolism in the TM and prevent Lmx1b-mediated ocular hypertension. Adding nicotinamide to the drinking water significantly prevented Lmxb1 mutant mice from developing high intraocular pressure. This is a laudable example of dissecting the molecular pathogenic mechanisms responsible for a disease (glaucoma) and then discovering and testing a potential therapy that directly intervenes in the disease process and thereby protects from the disease.

Strengths:<br /> There are numerous strengths in this comprehensive study including:<br /> • Deep scRNA sequencing that was confirmed by an independent dataset in another mouse strain at another university.<br /> • Identification and validation of molecular markers for each mouse TM cell subset along with localization of these subsets within the mouse aqueous outflow pathway.<br /> • Rigorous bioinformatics analysis of these data as well as comparison of the current data with previously published mouse and human scRNAseq data.<br /> • Correlating their current data with GWAS glaucoma and IOP "hits".<br /> • Discovering gene expression changes in the 3 TM subgroups in the mouse mutant Lmx1b model of glaucoma.<br /> • Further pursuing the indication of dysfunctional mitochondrial metabolism in TM3 cells from Lmx1b mutant mice to test the efficacy of dietary supplementation with nicotinamide. The authors nicely demonstrate the disease modifying efficacy of nicotinamide in preventing IOP elevation in these Lmx1b mutant mice, preventing the development of glaucoma. These results have clinical implications for new glaucoma therapies.

Weaknesses:<br /> • Occasional over-interpretation of data. The authors have used changes in gene expression (RNAseq) to implicate functions and signaling pathways. For example: they have not directly measured "changes in metabolism", "mitochondrial dysfunction" or "activity of Lmx1b".<br /> • In their very thorough data set, there is enrichment of or changes in gene expression that support other pathways that have been previously reported to be associated with glaucoma (such as TGFβ2, BMP signaling, actin cytoskeletal organization (CLANs), WNT signaling, ossification, etc. that appears to be a lost opportunity to further enhance the significance of this work.

Reviewer #3 (Public review):

Summary:In this study, the authors perform multimodal single-cell transcriptomic and epigenomic profiling of 9,394 mouse TM cells, identifying three transcriptionally distinct TM subtypes with validated molecular signatures. TM1 cells are enriched for extracellular matrix genes, TM2 for secreted ligands supporting Schlemm's canal, and TM3 for contractile and mitochondrial/metabolic functions. The transcription factor LMX1B, previously linked to glaucoma, shows the highest expression in TM3 cells and appears to regulate mitochondrial pathways. In Lmx1bV265D mutant mice, TM3 cells exhibit transcriptional signs of mitochondrial dysfunction associated with elevated IOP. Notably, vitamin B3 treatment significantly mitigates IOP elevation, suggesting a potential therapeutic avenue.

This is an excellent and collaborative study involving investigators from two institutions, offering the most detailed single-cell transcriptomic and epigenetic profiling of the mouse limbal tissues-including both TM and Schlemm's canal (SC), from wild-type and Lmx1bV265D mutant mice. The study defines three TM subtypes and characterizes their distinct molecular signatures, associated pathways, and transcriptional regulators. The authors also compare their dataset with previously published murine and human studies, including those by Van Zyl et al., providing valuable cross-species insights.

Strengths:

(1) Comprehensive dataset with high single-cell resolution<br /> (2) Use of multiple bioinformatic and cross-comparative approaches<br /> (3) Integration of 3D imaging of TM and SC for anatomical context<br /> (4) Convincing identification and validation of three TM subtypes using molecular markers.

Weaknesses:

(1) Insufficient evidence linking mitochondrial dysfunction to TM3 cells in Lmx1bV265D mice: While the identification of TM3 cells as metabolically specialized and Lmx1b-enriched is compelling, the proposed link between Lmx1b mutation and mitochondrial dysfunction remains underdeveloped. It is unclear whether mitochondrial defects are a primary consequence of Lmx1b-mediated transcriptional dysregulation or a secondary response to elevated IOP. Additional evidence is needed to clarify whether Lmx1b directly regulates mitochondrial genes (e.g., via ChIP-seq, motif analysis, or ATAC-seq), or whether mitochondrial changes are downstream effects.<br /> Furthermore, the protective effects of nicotinamide (NAM) are interpreted as evidence of mitochondrial involvement, but no direct mitochondrial measurements (e.g., immunostaining, electron microscopy, OCR assays) are provided. It is essential to validate mitochondrial dysfunction in TM3 cells using in vivo functional assays to support the central conclusion of the paper. Without this, the claim that mitochondrial dysfunction drives IOP elevation in Lmx1bV265D mice remains speculative. Alternatively, authors should consider revising their claims that mitochondrial dysfunction in these mice is a central driver of TM dysfunction.

(2) Mechanism of NAM-mediated protection is unclear: The manuscript states that NAM treatment prevents IOP elevation in Lmx1bV265D mice via metabolic support, yet no data are shown to confirm that NAM specifically rescues mitochondrial function. Do NAM-treated TM3 cells show improved mitochondrial integrity? Are reactive oxygen species (ROS) reduced? Does NAM also protect RGCs from glaucomatous damage? Addressing these points would clarify whether the therapeutic effects of NAM are indeed mitochondrial.

(3) Lack of direct evidence that LMX1B regulates mitochondrial genes: While transcriptomic and motif accessibility analyses suggest that LMX1B is enriched in TM3 cells and may influence mitochondrial function, no mechanistic data are provided to demonstrate direct regulation of mitochondrial genes. Including ChIP-seq data, motif enrichment at mitochondrial gene loci, or perturbation studies (e.g., Lmx1b knockout or overexpression in TM3 cells) would greatly strengthen this central claim.

(4)Focus on LMX1B in Fig. 5F lacks broader context: Figure 5F shows that several transcription factors (TFs)-including Tcf21, Foxs1, Arid3b, Myc, Gli2, Patz1, Plag1, Npas2, Nr1h4, and Nfatc2-exhibit stronger positive correlations or motif accessibility changes than LMX1B. Yet the manuscript focuses almost exclusively on LMX1B. The rationale for this focus should be clarified, especially given LMX1B's relatively lower ranking in the correlation analysis. Were the functions of these other highly ranked TFs examined or considered in the context of TM biology or glaucoma? Discussing their potential roles would enhance the interpretation of the transcriptional regulatory landscape and demonstrate the broader relevance of the findings.

Other weaknesses:

(1) In abstract, they say a number of 9,394 wild-type TM cell transcriptomes. The number of Lmx1bV265D/+ TM cell transcriptomes analyzed is not provided. This information is essential for evaluating the comparative analysis and should be clearly stated in the Abstract and again in the main text (e.g., lines 121-123). Including both wild-type and mutant cell counts will help readers assess the balance and robustness of the dataset.

(2) Did the authors monitor mouse weight or other health parameters to assess potential systemic effects of treatment? It is known that the taste of compounds in drinking water can alter fluid or food intake, which may influence general health. Also, does Lmx1bV265D/+ have mice exhibit non-ocular phenotypes, and if so, does nicotinamide confer protection in those tissues as well? Additionally, starting the dose of the nicotinamide at postnatal day 2, how long the mice were treated with water containing nicotinamide, and after how many days or weeks IOP was reduced, and how long the decrease in the IOP was sustained.<br /> (3) While the IOP reduction observed in NAM-treated Lmx1bV265D/+ mice appears statistically significant, it is unclear whether this reflects meaningful biological protection. Several untreated mice exhibit very high IOP values, which may skew the analysis. The authors should report the mean values for IOP in both untreated and NAM-treated groups to clarify the magnitude and variability of the response.<br /> (4) Additionally, since NAM has been shown to protect RGCs in other glaucoma models directly, the authors should assess whether RGCs are preserved in NAM-treated Lmx1b V265D/+ mice. Demonstrating RGC protection would support a synergistic effect of NAM through both IOP reduction and direct neuroprotection, strengthening the translational relevance of the treatment.<br /> (5) Can the authors add any other functional validation studies to explore to understand the pathways enriched in all the subtypes of TM1, TM2, and TM3 cells, in addition to the ICH/IF/RNAscope validation?<br /> (6) The authors should include a representative image of the limbal dissection. While Figure S1 provides a schematic, mouse eyes are very small, and dissecting unfixed limbal tissue is technically challenging. It is also difficult to reconcile the claim that the majority of cells in the limbal region are TM and endothelium. As shown in Figure S6, DAPI staining suggests a much higher abundance of scleral cells compared to TM cells within the limbal strip. Additional clarification or visual evidence would help validate the dissection strategy and cellular composition of the captured region.

eLife Assessment

This is a valuable methodological contribution towards accurate characterization of viral genetic diversity using long-read sequencing and unique molecular identifiers (UMIs). However, the methods are currently incomplete and the sensitivity is not rigorously demonstrated. Addressing these gaps would strengthen the manuscript and make it a key addition to the field.

Reviewer #1 (Public review):

Tamao et al. aimed to quantify the diversity and mutation rate of the influenza (PR8 strain) in order to establish a high-resolution method for studying intra-host viral evolution. To achieve this, the authors combined RNA sequencing with single-molecule unique molecular identifiers (UMIs) to minimize errors introduced during technical processing. They proposed an in vitro infection model with a single viral particle to represent biological genetic diversity, alongside a control model using in vitro transcribed RNA for two viral genes, PB2 and HA.

Through this approach, the authors demonstrated that UMIs reduced technical errors by approximately tenfold. By analyzing four viral populations and comparing them to in vitro transcribed RNA controls, they estimated that ~98.1% of observed mutations originated from viral replication rather than technical artifacts. Their results further showed that most mutations were synonymous and introduced randomly. However, the distribution of mutations suggested selective pressures that favored certain variants. Additionally, comparison with a closely related influenza strain (A/Alaska/1935) revealed two positively selected mutations, though these were absent in the strain responsible for the most recent pandemic (CA01).

Overall, the study is well-designed, and the interpretations are strongly supported by the data. However, the following clarifications are recommended:

(1) The methods section is overly brief. Even if techniques are cited, more experimental details should be included. For example, since the study focuses heavily on methodology, details such as the number of PCR cycles in RT-PCR or the rationale for choosing HA and PB2 as representative in vitro transcripts should be provided.

(2) Information on library preparation and sequencing metrics should be included. For example, the total number of reads, any filtering steps, and quality score distributions/cutoff for the analyzed reads.

(3) In the Results section (line 115, "Quantification of error rate caused by RT"), the mutation rate attributed to viral replication is calculated. However, in line 138, it is unclear whether the reported value reflects PB2, HA, or both, and whether the comparison is based on the error rate of the same viral RNA or the mean of multiple values (as shown in Figure 3A). Please clarify whether this number applies universally to all influenza RNAs or provide the observed range.

(4) Since the T7 polymerase introduced errors are only applied to the in vitro transcription control, how were these accounted for when comparing mutation rates between transcribed RNA and cell-culture-derived virus?

(5) Figure 2 shows that a UMI group size of 4 has an error rate of zero, but this group size is not mentioned in the text. Please clarify.

Reviewer #2 (Public review):

Summary:

This manuscript presents a technically oriented application of UMI-based long-read sequencing to study intra-host diversity in influenza virus populations. The authors aim to minimize sequencing artifacts and improve the detection of rare variants, proposing that this approach may inform predictive models of viral evolution. While the methodology appears robust and successfully reduces sequencing error rates, key experimental and analytical details are missing, and the biological insight is modest. The study includes only four samples, with no independent biological replicates or controls, which limits the generalizability of the findings. Claims related to rare variant detection and evolutionary selection are not fully supported by the data presented.

Strengths:

The study addresses an important technical challenge in viral genomics by implementing a UMI-based long-read sequencing approach to reduce amplification and sequencing errors. The methodological focus is well presented, and the work contributes to improving the resolution of low-frequency variant detection in complex viral populations.

Weaknesses:

The application of UMI-based error correction to viral population sequencing has been established in previous studies (e.g., in HIV), and this manuscript does not introduce a substantial methodological or conceptual advance beyond its use in the context of influenza.

The study lacks independent biological replicates or additional viral systems that would strengthen the generalizability of the conclusions. Potential sources of technical error are not explored or explicitly controlled. Key methodological details are missing, including the number of PCR cycles, the input number of molecules, and UMI family size distributions. These are essential to support the claimed sensitivity of the method.

The assertion that variants at {greater than or equal to}0.1% frequency can be reliably detected is based on total read count rather than the number of unique input molecules. Without information on UMI diversity and family sizes, the detection limit cannot be reliably assessed.

Although genetic variation is described, the functional relevance of observed mutations in HA and NA is not addressed or discussed in the context of known antigenic or evolutionary features of influenza. The manuscript is largely focused on technical performance, with limited exploration of the biological implications or mechanistic insights into influenza virus evolution.

The experimental scale is small, with only four viral populations derived from single particles analyzed. This limited sample size restricts the ability to draw broader conclusions about quasispecies dynamics or evolutionary pressures.

Author response:

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

Reviewer #1 (Public Review):

The authors sought to test whether anterior insular cortex neurons increase or decrease firing during fear behavior and freezing, bi-directionally control fear via separate, anatomically defined outputs. Using a fairly simple behavior where mice were exposed to tone-shock pairings, they found roughly equal populations that do indeed either increase or decrease firing during freezing. Next, they sought to test whether these distinct populations may also have distinct outputs. Using retrograde tracers they found that the anterior insular cortex contains non-overlapping neurons which project to the mediodorsal thalamus or amygdala. Mediodorsal thalamus-projecting neurons tended to cluster in deep cortical layers while amygdala-projecting neurons were primarily in more superficial layers. Stimulation of insula-thalamus projection decreased freezing behavior, and stimulation of insula-amygdala projections increased fear behavior. Given that the neurons that increased firing were located in deep layers, that thalamus projections occurred in deep layers, and that stimulation of insula-thalamus neurons decreased freezing, the authors concluded that the increased firing neurons may be thalamus projections. Similarly, given that decreased-firing neurons tended to occur in more superficial layers, that insula-amygdala projections were primarily superficial, and that insula-amygdala stimulation increased freezing behavior, authors concluded that the decreased firing cells may be amygdala projections. The study has several strengths though also some caveats.

Strengths:

The potential link between physiological activity, anatomy, and behavior is well laid out and is an interesting question. The activity contrast between the units that increase/decrease firing during freezing is clear.

It is nice to see the recording of extracellular spiking activity, which provides a clear measure of neural output, whereas similar studies often use bulk calcium imaging, a signal that rarely matches real neural activity even when anatomy suggests it might (see London et al 2018 J Neuro - there are increased/decreased spiking striatal populations, but both D1 and D2 striatal neurons increase bulk calcium).

Weaknesses:

The link between spiking, anatomy, and behavior requires assumptions/inferences: the anatomically/genetically defined neurons which had distinct outputs and opposite behavioral effects can only be assumed the increased/decreased spiking neurons, based on the rough area of the cortical layer they were recorded.

Yes, we are aware that we could not provide a direct link between spiking, anatomy and behavior. We have specifically noted this in the discussion section and added a possible experiment that could be carried out to provide a more direct link in a future study.

[Lines 371-375] We would like to provide a more direct evidence between the neuronal response types and projection patterns in future studies by electrophysiologically identifying freezing-excited and freezing-inhibited aIC neurons and testing whether those neurons activates to optogenetic activation of amygdala or medial thalamus projecting aIC neurons.

The behavior would require more control to fully support claims about the associative nature of the fear response (see Trott et al 2022 eLife) - freezing, in this case, could just as well be nonassociative. In a similar vein, fixed intertrial intervals, though common practice in the fear literature, pose a problem for neurophysiological studies. The first is that animals learn the timing of events, and the second is that neural activity is dynamic and changes over time. Thus it is very difficult to determine whether changes in neural activity are due to learning about the tone-shock contingency, timing of the task, simply occur because of time and independently of external events, or some combination of the above.

Trott et al. (2022) stated that "...freezing was the purest reflection of associative learning." The nonassociative processes mentioned in the study were related to running and darting behaviors, which the authors argue are suppressed by associative learning. Moreover, considerable evidence from immediate postshock freezing and immediate postshock context shift studies all indicate that the freezing response is an associative (and not nonassociative) response (Fanselow, 1980 and 1986; and Landeira-Fernandez et al., 2006). Thus, our animals' freezing response to the tone CS presentation in a novel context, following three tone CS-footshock US pairings, most likely reflects associative learning. 

Concerning the issue of fixed inter-trial intervals (ITIs), which are standard in fear conditioning studies, particularly those with few CS-US paired trials, we acknowledge the challenge in interpreting the neural correlates of behavior. However, the ITIs in our extinction study was variable and we still found neural activities that had significant correlation with freezing. The results of our extinction study, carried out with variable it is, suggest that the aIC neural activity changes measured in this study is likely due to freezing behavior associated with fear learning, not due to learning the contingencies of fixed ITIs.

Reviewer #2 (Public Review):

In this study, the authors aim to understand how neurons in the anterior insular cortex (insula) modulate fear behaviors. They report that the activity of a subpopulation of insula neurons is positively correlated with freezing behaviors, while the activity of another subpopulation of neurons is negatively correlated to the same freezing episodes. They then used optogenetics and showed that activation of anterior insula excitatory neurons during tones predicting a footshock increases the amount of freezing outside the tone presentation, while optogenetic inhibition had no effect. Finally, they found that two neuronal projections of the anterior insula, one to the amygdala and another to the medial thalamus, are increasing and decreasing freezing behaviors respectively. While the study contains interesting and timely findings for our understanding of the mechanisms underlying fear, some points remain to be addressed.

We are thankful for the detailed and constructive comments by the reviewer and addressed the points. Specifically, we included possible limitations of using only male mice in the study, included two more studies about the insula as references, specified the L-ratio and isolated distance used in our study, added the ratio of putative-excitatory and putative-inhibitory neurons obtained from our study, changed the terms used to describe neuronal activity changes (freezing-excited and freezing-inhibited cells), added new analysis (Figure 2H), rearranged Figure 2 for clarity, added new histology images, and added atlas maps with viral expressions (three figure supplements).

Reviewer #1 (Recommendations For The Authors):

- I would suggest keeping the same y-axis for all figures that display the same data type - Figure 5D, for example.

Thank you for the detailed suggestion. We corrected the y-axis that display the same data type to be the same for all figures.

- In the methods, it says 30s bins were used for neural analysis (line 435). I cannot imagine doing this, and looking at the other figures, it does not look like this is the case so could you please clarify what bins, averages, etc were used for neural and behavioral analysis?

Bin size for neural analysis varied; 30s, 5s, 1s bins were used depending on the analysis. We corrected this and specified what time bin was used for which figure in the methods.

Bin size for neural and freezing behavior was 30s and we also added this to the methods.

- I would not make any claims about the fear response here being associative/conditional. This would require a control group that received an equal number of tone and shock exposures, whether explicitly unpaired or random.

The unpaired fear conditioning paradigm, unpaired tone and shock, suggested by the reviewer is well characterized not to induce fear behavior by CS (Moita et al., 2003 and Kochli et al., 2015). In addition, considerable evidence from immediate post-shock freezing and immediate post-shock context shift studies all indicate that the freezing response is an associative (and not nonassociative) response (Fanselow, 1980 and 1986; and Landeira-Fernandez et al., 2006). Thus, our animals' freezing response to the tone CS presentation in a novel context, following three tone CS-footshock US pairings, most likely reflects associative learning.

- I appreciate the discussion about requiring some inference to conclude that anatomically defined neurons are the physiologically defined ones. This is a caveat that is fully disclosed, however, I might suggest adding to the discussion that future experiments could address this by tagging insula-thalamus or insula-amygdala neurons with antidromic (opto or even plain old electric!) stimulation. These experiments are tricky to perform, of course, but this would be required to fully close all the links between behavior, physiology, and anatomy.

As suggested, we have included that, in a future study, we would like to elucidate a more direct link between physiology, anatomy and behaviors by optogenetically tagging the insula-thalamus/insula-amygdala neurons and identifying whether it may be a positive or a negative cell (now named the freezing-excited and freezing-inhibited cells, respectively) in the discussion.

[Lines 371-375] We would like to provide a more direct evidence between the neuronal response types and projection patterns in future studies by electrophysiologically identifying freezing-excited and freezing-inhibited aIC neurons and testing whether those neurons activates to optogenetic activation of amygdala or medial thalamus projecting aIC neurons.

Reviewer #2 (Recommendations For The Authors):

Major comments:

(1) As all experiments have been performed only in male mice, the authors need to clearly state this limit in the introduction, abstract, and title of the manuscript.

With increasing number of readers becoming interested in the biological sex used in preclinical studies, we also feel that it should be mentioned in the beginning of the manuscript. As suggested, we explicitly wrote that we only used male mice in the title, abstract, and introduction. In addition, we discussed possible limitations of only using male mice in the discussion section as follows:

[Lines 381-386] Another factor to consider is that we have only used male mice in this study. Although many studies report that there is no biological sex difference in cued fear conditioning (42), the main experimental paradigm used in this study, it does not mean that the underlying brain circuit mechanism would also be similar. The bidirectional fear modulation by aIC→medial thalamus or the aIC→amygdala projections may be different in female mice, as some studies report reduced cued fear extinction in females (42).

(2) The authors are missing important publications reporting findings on the insular cortex in fear and anxiety. For example, the authors should cite studies showing that anterior insula VIP+ interneurons inhibition reduces fear memory retrieval (Ramos-Prats et al., 2022) and that posterior insula neurons are a state-dependent regulator of fear (Klein et al., 2021). Also, regarding the anterior insula to basolateral amygdala projection (aIC-BLA), the author should include recent work showing that this population encodes both negative valence and anxiogenic spaces (Nicolas et al., 2023). 

We appreciate the detailed suggestions and we added appropriate publications in the discussion section. The anterior insula VIP+ interneuron study (Ramos-Prats et al., 2022) is interesting, but based on the evidence provided in the paper, we felt that the role of aIC VIP+ interneuron in fear conditioning is low. VIP+ interneurons in the aIC seem to be important in coding sensory stimuli, however, it’s relevance to conditioned stimuli seems to be low; overall VIP intracellular calcium activity to CS was low and did not differ between acquisition and retrieval. Also, inhibition of VIP did not influence fear acquisition. VIP inhibition during fear acquisition did reduce fear retrieval (CS only, no light stimulation), but this does not necessarily mean that VIP activity will be involved in fear memory storage or retrieval, especially because intracellular calcium activity of VIP+ neurons was low during fear conditioning and retrieval.

Studies by Klein et al. (2021) and Nicolas et al. (2023) are integrated in the discussion section as follows.

[Lines 297-301] Group activity of neurons in the pIC measured with fiberphotometry, interestingly, exhibited fear state dependent activity changes—decreased activity with high fear behavior and increased activity with lower fear behavior (29)—suggesting that group activity of the pIC may be involves in maintain appropriate level of fear behavior.

[Lines 316-319] Another distinction between the aIC and pIC may be related with anxiety, as a recent study showed that group activity of aIC neurons, but not that of the pIC, increased when mice explored anxiogenic space (open arms in an elevated plus maze, center of an open field box) (32).

(3) The authors should specify how many neurons they excluded after controlling the L-ratio and isolation distance. It is also important to specify the percentage of putative excitatory and inhibitory interneurons recorded among the 11 mice based on their classification (the number of putative inhibitory interneurons in Figure 1D seems too low to be accurate).

We use manual cluster cutting and only cut clusters that are visually well isolated. So we hardly have any neurons that are excluded after controlling for L-ratio and isolation distance. The criterion we used was L-ratio<0.3 and isolation distance>15, and we specified this in the methods as follows.

[Lines 454-458] We only used well-isolated units (L-ratio<0.3, isolation distance>15) that were confirmed to be recorded in the aIC (conditioned group: n = 116 neurons, 11 mice; control group: n = 14 neurons, 3 mice) for the analysis (46). The mean of units used in our analysis are as follows: L-ratio = 0.09 ± 0.012, isolation distance = 44.97 ± 5.26 (expressed as mean ± standard deviation).

As suggested, we also specified the percentage of putative excitatory and inhibitory interneurons recorded from our study in the results and methods section. The relative percentage of putative excitatory and inhibitory interneurons were similar for both the conditioned and the control groups (conditioned putative-excitatory: 93.1%, putative-inhibitory: 6.9%; control putative-excitatory: 92.9%, putative-inhibitory: 7.1%). Although the number of putative-interneurons isolated from our recordings is low that is what we obtained. Putative inhibitory neurons, probably because of their relatively smaller size, has a tendency to be underrepresented than the putative excitatory cells.

[Lines 83-87] Of the recorded neurons, we analyzed the activity of 108 putative pyramidal neurons (93% of total isolated neurons) from 11 mice, which were distinguished from putative interneurons (n = 8 cells, 7% of total isolated neurons) based on the characteristics of their recorded action potentials (Figure 1D; see methods for details).

[Lines 464-467] The percentage of putative excitatory neurons and putative inhibitory interneurons obtained from both groups were similar (conditioned putative-excitatory: 93.1%, putative-inhibitory: 6.9%; control putative-excitatory: 92.9%, putative-inhibitory: 7.1%).

(4) While the use of correlation of single-unit firing frequency with freezing is interesting, classically, studies analyze the firing in comparison to the auditory cues. If the authors want to keep the correlation analysis with freezing, rather than correlations to the cues, they should rename the cells as "freezing excited" and "freezing inhibited" cells instead of positive and negative cells.

As suggested, we used the terms “freezing-excited” and “freezing-inhibited” cells instead of positive and negative cells.

(5) To improve clarity, Figure 2 should be reorganized to start with the representative examples before including the average of population data. Thus Panel D should be the first one. The authors should also consider including the trace of the firing rate of these representative units over time, on top of the freezing trace, as well as Pearson's r and p values for both of them. Then, the next panels should be ordered as follows: F, G, H, C, A, B, I, and finally E.

We have rearranged Figure 2 based on the suggestions.

(6) It is unclear why the freezing response in Figure 2 is different in current panels F, G, and H. Please clarify this point.

It was because the freezing behaviors of slightly different population of animals were averaged. Some animals did not have positive/negative (or both) cells and only the behavior of animals with the specified cell-type were used for calculating the mean freezing response. With rearrangement of Figure 2, now we do not have plots with juxtaposed mean neuronal response-types and behavior.

(7) Even though the peak of tone-induced firing rate change between negative and positive cells is 10s later for positive cells, the conclusion that this 'difference suggests differential circuits may regulate the activities of different neuron types in response to fear' is overstating the observation. This statement should be rephrased. Indeed, it could be the same circuits that are regulated by different inputs (glutamatergic, GABA, or neuromodulatory inputs).

We agree and delete the statement from the manuscript.

(8) The authors mention they did not find tone onset nor tone offset-induced responses of anterior insula neurons. It would be helpful to represent this finding in a Figure, especially, which were the criteria for a cell to be tone onset or tone offset responding.

We added how tone-onset and tone-offset were analyzed in the methods section and added a plot of the analysis in Figure 2H.

(9) Based on the spread of the viral expression shown in Figure 3B, it appears that the authors are activating/inhibiting insula neurons in the GI layer, whereas single-unit recordings report the electrodes were located in DI, AID, and AIV layers. The authors should provide histology maps of the viral spread for ChR2, NpHR3, and eYFP expression.

Thank you for the excellent suggestion. Now the histological sample in Figure 3B is a sample with expression in the GI/DI/AID layers and it also has an image taken at higher resolution (x40) to show that viral vectors are expressed inside neurons. We also added histological maps with overlay of viral expression patterns of the ChR2, eYFP, and NpHR3 groups in Figure 3—figure supplement 1.

(10) In Figure 5B, the distribution of terminals expressing ChR2 appears much denser in CM than in MD. This should be quantified across mice and if consistent with the representative image, the authors should refer to aIC-CM rather than aIC-MD terminals.

Overall, we referred to the connection as aIC-medial thalamus, which collectively includes both the CM and the MD. Microscopes we have cannot determine whether terminals end at the CM or MD, but the aIC projections seems to pass through the CM to reach the MD. The Allen Brain Institute’s Mouse brain connectivity map (https://connectivity.brain-map.org/projection/experiment/272737914) of a B6 mouse, the mouse strain we used in our study, with tracers injected in similar location as our study also supports our speculation and shows that aIC neuronal projections terminate more in the MD than in the CM. In addition, the power of light delivered for optogenetic manipulation is greatly reduced over distance, and therefore, the MD projecting terminals which is closer to the optic fiber will be more likely to be activated than the CM projecting terminals. However, since we could not determine whether the aIC terminate at the CM or the MD, we collectively referred to the connection as the aIC-medial thalamus throughout the manuscript.

Author response image 1.

(11) Histological verifications for each in vivo electrophysiology, optogenetic, and tracing experiments need to include a representative image of the implantation/injection site, as well as a 40x zoom-in image focusing on the cell bodies or terminals right below the optic fiber (for optogenetic experiments). Moreover, an atlas map including all injection locations with the spread of the virus and fiber placement should be added in the Supplement Figures for each experiment (see Figure S1 Klein et al., 2021). Similarly, the authors need to add a representation of the spread of the retrograde tracers for each mouse used for this tracing experiment.

As suggested, we added a histology sample showing electrode recording location for in-vivo electrophysiology in Figure 1 and added atlas maps for the optogenetic and tracing experiments in supplementary figures. We also provide a 40x zoom-in image of the expression pattern for the optogenetic experiments (Figure 3B).

(12) To target anterior insula neurons, authors mention coordinates that do not reach the insula on the Paxinos atlas (AP: +1.2 mm, ML: -3.4 mm, DV: -1.8 mm). If the DV was taken from the brain surface, this has to be specified, and if the other coordinates are from Bregma, this also needs to be specified. Finally, the authors cite a review from Maren & Fanselow (1996), for the anterior insula coordinates, but it remains unclear why.

AP and ML coordinates are measurement made in reference to the bregma. DV was calculated from the brain surface. We specified these in the Methods. We did not cite a review from Maren & Fenselow for the aIC coordinates.

Minor comments:

(1) A schematic of the microdrive and tetrodes, including the distance of each tetrode would also be helpful.

We used a handcrafted Microdrives with four tetrodes. Since they were handcrafted, the relative orientation of the tetrodes varies and tetrode recording locations has to be verified histologically. We, however, made sure that the distance between tetrodes to be more than 200 μm apart so that distinct single-units will be obtained from different tetrodes. We added this to the methods as follows.

[Lines 430-431] The distance between the tetrodes were greater than 200 μm to ensure that distinct single-units will be obtained from different tetrodes.

(2) Figure 2E: representation of the baseline firing (3-min period before the tone presentation) is missing.

Figure 2E is the 3 min period before tone presentation

(3) Figure 2: Averages Pearson's correlation r and p values should be stated on panels F, G, and H (positive cell r = 0.81, P < 0.05; negative cell r = -0.68, P < 0.05).

They were all originally stated in the figures. But with reorganization of Figure 2, we now have a plot of the Pearson’s Correlation with r and p values in Figure 2F.

(4) Figure 2I: Representation of the absolute value of the normalized firing is highly confusing. Indeed, as the 'negative cells' are inhibited to freezing, firing should be represented as normalized, and negative for the inhibited cells.

To avoid confusion, we did not take an absolute value of the “negative cells”, which are now called the “freezing-inhibited cells”.

(5) Figure 4E (retrograde tracing): representation of individual values is missing.

Figure 4E now has individual values.

References:

London, T. D., Licholai, J. A., Szczot, I., Ali, M. A., LeBlanc, K. H., Fobbs, W. C., & Kravitz, A. V. (2018). Coordinated ramping of dorsal striatal pathways preceding food approach and consumption. Journal of Neuroscience, 38(14), 3547-3558.

Trott, J. M., Hoffman, A. N., Zhuravka, I., & Fanselow, M. S. (2022). Conditional and unconditional components of aversively motivated freezing, flight and darting in mice. Elife, 11, e75663.

Fanselow, M. S. (1980). Conditional and unconditional components of post-shock freezing. The Pavlovian journal of biological science: Official Journal of the Pavlovian, 15(4), 177-182.

Fanselow, M. S. (1986). Associative vs topographical accounts of the immediate shock-freezing deficit in rats: implications for the response selection rules governing species-specific defensive reactions. Learning and Motivation, 17(1), 16-39.

Landeira-Fernandez, J., DeCola, J. P., Kim, J. J., & Fanselow, M. S. (2006). Immediate shock deficit in fear conditioning: effects of shock manipulations. Behavioral neuroscience, 120(4), 873.

Moita, M. A., Rosis, S., Zhou, Y., LeDoux, J. E., & Blair, H. T. (2003). Hippocampal place cells acquire location-specific responses to the conditioned stimulus during auditory fear conditioning. Neuron, 37(3), 485-497.

Kochli, D. E., Thompson, E. C., Fricke, E. A., Postle, A. F., & Quinn, J. J. (2015). The amygdala is critical for trace, delay, and contextual fear conditioning. Learning & memory, 22(2), 92-100.

Ramos-Prats, A., Paradiso, E., Castaldi, F., Sadeghi, M., Mir, M. Y., Hörtnagl, H., ... & Ferraguti, F. (2022). VIP-expressing interneurons in the anterior insular cortex contribute to sensory processing to regulate adaptive behavior. Cell Reports, 39(9).

Klein, A. S., Dolensek, N., Weiand, C., & Gogolla, N. (2021). Fear balance is maintained by bodily feedback to the insular cortex in mice. Science, 374(6570), 1010-1015.

Nicolas, C., Ju, A., Wu, Y., Eldirdiri, H., Delcasso, S., Couderc, Y., ... & Beyeler, A. (2023). Linking emotional valence and anxiety in a mouse insula-amygdala circuit. Nature Communications, 14(1), 5073.

Maren, S., & Fanselow, M. S. (1996). The amygdala and fear conditioning : Has the nut been cracked? Neuron, 16(2), 237‑240. https://doi.org/10.1016/s0896-6273(00)80041-0

Author response:

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

Reviewer #1 (Public review):

(1) The primary weakness of the paper concerns its conclusion of having generated "homogenous mature microglia", partly based on the RNAseq analysis. However, the comparison of gene profiles was carried out only between "hiPSC-derived mature microglia" and the proliferating myeloid progenitors. While the transcriptome profiles revealed a trend of enrichment of microglia-like gene expression in "hiPSC-derived mature microglia" compared to proliferating myeloid progenitors, this is not sufficient to claim they are "mature microglia". It is important that one carries out a comparative analysis of the RNAseq data with those of primary human microglia, which may be done by leveraging the public database. To convincingly claim these cells are mature microglia, questions need to be addressed including how similar the molecular signatures of these cells are compared with the fully differentiated primary microglia cell or if they remain progenitor-like or take on mosaic properties, and how they distinguish from macrophages.

We greatly appreciate the insightful comments and suggestions from the reviewers, which were instrumental in enhancing our data analysis and organization. In response to the feedback, we have updated the terminology from “mature microglia” to simply “microglia” while clarifying in our text that these are fully differentiated microglia under single-type cell culture conditions.

Guided by the reviewer's advice, we incorporated RNA-seq data from human brain microglia studies conducted by Dr. Poon and Dr. Blurton-Jones' Lab (Abud et al., Neuron, 2017) and Dr. Huitinga's Lab (van der Poel et al., Nat Commun, 2019). We then conducted a comparative analysis of the gene expression profiles between our fully differentiated hiPSC-derived microglia and those from fetal/adult brain microglia (see Fig.2. Suppl. B, C and D; Suppl. table 1 and table 2). The correlation analysis revealed that our hiPSC-derived microglia closely resemble fetal and adult brain microglia, distinguishing them significantly from monocytes and inflammatory monocytes.

(2) While the authors attempted to demonstrate the functional property of "hiPSC-derived mature microglia" in culture, they used LPS challenge, which is an inappropriate assay. This is because human microglia respond poorly to LPS alone but need to be activated by a combination of LPS with other factors, such as IFNγ. Their data that "hiPSC-derived mature microglia" showed robust responses to LPS indeed implicates that these cells do not behave like mature human microglia.

We appreciate the feedback received. In response, we cultured hiPSC-derived microglia cells and subjected them to treatments with IFNγ, LPS, and a combination of both IFNγ+LPS, as illustrated in Figure 3 suppl. Our findings revealed that the IFNγ+LPS combination notably enhanced the expression of IL1a, IL1b, TNFa, CCL8, and CXCL10, whereas IL6 and CCL2 levels remained unchanged. Treatment with IFNγ alone significantly elevated the expression of TNFa, CCL8, CXCL10, and CCL2. These outcomes align with the findings reported by Rustenhoven et al. (Sci Rep, 2016), suggesting that the functionality of our hiPSC-derived microglia cells closely mirrors that of primary human adult microglia cells.

(3) The resolution of Figs. 4 - 6 is so low that even some of the text and labels are hardly readable. Based on the morphology shown in Fig. 4 and the statement in line 147, these hiPSC-derived "cells altered their morphology to a rounded shape within an hour of incubation and rapidly internalized the fluorescent-labeled particles". This is a peculiar response. Usually, microglia do not respond to fluorescent-labeled zymosan by turning into a rounded shaped within an hour when they internalize them. Such a behavior usually implicates weak phagocytotic capacity.

Thank you for your insightful comments. During submission, the main text's PDF version was converted online, resulting in low-quality output. We have since updated this with a high-resolution version. The observed alterations in cell morphology following zymosan phagocytosis may be attributed to the high zymosan concentration used (2mg/ml). We conducted an assessment to understand the impact of zymosan concentration on the morphology of hiPSC-derived microglial cells, as shown in Figure 4 suppl B. Our findings indicate that microglia cells adopt an amoeboid, rounded shape at zymosan concentrations exceeding 20ug/ml. To clarify this point, we have amended the text to read: "The cells altered their morphology and rapidly internalized the fluorescent-labeled particles."

(4) Data presented in Fig. 5 are not very convincing to support that transplanted cells were immunopositive for "human CD11b (Fig.5C), as well as microglia signature markers P2ry12 and TMEM119 (Fig.5D)" (line 167). The resolution and magnification of Fig. 5D is too low to tell the colocalization of tdT and human microglial marker immunolabeling. In the flat-mount images (C, I), hCD11b immunolabeling is not visible in the GCL or barely visible in the IPL. This should be discussed.

We are grateful for the reviewer's comments. As previously mentioned, the low quality of the images was due to the online conversion of the PDF version. We have now submitted both high-quality PDF and Word versions for the reviewer's assessment. In these high-quality versions, the colocalization of tdT with human P2ry12 and TMEM119 is distinctly visible. Additionally, we have updated the hTMEM119 staining images in Figure 5D. The results from hCD11b staining align with those observed in mouse CD11b staining, notably showing more effective staining in the outer plexiform layer (OPL) microglia cells. The reason for this—whether it pertains to a staining issue, a variance in CD11b expression among microglia cells in the OPL and ganglion layer (GL), or differences in the samples due to varying conditions—is not yet clear and warrants further investigation.

(5) Microglia respond to injury by becoming active and lose their expression of the resting state microglial marker, such as P2ry12, which is used in Fig. 6 for detection of migrated microglia. To confirm that these cells indeed respond to injury like native microglia, one should check for activated microglial markers and induction of pro-inflammatory cytokines in the sodium iodate-injury model.

The reviewer's insights are spot-on. We utilized preserved retinas to extract mRNA, which was then reverse-transcribed to cDNA for conducting qRT-PCR using human-specific primers, as detailed in the updated Table 5. The findings revealed that following retinal pigment epithelium (RPE) injury for 3 days, the transplanted hiPSC-derived microglial cells exhibited an increase in the production of inflammatory cytokines and upregulated genes related to phagocytosis, migration, and adhesion. Conversely, there was a decrease in the expression of microglia-specific signature genes and neurotrophic factors, as demonstrated in Figure 7 suppl.

Reviewer #1 (Recommendations For The Authors):

Line 52: "Microglia cell repopulation research suggests that: 1) if no injury or infection occurs, retinal microglia cells can sustain their homeostasis indefinitely" - this statement is too strong or delivers a confusing message; it needs clarification or to be backed up by evidence. Recent single cell RNA sequencing analyses suggest that even under a normal condition, residential microglia do not present as a single homeostatic cell cluster, rather a subpopulation of activated inflammatory microglia are constantly detectable in the normal retina. This is likely because normal retinal neurons can be stressed due to various reasons, such as the temporal accumulation of misfolded proteins, exposed to strong light, or ageing, etc.

We appreciate the comments. We changed the sentence to read, "Microglia cell repopulation research suggests that: 1) retinal resident microglia cells can sustain their population with the local dividing and migration if any perturbations do not exceed the threshold of the recovery speed by local neighbor microglia cells."

Line 83: "we applied an appropriate protocol for culturing human iPSC-derived microglia cells" - it would be more appropriate if the word "appropriate" can be replaced by either "unique" or a phrase like "we adopted a (previously published) protocol...".

Thanks! We changed it to “We modified a previously published protocol to culture human iPSC-derived microglia cells.".

Fig. 1F,G: A method of flow cytometry will provide more comprehensive cell quantification for percentages of positively labeled cells than cell counts under high magnification confocal images.

Thanks for the comments! We agreed with the reviewer. Given the experimental resources available, the quantifications of confocal images did provide a reasonable assessment. We will perform flow cytometry analysis in future experiments.

Reviewer #2 (Public review):

Weaknesses:

Gene expression analysis of mature microglia cells should be better interpreted and it would be beneficial to compare the iPSC-derived microglia gene set to a human microglial cell line (for example, HMC3) instead of myeloid progenitor cells.<br /> The way that the manuscript has been written, unfortunately, is not optimal. I recommend that the entire manuscript be edited and proofread in English. The text contains spelling and grammar mistakes, and the manuscript is inconsistent in several parts. The manuscript should also be revised for a scientific paper format.

We appreciate the reviewer's comments and have taken them into consideration along with similar inquiries from Reviewer 1. Following the suggestions, we conducted a comparison of gene expression profiles between our hiPSC-derived microglia and those from fetal/adult brain microglia, as depicted in the updated Fig.2. Suppl. B, C and D; as well as in the Suppl. table 1 and table 2. The correlation analysis demonstrated that the hiPSC-derived microglia cells closely resemble fetal and adult brain microglia, significantly differing from monocytes and inflammatory monocytes. Additionally, we have revised the manuscript to adhere more closely to the conventional scientific format.

Reviewer #2 (Recommendations For The Authors):

Specific suggestions for improvement:

- Regarding the characterization of human iPSC-derived microglia, P2RY12 is a general hematopoietic cell marker. One cannot judge the maturity of microglia only by P2RY12 expression (for example, line 261). The expression of more specific markers such as TMEM119 and PROS1 should be studied and discussed.

We are thankful for the reviewer's valuable feedback. In response:

We have removed the term "mature" and clarified that the hiPSC-derived microglia we studied are fully differentiated within single-type cell culture conditions.

We performed a comparative analysis of the gene expression profiles between our hiPSC-derived microglia and microglia from human brains, as illustrated in the updated Fig.2. Suppl. B, C and D. The results affirm that hiPSC-derived microglia closely resemble human fetal and adult microglia.

We noted that the expression of TMEM119 in hiPSC-derived microglia under in vitro single-type cell culture conditions is notably low, as shown in the below A. This suggests that the stimulatory factors in our single-type cell culture might not sufficiently induce TMEM119 expression in microglia. The necessity for a retinal environment or interaction with neuronal and/or other glial cells for TMEM119 expression mirrors the behavior of infiltrating peripheral monocytes in pathological conditions, which initially lack TMEM119 but later differentiate into microglial-like macrophages that express TMEM119, as reported by Ma et al. in Sci Rep (2017).

Additionally, our findings suggest that PROS1 is not uniquely characteristic of microglia but is expressed across a variety of cell types. Within our specific culture conditions, we noted a higher expression of PROS1 in microglial progenitor cells, as shown in Author response image 1B and C.

Author response image 1.

- In Figure 2, Part E, the names of the genes or pathways in the figure are not clear, and are these genes the set that are the most differentially expressed between iPSCs-derived microglia and MPC? The analysis needs more explanation.

We regret any confusion caused by our previous explanation. To clarify, we compiled a list of microglia-enriched genes from the research conducted by Barres BA Lab (Bennett et al., Proc Natl Acad Sci U S A, 2016) and from our own RNA sequencing data of mouse retinal microglia, identifying a total of 130 genes predominantly expressed in microglia (Suppl. Table 3). We then applied this gene list to analyze our hiPSC-derived microglia RNA sequencing data, resulting in the identification of 71 microglia-specific genes. These 71 genes were subjected to Ingenuity Pathway Analysis (IPA) to visualize the signaling pathways involved. The details of these microglia genes can be found in the updated suppl. table 3.

- Lines 124 to 128 mention that high expression of Stat3, IL1b, and IL6 and their central role in pathway analysis emphasize the efficiency of the maturation protocol. Regarding the fact that Stat3, IL1b, and IL6 are contributors to proinflammatory pathways, it is not convincing that the high expression of these genes in iPSC-derived microglia demonstrates the efficiency of the maturation protocol, given that microglia are not stimulated.

Thanks for the comments! We added the sentences about the comparison results between hiPSC-derived microglia and human brain microglia. We have also replaced the “mature” with “functional.” The sentence reads, “Thus, our method of obtaining differentiated microglia is a reliable method to generate a large number of homogenous functional microglia cells.”

- Statistical analysis is missing for some graphs, for example, figures 1-3 and 5.

We appreciate the comments. We have added the statistical results in the revised version.

- The legend for Figure 3 needs to be rewritten. The graphs or applied assays should be explained in the legend, not the interpretation of the data.

The legend was rewritten.

- There is no Figure 3 in the supplement figures file.

We added Figure 3. Suppl.

- hTMEM119 staining in Figure 5, Part D, is mostly background. Please provide another image.

The images were unclear after on-line converting due to the low number of pixels. We replaced them with new hTMEM119 staining images in Figure 5D.

- In line 176, figure 5I has been forgotten to be mentioned.

Thank you very much! We added 5I.

- Lines 241 to 244 state that more than 50% of the AMD-associated genes are highly expressed in retinal microglia according to Fig. discussion suppl A & B. It is not clear that the gene set that was used for analysis is from a healthy retinal microglia or AMD-related ones. Please explain precisely.

Thank you for your feedback. The gene list we referenced originates from a Genome-Wide Association Study (GWAS) that compared patients with Age-related Macular Degeneration (AMD) to healthy cohorts. We did not directly utilize this list in our experiments but referred to it to underscore the importance of microglia cells in the context of AMD.

Some of the English proofreading and manuscript format comments:

Line 805: Iba1 is written in lowercase. Is it human IBA1? It is not consistent with the way it is written in the text (in line 117, for example).

Thank you for pointing out the error. We reformed all Iba1 as “Iba1”. The Iba1 we used here are all from Wako (#019–19741), which labels both mouse and human microglial cells.

Line 814: microglia-enriched gene expression instead of microglia-enrich gene expression

Thank you! We changed it.

Line 345: Starting a sentence with lower case letter.

Thank you! We changed it.

Line 342: Myeloid lineage instead of myeloid cell linage.

Thank you! We changed it.

Line 815: What does FPKM stand for? The abbreviations should be explained.

The FPKM is the abbreviation of Fragments Per Kilobase of transcript per Million mapped reads. We added it in the text.

Line 309: The manuscript has occasionally referred to PLX-5622 without a minus. Please follow a uniform format.

We changed all “PLX5622” to “PLX-5622”.

Lines 327-331: should be rewritten.

The mentioned paragraph was rewritten.

Lines 335-340: should be rewritten.

The mentioned sentence was rewritten.

Line 135: qRT-PCR instead of QPCR," as it is also mentioned in the methods and material. The correction also applies to all the QPCRs in the text.

We changed “QPCR” with “qRT-PCR”

Figure 3: Graph B should be right side of graph A

Images description: It is better to have the images description in the left side of the image, for example, figure 5 part B, GL, IPL and OPL

Thanks for the suggestion. We changed the image organization as per the reviewer’s advice.

Lines 258 to 260 in the discussion have also been repeated with the same words in the introduction.

The mentioned paragraph was rewritten.

Lines 327-331 should be rewritten.

The mentioned paragraph was rewritten.

Lines 335-340 should be rewritten.

The mentioned paragraph was rewritten.

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

Qin et al. set out to investigate the role of mechanosensory feedback during swallowing and identify neural circuits that generate ingestion rhythms. They use Drosophila melanogaster swallowing as a model system, focusing their study on the neural mechanisms that control cibarium filling and emptying in vivo. They find that pump frequency is decreased in mutants of three mechanotransduction genes (nompC, piezo, and Tmc), and conclude that mechanosensation mainly contributes to the emptying phase of swallowing. Furthermore, they find that double mutants of nompC and Tmc have more pronounced cibarium pumping defects than either single mutants or Tmc/piezo double mutants. They discover that the expression patterns of nompC and Tmc overlap in two classes of neurons, md-C and md-L neurons. The dendrites of md-C neurons warp the cibarium and project their axons to the subesophageal zone of the brain. Silencing neurons that express both nompC and Tmc leads to severe ingestion defects, with decreased cibarium emptying. Optogenetic activation of the same population of neurons inhibited filling of the cibarium and accelerated cibarium emptying. In the brain, the axons of nompC∩Tmc cell types respond during ingestion of sugar but do not respond when the entire fly head is passively exposed to sucrose. Finally, the authors show that nompC∩Tmc cell types arborize close to the dendrites of motor neurons that are required for swallowing, and that swallowing motor neurons respond to the activation of the entire Tmc-GAL4 pattern.

Strengths:

• The authors rigorously quantify ingestion behavior to convincingly demonstrate the importance of mechanosensory genes in the control of swallowing rhythms and cibarium filling and emptying

• The authors demonstrate that a small population of neurons that express both nompC and Tmc oppositely regulate cibarium emptying and filling when inhibited or activated, respectively

• They provide evidence that the action of multiple mechanotransduction genes may converge in common cell types

Thank you for your insightful and detailed assessment of our work. Your constructive feedback will help to improve our manuscript.

Weaknesses:

• A major weakness of the paper is that the authors use reagents that are expressed in both md-C and md-L but describe the results as though only md-C is manipulated-Severing the labellum will not prevent optogenetic activation of md-L from triggering neural responses downstream of md-L. Optogenetic activation is strong enough to trigger action potentials in the remaining axons. Therefore, Qin et al. do not present convincing evidence that the defects they see in pumping can be specifically attributed to md-C.

Thank you for your comments. This is important point that we did not adequately address in the original preprint. We have obtained imaging and behavioral results that strongly suggest md-C, rather than md-L, are essential for swallowing behavior.

36 hours after the ablation of the labellum, the signals of md-L were hardly observable when GFP expression was driven by the intersection between Tmc-GAL4 & nompC-QF (see F Figure 3—figure supplement 1A). This observation indicates that the axons of md-L likely degenerated after 36 hours, and were unlikely to influence swallowing. Moreover, the projecting pattern of Tmc-GAL4 & nompC-QF>>GFP exhibited no significant changes in the brain post labellum ablation.

Furthermore, even after labellum ablation for 36 hours, flies exhibited responses to light stimulation (see Figure 3—figure supplement 1B-C, Video 5) when ReaChR was expressed in md-C. We thus reasoned that md-C but not md-L, plays a crucial role in the swallowing process.

• GRASP is known to be non-specific and prone to false positives when neurons are in close proximity but not synaptically connected. A positive GRASP signal supports but does not confirm direct synaptic connectivity between md-C/md-L axons and MN11/MN12.

In this study, we employed the nSyb-GRASP, wherein the GRASP is expressed at the presynaptic terminals by fusion with the synaptic marker nSyb. This method demonstrates an enhanced specificity compared to the original GRASP approach.

Additionally, we utilized +/ UAS-nSyb-spGFP1-10, lexAop-CD4-spGFP11 ; + / MN-LexA fruit flies as a negative control to mitigate potential false signals originating from the tool itself (Author response image 1, scale bar = 50μm). Beside the genotype Tmc-Gal4, Tub(FRT. Gal80) / UAS-nSyb-spGFP1-10, lexAop-CD4-spGFP11 ; nompC-QF, QUAS-FLP / MN-LexA fruit flies discussed in this manuscript, we also incorporated genotype Tmc-Gal4, Tub(FRT. Gal80) / lexAop-nSyb-spGFP1-10, UAS-CD4-spGFP11 ; nompC-QF, QUAS-FLP / MN-LexA fruit flies as a reverse control (Author response image 2). Unexpectedly, similar positive signals were observed, indicating that, positive signals may emerge due to close proximity between neurons even with nSyb-GRASP.

Author response image 1.

It should be noted that the existence of synaptic projections from motor neurons (MN) to md-C cannot be definitively confirmed at this juncture. At present, we can only posit the potential for synaptic connections between md-C and motor neurons. A more conclusive conclusion may be attainable with the utilization of comprehensive whole-brain connectome data in future studies.

Author response image 2.

• As seen in Figure 2—figure supplement 1, the expression pattern of Tmc-GAL4 is broader than md-C alone. Therefore, the functional connectivity the authors observe between Tmc expressing neurons and MN11 and 12 cannot be traced to md-C alone

It is true that the expression pattern of Tmc-GAL4 is broader than that of md-C alone. Our experiments, including those flies expressing TNT in Tmc+ neurons, demonstrated difficulties in emptying (Figure 2A, 2D). Notably, we encountered challenges in finding fly stocks bearing UAS>FRT-STOP-P2X2. Consequently, we opted to utilize Tmc-GAL4 to drive UAS-P2X2 instead. We believe that the results further support our hypothesis on the role of md-C in the observed behavioral change in emptying.

Overall, this work convincingly shows that swallowing and swallowing rhythms are dependent on several mechanosensory genes. Qin et al. also characterize a candidate neuron, md-C, that is likely to provide mechanosensory feedback to pumping motor neurons, but the results they present here are not sufficient to assign this function to md-C alone. This work will have a positive impact on the field by demonstrating the importance of mechanosensory feedback to swallowing rhythms and providing a potential entry point for future investigation of the identity and mechanisms of swallowing central pattern generators.

Reviewer #2 (Public Review):

In this manuscript, the authors describe the role of cibarial mechanosensory neurons in fly ingestion. They demonstrate that pumping of the cibarium is subtly disrupted in mutants for piezo, TMC, and nomp-C. Evidence is presented that these three genes are co-expressed in a set of cibarial mechanosensory neurons named md-C. Silencing of md-C neurons results in disrupted cibarial emptying, while activation promotes faster pumping and/or difficulty filling. GRASP and chemogenetic activation of the md-C neurons is used to argue that they may be directly connected to motor neurons that control cibarial emptying.

The manuscript makes several convincing and useful contributions. First, identifying the md-C neurons and demonstrating their essential role for cibarium emptying provides reagents for further studying this circuit and also demonstrates the important of mechanosensation in driving pumping rhythms in the pharynx. Second, the suggestion that these mechanosensory neurons are directly connected to motor neurons controlling pumping stands in contrast to other sensory circuits identified in fly feeding and is an interesting idea that can be more rigorously tested in the future.

At the same time, there are several shortcomings that limit the scope of the paper and the confidence in some claims. These include:

a) the MN-LexA lines used for GRASP experiments are not characterized in any other way to demonstrate specificity. These were generated for this study using Phack methods, and their expression should be shown to be specific for MN11 and MN12 in order to interpret the GRASP experiments.

Thanks for the suggestion. We have checked the expression pattern of MN-LexA, which is similar to MN-GAL4 used in previous work (Manzo et al., PNAS., 2012, PMID:22474379) . Here is the expression pattern:

Author response image 3.

b) There is also insufficient detail for the P2X2 experiment to evaluate its results. Is this an in vivo or ex vivo prep? Is ATP added to the brain, or ingested? If it is ingested, how is ATP coming into contact with md-C neuron if it is not a chemosensory neuron and therefore not exposed to the contents of the cibarium?

The P2X2 experimental preparation was done ex vivo. We immersed the fly in the imaging buffer, as described in the Methods section under Functional Imaging. Following dissection and identification of the subesophageal zone (SEZ) area under fluorescent microscopy, we introduced ATP slowly into the buffer, positioned at a distance from the brain

c) In Figure 3C, the authors claim that ablating the labellum will remove the optogenetic stimulation of the md-L neuron (mechanosensory neuron of the labellum), but this manipulation would presumably leave an intact md-L axon that would still be capable of being optogenetically activated by Chrimson.

Please refer to the corresponding answers for reviewer 1 and Figure 3—figure supplement 1.

d) Average GCaMP traces are not shown for md-C during ingestion, and therefore it is impossible to gauge the dynamics of md-C neuron activation during swallowing. Seeing activation with a similar frequency to pumping would support the suggested role for these neurons, although GCaMP6s may be too slow for these purposes.

Profiling the dynamics of md-C neuron activation during swallowing is crucial for unraveling the operational model of md-C and validating our proposed hypothesis. Unfortunately, our assay faces challenges in detecting probable 6Hz fluorescent changes with GCaMP6s.

In general, we observed an increase of fluorescent signals during swallowing, but movement of alive flies during swallowing influenced the imaging recording, so we could not depict a decent tracing for calcium imaging for md-C neurons. To enhance the robustness of our findings, patching the md-C neurons would be a more convincing approach. As illustrated in Figure 2, the somata of md-C neurons are situated in the cibarium rather than the brain. patching of the md-C neuron somata in flies during ingestion is difficult.

e) The negative result in Figure 4K that is meant to rule out taste stimulation of md-C is not useful without a positive control for pharyngeal taste neuron activation in this same preparation.

We followed methods used in the previous work (Chen et al., Cell Rep., 2019, PMID:31644916), which we believe could confirm that md-C do not respond to sugars.

In addition to the experimental limitations described above, the manuscript could be organized in a way that is easier to read (for example, not jumping back and forth in figure order).

Thanks for your suggestion and the manuscript has been reorganized.

Reviewer #3 (Public Review):

Swallowing is an essential daily activity for survival, and pharyngo-laryngeal sensory function is critical for safe swallowing. In Drosophila, it has been reported that the mechanical property of food (e.g. Viscosity) can modulate swallowing. However, how mechanical expansion of the pharynx or fluid content sense and control swallowing was elusive. Qin et al. showed that a group of pharyngeal mechanosensory neurons, as well as mechanosensory channels (nompC, Tmc, and Piezo), respond to these mechanical forces for regulation of swallowing in Drosophila melanogaster.

Strengths:

There are many reports on the effect of chemical properties of foods on feeding in fruit flies, but only limited studies reported how physical properties of food affect feeding especially pharyngeal mechanosensory neurons. First, they found that mechanosensory mutants, including nompC, Tmc, and Piezo, showed impaired swallowing, mainly the emptying process. Next, they identified cibarium multidendritic mechanosensory neurons (md-C) are responsible for controlling swallowing by regulating motor neuron (MN) 12 and 11, which control filling and emptying, respectively.

Weaknesses:

While the involvement of md-C and mechanosensory channels in controlling swallowing is convincing, it is not yet clear which stimuli activate md-C. Can it be an expansion of cibarium or food viscosity, or both? In addition, if rhythmic and coordinated contraction of muscles 11 and 12 is essential for swallowing, how can simultaneous activation of MN 11 and 12 by md-C achieve this? Finally, previous reports showed that food viscosity mainly affects the filling rather than the emptying process, which seems different from their finding.

We have confirmed that swallowing sucrose water solution activated md-C neurons, while sucrose water solution alone could not (Figure 4J-K). We hypothesized that the viscosity of the food might influence this expansion process.

While we were unable to delineate the activation dynamics of md-C neurons, our proposal posits that these neurons could be activated in a single pump cycle, sequentially stimulating MN12 and MN11. Another possibility is that the activation of md-C neurons acts as a switch, altering the oscillation pattern of the swallowing central pattern generator (CPG) from a resting state to a working state.

In the experiments with w1118 flies fed with MC (methylcellulose) water, we observed that viscosity predominantly affects the filling process rather than the emptying process, consistent with previous findings. This raises an intriguing question. Our investigation into the mutation of mechanosensitive ion channels revealed a significant impact on the emptying process. We believe this is due to the loss of mechanosensation affecting the vibration of swallowing circuits, thereby influencing both the emptying and filling processes. In contrast, viscosity appears to make it more challenging for the fly to fill the cibarium with food, primarily attributable to the inherent properties of the food itself.

Reviewer #4 (Public Review):

A combination of optogenetic behavioral experiments and functional imaging are employed to identify the role of mechanosensory neurons in food swallowing in adult Drosophila. While some of the findings are intriguing and the overall goal of mapping a sensory to motor circuit for this rhythmic movement are admirable, the data presented could be improved.

The circuit proposed (and supported by GRASP contact data) shows these multi-dendritic neurons connecting to pharyngeal motor neurons. This is pretty direct - there is no evidence that they affect the hypothetical central pattern generator - just the execution of its rhythm. The optogenetic activation and inhibition experiments are constitutive, not patterned light, and they seem to disrupt the timing of pumping, not impose a new one. A slight slowing of the rhythm is not consistent with the proposed function.

Motor neurons implicated in patterned motions can be considered effectors of Central Pattern Generators (CPGs)(Marder et al., Curr Biol., 2001, PMID: 11728329; Hurkey et al., Nature., 2023, PMID:37225999). Given our observation of the connection between md-C neurons and motor neurons, it is reasonable to speculate that md-C neurons influence CPGs. Compared to the patterned light (0.1s light on and 0.1s light off) used in our optogenetic experiments, we noted no significant changes in their responses to continuous light stimulation. We think that optogenetic methods may lead to overstimulation of md-C neurons, failing to accurately mimic the expansion of the cibarium during feeding.

Dysfunction in mechanosensitive ion channels or mechanosensory neurons not only disrupts the timing of pumping but also results in decreased intake efficiency (Figure 1E). The water-swallowing rhythm is generally stable in flies, and swallowing is a vital process that may involve redundant ion channels to ensure its stability.

The mechanosensory channel mutants nompC, piezo, and TMC have a range of defects. The role of these channels in swallowing may not be sufficiently specific to support the interpretation presented. Their other defects are not described here and their overall locomotor function is not measured. If the flies have trouble consuming sufficient food throughout their development, how healthy are they at the time of assay? The level of starvation or water deprivation can affect different properties of feeding - meal size and frequency. There is no description of how starvation state was standardized or measured in these experiments.

Defects in mechanosensory channel mutants nompC, piezo, and TMC, have been extensively investigated (Hehlert et al., Trends Neurosci., 2021, PMID:332570000). Mutations in these channels exhibit multifaceted effects, as illustrated in our RNAi experiments (see Figure 2E). Deprivation of water and food was performed in empty fly vials. It's important to note that the duration of starvation determines the fly's willingness to feed but not the pump frequency (Manzo et al., PNAS., 2012, PMID:22474379).

In most cases, female flies were deprived water and food in empty vials for 24 hours because after that most flies would be willing to drink water. The deprivation time is 12 hours for flies with nompC and Tmc mutated or flies with Kir2.1 expressed in md-C neurons, as some of these flies cannot survive 24h deprivation.

The brain is likely to move considerably during swallow, so the GCaMP signal change may be a motion artifact. Sometimes this can be calculated by comparing GCaMP signal to that of a co-expressed fluorescent protein, but there is no mention that this is done here. Therefore, the GCaMP data cannot be interpreted.

We did not co-express a fluorescent protein with GCaMP for md-C. The head of the fly was mounted onto a glass slide, and we did not observe significant signal changes before feeding.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

.>Abstract: I disagree that swallow is the first step of ingestion. The first paragraph also mentions the final checkpoint before food ingestion. Perhaps sufficient to say that swallow is a critical step of ingestion.

Indeed, it is not rigorous enough to say “first step”. This has been replaced by “early step”.

Introduction:

Line 59: "Silence" should be "Silencing"

This has been replaced.

Results:

Lines 91-92: I am not clear about what this means. 20% of nompC and 20% of wild-type flies exhibit incomplete filling? So nompC is not different from wild-type?

Sorry for the mistake. Viscous foods led to incomplete emptying (not incomplete filling), as displayed in Video 4. The swallowing behavior differs between nompC mutants and wild-type flies, as illustrated in Figure 1C, Figure 1—figure supplement 1A-C and video 1&5.

When fed with 1% MC water solution (Figure 1—figure supplement 1E-H). We found that when fed with 1% MC watere solution, Tmc or piezo mutants displayed incomplete emptying, which could constitute a long time proportion of swallowing behavior; while only 20% of nompC flies and 20% of wild-type flies sporadically exhibit incomplete emptying, which is significantly different. Though the percent of flies displaying incomplete pump is similar between nompC mutant and wild-type files, you can find it quite different in video 1 and 5.

Line 94: Should read: “while for foods with certain viscosity, the pump of Tmc or piezo mutants might"

What evidence is there for weakened muscle motion? The phenotypes of all three mutants is quite similar, so concluding that they have roles in initiation versus swallowing strength is not well supported -this would be better moved to the discussion since it is speculative.

Muscles are responsible for pumping the bolus from the mouth to the crop. In the case of Tmc or piezo mutants, as evidenced by incomplete filling for viscous foods (see Video 4), we speculate that the loss of sensory stimuli leads to inadequate muscle contraction. The phenotypes observed in Tmc and piezo mutants are similar yet distinct from those of the wild-type or nompC mutant, as shown in Video 1 and 4. The phrase "due to weakened muscle motion" has been removed for clarity.

Line 146: If md-L neurons are also labeled by this intersection, then you are not able to know whether the axons seen in the brain are from md-L or md-C neurons. Line 148: cutting the labellum is not sufficient to ablate md-L neurons. The projections will still enter the brain and can be activated with optogenetics, even after severing the processes that reside in the labellum.

Please refer to the responses for reviewer #1 (Public Review):” A major weakness of the paper…” and Figure 4.

Line 162: If the fly head alone is in saline, do you know that the sucrose enters the esophagus? The more relevant question here is whether the md-C neurons respond to mechanical force. If you could artificially inflate the cibarium with air and see the md-C neurons respond that would be a more convincing result. So far you only know that these are activated during ingestion, but have not shown that they are activated specifically by filling or emptying. In addition, you are not only imaging md-C (md-L is also labeled). This caveat should be mentioned.

We followed the methods outlined in the previous work (Chen et al., Cell Rep., 2019, PMID:31644916), which suggested that md-C neurons do not respond to sugars. While we aimed to mechanically stimulate md-C neurons, detecting signal changes during different steps of swallowing is challenging. This aspect could be further investigated in subsequent research with the application of adequate patch recording or two-photon microscopy (TPM).

Figure 3: It is not clear what the pie charts in Figure 3 A refer to. What are the three different rows, and what does blue versus red indicate?

Figure 3A illustrates three distinct states driven by CsChrimson light stimulation of md-C neurons, with the proportions of flies exhibiting each state. During light activation, flies may display difficulty in filling, incomplete filling, or a normal range of pumping. The blue and red bars represent the proportions of flies showing the corresponding state, as indicated by the black line.

Figure 4: Where are the example traces for J? The comparison in K should be average dF/F before ingestion compared with average dF/F during ingestion. Comparing the in vitro response to sucrose to the in vivo response during ingestion is not a useful comparison.

Please refer to the answers for reviewer #2 question d).

Reviewer #2 (Recommendations For The Authors):

Suggested experiments that would address some of my concerns listed in the public review include:

a) high resolution SEZ images of MN-LexA lines crossed to LexAop-GFP to demonstrate their specificity

b) more detail on the P2X2 experiment. It is hard to make suggestions beyond that without first seeing the details.

c) presenting average GCaMP traces for all calcium imaging results

d) to rule out taste stimulation of md-C (Figure 4K) I would suggest performing more extensive calcium imaging experiments with different stimuli. For example, sugar, water, and increasing concentrations of a neutral osmolyte (e.g. PEG) to suppress the water response. I think that this is more feasible than trying to get an in vitro taste prep to be convincing.

Please refer to the responses for public review of reviewer #2.

Reviewer #3 (Recommendations For The Authors):

Below I list my suggestions as well as criticisms.

(1) It would be excellent if the authors could demonstrate whether varying levels of food viscosity affect md-C activation.

That is a good point, and could be studied in future work.

(2) It is not clear whether an intersectional approach using TMC-GAL4 and nompC-QF abolishes labelling of the labellar multidendritic neurons. If this is the case, please show labellar multidendritic neurons in TMC-GAL4 only flies and flies using the intersectional approach. Along with this question, I am concerned that labellum-removed flies could be used for feeding assay.

Intersectional labelling using TMC-GAL4 and nompC-QF could not abolish labelling of the labellar multidendritic neurons (Author response image 4). Labellum-removed flies could be used for feeding assay (Figure 3—figure supplement 1B-C, video 5), but once LSO or cibarium of fly was damaged, swallowing behavior would be affected. Removing labellum should be very careful.

Author response image 4.

(3) Please provide the detailed methods for GRASP and include proper control.

Please refer to the responses for public review of reviewer #1.

(4) The authors hypothesized that md-C sequentially activates MN11 and 12. Is the time gap between applying ATP on md-C and activation of MN11 or MN12 different? Please refer to the responses for public review of reviewer #3. The time gap between applying ATP on md-C and activation of MN11 or MN12 didn’t show significant differences, and we think the reason is that the ex vivo conditions could not completely mimic in vivo process.

I found the manuscript includes many errors, which need to be corrected.

(1) The reference formatting needs to be rechecked, for example, lines 37, 42, and 43.

(2) Line 44-46: There is some misunderstanding. The role of pharyngeal mechanosensory neurons is not known compared with chemosensory neurons.

(3) Line 49: Please specify which type of quality of food. Chemical or physical?

(4) Line 80 and Figure 1B-D Authors need to put filling and emptying time data in the main figure rather than in the supplementary figure. Otherwise, please cite the relevant figures in the text(S1A-C).

(5) Line 84-85; Is "the mutant animals" indicating only nompC? Please specify it.

(6) Figure 1a: It is hard to determine the difference between the series of images. And also label filling and emptying under the time.

(7) S1E-H: It is unclear what "Time proportion of incomplete pump" means. Please define it.

(8) Please reorganize the figures to follow the order of the text, for example, figures 2 and 4

(9) Figure 4A. There is mislabelling in Figure 4A. It is supposed to be phalloidin not nc82.

(10) Figure 4K: It does not match the figure legend and main text.

(11) Figure 4D and G: Please indicate ATP application time point.

Thanks for your correction and all the points mentioned were revised.

Reviewer #4 (Recommendations For The Authors):

The figures need improvement. 1A has tiny circles showing pharynx and any differences are unclear.

The expression pattern of some of these drivers (Supplement) seems quite broad. The tmc nompC intersection image in Figure 1F is nice but the cibarium images are hard to interpret: does this one show muscle expression? What are "brain" motor neurons? Where are the labellar multi-dendritic neurons?

Tmc nompC intersection image show no expression in muscles. Somata of motor neurons 12 or 11 situated at SEZ area of brain, while somata of md-C neurons are in the cibarium. Image of md-L neurons was posted in response for reviewer #3 (Recommendations For The Authors):

Why do the assays alternate between swallowing food and swallowing water?

Thank for your suggestion, figure 1A has been zoomed-in. The Tmc nompC intersection image in Figure 2F displayed the position of md-C neurons in a ventral perspective, and muscles were not labelled. We stained muscles in cibarium by phalloidin and the image is illustrated in Figure 4A, while we didn’t find overlap between md-C neurons and muscles. Image of md-L neurons were posted as Author response image 4.

In the majority of our experiments, we employed water to test swallowing behavior, while we used methylcellulose water solution to test swallowing behavior of mechanoreceptor mutants, and sucrose solution for flies with md-C neurons expressing GCaMP since they hardly drank water when their head capsules were open.

How starved or water-deprived were the flies?

One day prior to the behavioral assays, flies were transferred to empty vials (without water or food) for 24 hours for water deprivation. Flies who could not survive 24h deprivation would be deprived for 12h.

How exactly was the pumping frequency (shown in Fig 1B) measured? There is no description in the methods at all. If the pump frequency is scored by changes in blue food intensity (arbitrary units?), this seems very subjective and maybe image angle dependent. What was camera frame rate? Can it capture this pumping speed adequately? Given the wealth of more quantitative methods for measuring food intake (eg. CAFE, flyPAD), it seems that better data could be obtained.

How was the total volume of the cibarium measured? What do the pie charts in Figure 3A represent?

The pump frequency was computed as the number of pumps divided by the time scale, following the methodology outlined in Manzo et al., 2012. Swallowing curves were plotted using the inverse of the blue food intensity in the cibarium. In this representation, ascending lines signify filling, while descending lines indicate emptying (see Figure 2D, 3B). We maintain objectivity in our approach since, during the recording of swallowing behavior, the fly was fixed, and we exclusively used data for analysis when the Region of Interest (ROI) was in the cibarium. This ensures that the intensity values accurately reflect the filling and emptying processes. Furthermore, we conducted manual frame-by-frame checks of pump frequency, and the results align with those generated by the time series analyzer V3 of ImageJ.

For the assessment of total volume of ingestion, we referred the methods of CAFE, utilizing a measurable glass capillary. We then calculated the ingestion rate (nL/s) by dividing the total volume of ingestion by the feeding time.

The changes seem small, in spite of the claim of statistical significance.

The observed stability in pump frequency within a given genotype underscores the significance of even seemingly small changes, which is statistically significant. We speculate that the stability in swallowing frequency suggests the existence of a redundant mechanism to ensure the robustness of the process. Disruption of one channel might potentially be partially compensated for by others, highlighting the vital nature of the swallowing mechanism.

How is this change in pump frequency consistent with defects in one aspect of the cycle - either ingestion (activation) or expulsion (inhibition)?

Please refer to Figure 2, 3. Both filling and emptying process were affects, while inhibition mainly influences emptying time (Figure 1—figure supplement 1).

for the authors:

Line 48: extensively

Line 62 - undiscovered.

Line 107, 463: multi

Line 124: What is "dysphagia?" This is an unusual word and should be defined.

Line 446: severe

Line 466: in the cibarium or not?

Thanks for your correction and all the places mentioned were revised.

Reviewer #2 (Public Review):

The goal of the present study is to better understand the 'control objectives' that subjects adopt in a video-game-like virtual-balancing task. In this task, the hand must move in the opposite direction from a cursor. For example, if the cursor is 2 cm to the right, the subject must move their hand 2 cm to the left to 'balance' the cursor. Any imperfection in that opposition causes the cursor to move. E.g., if the subject were to move only 1.8 cm, that would be insufficient, and the cursor would continue to move to the right. If they were to move 2.2 cm, the cursor would move back toward the center of the screen. This return to center might actually be 'good' from the subject's perspective, depending on whether their objective is to keep the cursor still or keep it near the screen's center. Both are reasonable 'objectives' because the trial fails if the cursor moves too far from the screen's center during each six-second trial.

This task was recently developed for use in monkeys (Quick et al., 2018), with the intention of being used for the study of the cortical control of movement, and also as a task that might be used to evaluate BMI control algorithms. The purpose of the present study is to better characterize how this task is performed. What sort of control policies are used. Perhaps more deeply, what kind of errors are those policies trying to minimize? To address these questions, the authors simulate control-theory style models and compare with behavior. They do in both in monkeys and in humans.

These goals make sense as a precursor to future recording or BMI experiments. The primate motor-control field has long been dominated by variants of reaching tasks, so introducing this new task will likely be beneficial. This is not the first non-reaching task, but it is an interesting one and it makes sense to expand the presently limited repertoire of tasks. The present task is very different from any prior task I know of. Thus, it makes sense to quantify behavior as thoroughly as possible in advance of recordings. Understanding how behavior is controlled is, as the authors note, likely to be critical to interpreting neural data.

From this perspective - providing a basis for interpreting future neural results - the present study is fairly successful. Monkeys seem to understand the task properly, and to use control policies that are not dissimilar from humans. Also reassuring is the fact that behavior remains sensible even when task-difficulty become high. By 'sensible' I simply mean that behavior can be understood as seeking to minimize error: position, velocity, or (possibly) both, and that this remains true across a broad range of task difficulties. The authors document why minimizing position and minimizing velocity are both reasonable objectives. Minimizing velocity is reasonable, because a near-stationary cursor can't move far in six seconds. Minimizing position error is reasonable, because the trial won't fail if the cursor doesn't stray far from the center. This is formally demonstrated by simulating control policies: both objectives lead to control policies that can perform the task and produce realistic single-trial behavior. The authors also demonstrate that, via verbal instruction, they can induce human subjects to favor one objective over the other. These all seem like things that are on the 'need to know' list, and it is commendable that this amount of care is being taken before recordings begin, as it will surely aid interpretation.

Yet as a stand-alone study, the contribution to our understanding of motor control is more limited. The task allows two different objectives (minimize velocity, minimize position) to be equally compatible with the overall goal (don't fail the trial). Or more precisely, there exists a range of objectives with those two at the extreme. So it makes sense that different subjects might choose to favor different objectives, and also that they can do so when instructed. But has this taught us something about motor control, or simply that there is a natural ambiguity built into the task? If I ask you to play a game, but don't fully specify the rules, should I be surprised that different people think the rules are slightly different?

The most interesting scientific claim of this study is not the subject-to-subject variability; the task design makes that quite likely and natural. Rather, the central scientific result is the claim that individual subjects are constantly switching objectives (and thus control policies), such that the policy guiding behavior differs dramatically even on a single-trial basis. This scientific claim is supported by a technical claim: that the authors' methods can distinguish which objective is in use, even on single trials. I am uncertain of both claims.

Consider Figure 8B, which reprises a point made in Figure 1&3 and gives the best evidence for trial-to-trial variability in objective/policy. For every subject, there are two example trials. The top row of trials shows oscillations around the center, which could be consistent with position-error minimization. The bottom row shows tolerance of position errors so long as drift is slow, which could be consistent with velocity-error minimization. But is this really evidence that subjects were switching objectives (and thus control policies) from trial to trial? A simpler alternative would be a single control policy that does not switch, but still generates this range of behaviors. The authors don't really consider this possibility, and I'm not sure why. One can think of a variety of ways in which a unified policy could produce this variation, given noise and the natural instability of the system.

Indeed, I found that it was remarkably easy to produce a range of reasonably realistic behaviors, including the patterns that the authors interpret as evidence for switching objectives, based on a simple fixed controller. To run the simulations, I made the simple assumption that subjects simply attempt to match their hand position to oppose the cursor position. Because subjects cannot see their hand, I assumed modest variability in the gain, with a range from -1 to -1.05. I assumed a small amount of motor noise in the outgoing motor command. The resulting (very simple) controller naturally displayed the basic range of behaviors observed across trials (see Image 1)

Peer review image 1.

Some trials had oscillations around the screen center (zero), which is the pattern the authors suggest reflects position control. In other trials the cursor was allowed to drift slowly away from the center, which is the pattern the authors suggest reflects velocity control. This is true even though the controller was the same on every trial. Trial-to-trial differences were driven both by motor noise and by the modest variability in gain. In an unstable system, small differences can lead to (seemingly) qualitatively different behavior on different trials.

This simple controller is also compatible with the ability of subjects to adapt their strategy when instructed. Anyone experienced with this task likely understands (or has learned) that moving the hand slightly more than 'one should' will tend to shepherd the cursor back to center, at the cost of briefly high velocity. Using this strategy more sparingly will tend to minimize velocity even if position errors persist. Thus, any subject using this control policy would be able to adapt their strategy via a modest change in gain (the gain linking visible cursor position to intended hand position).

This model is simple, and there may be reasons to dislike it. But it is presumably a reasonable model. The nature of the task is that you should move your hand opposite where the cursor is. Because you can't see your hand, you will make small mistakes. Due to the instability of the system, those small mistakes have large and variable effects. This feature is likely common to other controllers as well; many may explicitly or implicitly blend position and velocity control, with different trials appearing more dominated by one versus the other. Given this, I think the study presents only weak evidence that individual subjects are switching their objective on individual trials. Indeed, the more parsimonious explanation may be that they aren't. While the study certainly does demonstrate that the control policy can be influenced by verbal instructions, this might be a small adjustment as noted above.

I thus don't feel convinced that the authors can conclusively tell us the true control policy being used by human and monkey subjects, nor whether that policy is mostly fixed or constantly switching. The data are potentially compatible with any of these interpretations, depending on which control-style model one prefers.

I see a few paths that the authors might take if they chose.<br /> --First, my reasoning above might be faulty, or there might be additional analyses that could rule out the possibility of a unified policy underlying variable behavior. If so, the authors may be able to reject the above concerns and retain the present conclusions. The main scientifically novel conclusion of the present study is that subjects are using a highly variable control policy, and switching on individual trials. If this is indeed the case, there may be additional analyses that could reveal that.<br /> --Second, additional trial types (e.g., with various perturbations) might be used as a probe of the control policy. As noted below, there is a long history of doing this in the pursuit system. That additional data might better disambiguate control policies both in general, and across trials.<br /> --Third, the authors might find that a unified controller is actually a good (and more parsimonious) explanation. Which might actually be a good thing from the standpoint of future experiments. Interpretation of neural data is likely to be much easier if the control policy being instantiated isn't in constant flux.

In any case, I would recommend altering the strength of some conclusions, particularly the conclusion that the presented methods can reliably discriminate amongst objectives/policies on individual trials. This is mentioned as a major motivation on multiple occasions, but in most of these instances, the subsequent analysis infers the objective only across trial (e.g., one must observe a scatterplot of many trials). By Figure 7, they do introduce a method for inferring the control policy on individual trials, and while this seems to work considerably better than chance, it hardly appears reliable.

In this same vein I would suggest toning down aspects of the Introduction and Discussion. The Introduction in particular is overly long, and tries to position the present study as unique in ways that seem strained. Other studies have built links between human behavior, monkey behavior, and monkey neural data (for just one example, consider the corpus of work from the Scott lab that includes Pruszynski et al. 2008 and 2011). Other studies have used highly quantitative methods to infer the objective function used by subjects (e.g. Kording and Wolpert 2004). The very issue that is of interest in the present study - velocity-error-minimization versus position-error-minimization - has been extensively addressed in the smooth pursuit system. That field has long combined quantitative analyses of behavior in humans and monkeys, along with neural recordings. Many pursuit experiments used strategies that could be fruitfully employed to address the central questions of the present study. For example, error stabilization was important for dissecting the control policy used by the pursuit system. By artificially stabilizing the error (position or velocity) at zero, or at some other value, one can determine the system's response. The classic Rashbass step (1961) put position and velocity errors in opposition, to see which dominates the response. Step and sinusoidal perturbations were useful in distinguishing between models, as was the imposition of artificially imposed delays. The authors note the 'richness' of the behavior in the present task, and while one could say the same of pursuit, it was still the case that specific and well-thought through experimental manipulations were pretty critical. It would be better if the Introduction considered at least some of the above-mentioned work (or other work in a similar vein). While most would agree with the motivations outlined by the authors - they are logical and make sense - the present Introduction runs the risk of overselling the present conclusions while underselling prior work.

Author Response

Reviewer #1 (Public Review):

The manuscript, "A versatile high-throughput assay based on 3D ring-shaped cardiac tissues generated from human induced pluripotent stem cell-derived cardiomyocytes" developed a unique culture platform with PEG hydrogel that facilitates the in-situ measurement of contractile dynamics of the engineered cardiac rings. The authors optimized the tissue seeding conditions, demonstrated tissue morphology with expressions of cardiac and fibroblast markers, mathematically modeled the equation to derive contractile forces and other parameters based on imaging analysis, and ended by testing several compounds with known cardiac responses.

To strengthen the paper, the following comments should be considered:

1) This paper provided an intriguing platform that creates miniature cardiac rings with merely thousands of CMs per tissue in a 96-well plate format. The shape of the ring and the squeezing motion can recapitulate the contraction of the cardiac chamber to a certain degree. However, Thavandiran et al (PNAS 2013) created a larger version of the cardiac ring and found the electrical propagation revealed spontaneous infinite loop-like cycles of activation propagation traversing the ring. This model was used to mimic a reentrant wave during arrhythmia. Therefore, it presents great concerns if a large number of cardiac tissues experience arrhythmia by geometry-induced re-entry current and cannot be used as a healthy tissue model. It would be interesting to see the impulse propagation/calcium transient on these miniature cardiac rings and evaluate the % of arrhythmia occurrence.

The size is a key factor impacting the electrical propagation within the generated tissues. Our ring-shaped cardiac tissues have a diameter of 360µm, which is largely smaller than other tissues proposed so far, including in Thavandiran et al (PNAS 2013) where circular tissues had a reported size > 1mm. As shown in Figure 4E (and highlighted below in Author response image 1), tissues under basal conditions display regular beating rates without spontaneous arrhythmias. Videos also show that the tissue contraction is homogeneous around the pillar, suggesting that the smaller size favors the electrical propagation and limits the occurrence of spontaneous reentrant waves. Optical mapping measurements will be performed in the future to assess the occurrence of reentrant waves.

**Author response image 1. **

Poincaré plot showing the plots between successive RR intervals (Data from Figure 4E in basal conditions). Linear regression with 95% confidence interval indicates identity.

2) The platform can produce 21 cardiac rings per well in 96-well plates. The throughput has been the highest among competing platforms. The resulting tissues have good sarcomere striation due to the strain from the pillars. Now the emerging questions are culture longevity and reproducibility among tissues. According to Figure 1E, there was uneven ring formation around the pillar, which leads to the tissue thinning and breaking off. There is only 50% survival after 20 days of culture in the optimized seeding group. Is there any way to improve it? The tissues had two compartments, cardiac and fibroblast-rich regions, where fibroblasts are responsible for maintaining the attachment to the glass slides. Do the cardiac rings detach from the glass slides and roll up? The SD of the force measurement is a quarter of the value, which is not ideal with such a high replicate number. As the platform utilizes imaging analysis to derive contractile dynamics, calibration should be done based on the angle and the distance of the camera lens to the individual tissues to reduce the error. On the other hand, how reproducible of the pillars? It is highly recommended to mechanically evaluate the consistency of the hydrogel-based pillars across different wells and within the wells to understand the variance. Figure 2B reports the early results obtained as the system was tested and developed. Since then, we have tested different iPSC lines and confirm that the overall yield is higher (up to 20 tissues at D14 for some cell lines), however dependent of cell lines.

The tissues do not detach from the glass slides. It is very rare to see tissues roll up on the central pillar. As shown in Figure 1B, the pillars have a specific shape to avoid tissues to roll up as they develop and contract.

3) Does the platform allow the observation of non-synchronized beating when testing with compounds? This can be extremely important as the intended applications of this platform are drug testing and cardiac disease modeling. The author should elaborate on the method in the manuscript and explain the obtained results in detail. The arrhythmogenic effect of a drug can be derived from the regularity of the beat-to-beat time. Indeed, we show that dofetilide increases the variability in the beat-to-beat time by plotting for each beat, the beat-to-beat time with the next beat as a function of the beat-to-beat time with the previous beat.

4) The results of drug testing are interesting. Isoproterenol is typically causing positive chronotropic and positive inotropic responses, where inotropic responses are difficult to obtain due to low tissue maturity. It is inconsistent with other reported results that cardiac rings do not exhibit increased beating frequency, but slightly increased forces only. Zhao et al were using electrical pacing at a defined rate during force measurement, whereas the ring constructs are not.

We agree. The difference in the response to isoproterenol with previous papers may be explained by different incubation timing with the drug. In our case, the tissues were incubated for 5 minutes at 37•C before being recorded.

Overall, the manuscript is well written and the designed platform presented the unique advantages of high throughput cardiac tissue culture. Besides the contractile dynamics and IHC images, the paper lacks other cardiac functional evaluations, such as calcium handling, impulse propagation, and/or electrophysiology. The culture reproducibility (high SD) and longevity (<20 days) still remain unsolved.

Since the submission, we have managed to keep some tissues and analyze them up to 32 days. At that time point the tissues are still beating. Nevertheless, a specific study concerning tissue longevity has not been carried out as the tissues were usually fixed after 14 days to be stained and analyze their structure.

Reviewer #2 (Public Review):

The authors should be commended for developing a high throughput platform for the formation and study of human cardiac tissues, and for discussing its potential, advantages and limitations. The study is addressing some of the key needs in the use of engineered cardiac tissues for pharmacological studies: ease of use, reproducible preparation of tissues, and high throughput.

There are also some areas where the manuscript should be improved. The design of the platform and the experimental design should be described in more detail.

It would be of interest to comprehensively document the progression of tissue formation. To this end, it would be helpful to show the changes in tissue structure through a series of images that would correspond to the progression of contractile properties shown in Figure 3.

Our results indicate that the fibroblasts/cardiomyocytes segregation likely happens as soon as the tissue is formed, as the fibroblasts are critical for tissue generation. The change with time in the shape of the contractile ring is reported in Figure 1E, with a series of images which correspond to the timepoints of Figure 3.

The very interesting tissue morphology (separation into the two regions) that was observed in this study is inviting more discussion.

Finally, the reader would benefit from more specific comparisons of the contractile function of cardiac tissues measured in this study with data reported for other cardiac tissue models.

Author Response:

Assessment note: “Whereas the results and interpretations are generally solid, the mechanistic aspect of the work and conclusions put forth rely heavily on in vitro studies performed in cultured L6 myocytes, which are highly glycolytic and generally not viewed as a good model for studying muscle metabolism and insulin action.”

While we acknowledge that in vitro models may not fully recapitulate the complexity of in vivo systems, we believe that our use of L6 myotubes is appropriate for studying the mechanisms underlying muscle metabolism and insulin action. As mentioned below (reviewer 2, point 1), L6 myotubes possess many important characteristics relevant to our research, including high insulin sensitivity and a similar mitochondrial respiration sensitivity to primary muscle fibres. Furthermore, several studies have demonstrated the utility of L6 myotubes as a model for studying insulin sensitivity and metabolism, including our own previous work (PMID: 19805130, 31693893, 19915010).

In addition, we have provided evidence of the similarities between L6 cells overexpressing SMPD5 and human muscle biopsies at protein levels and the reproducibility of the negative correlation between ceramide and Coenzyme Q observed in L6 cells in vivo, specifically in the skeletal muscle of mice in chow diet. These findings support the relevance of our in vitro results to in vivo muscle metabolism.

Finally, we will supplement our findings by demonstrating a comparable relationship between ceramide and Coenzyme Q in mice exposed to a high-fat diet, to be shown in Supplementary Figure 4 H-I. Further animal experiments will be performed to validate our cell-line based conclusions. We hope that these additional results address the concerns raised by the reviewer and further support the relevance of our in vitro findings to in vivo muscle metabolism and insulin action.

Points from reviewer 1:

1. Although the authors' results suggest that higher mitochondrial ceramide levels suppress cellular insulin sensitivity, they rely solely on a partial inhibition (i.e., 30%) of insulin-stimulated GLUT4-HA translocation in L6 myocytes. It would be critical to examine how much the increased mitochondrial ceramide would inhibit insulin-induced glucose uptake in myocytes using radiolabel deoxy-glucose.

Response: The primary impact of insulin is to facilitate the translocation of glucose transporter type 4 (GLUT4) to the cell surface, which effectively enhances the maximum rate of glucose uptake into cells. Therefore, assessing the quantity of GLUT4 present at the cell surface in non-permeabilized cells is widely regarded as the most reliable measure of insulin sensitivity (PMID: 36283703, 35594055, 34285405). Additionally, plasma membrane GLUT4 and glucose uptake are highly correlated. Whilst we have routinely measured glucose uptake with radiolabelled glucose in the past, we do not believe that evaluating glucose uptake provides a better assessment of insulin sensitivity than GLUT4.

We will clarify the use of GLUT4 translocation in the Results section:

“...For this reason, several in vitro models have been employed involving incubation of insulin sensitive cell types with lipids such as palmitate to mimic lipotoxicity in vivo. In this study we will use cell surface GLUT4-HA abundance as the main readout of insulin response...”

1. Another important question to be addressed is whether glycogen synthesis is affected in myocytes under these experimental conditions. Results demonstrating reductions in insulin-stimulated glucose transport and glycogen synthesis in myocytes with dysfunctional mitochondria due to ceramide accumulation would further support the authors' claim.

Response: We have carried out supplementary experiments to investigate glycogen synthesis in our insulin-resistant models. Our approach involved L6-myotubes overexpressing the mitochondrial-targeted construct ASAH1 (as described in Fig. 3). We then challenged them with palmitate and measured glycogen synthesis using 14C radiolabeled glucose. Our observations indicated that palmitate suppressed insulin-induced glycogen synthesis, which was effectively prevented by the overexpression of ASAH1 (N = 5, * p<0.05). These results provide additional evidence highlighting the role of dysfunctional mitochondria in muscle cell glucose metabolism.

These data will be added to Supplementary Figure 4K and the results modified as follows:

“Notably, mtASAH1 overexpression protected cells from palmitate-induced insulin resistance without affecting basal insulin sensitivity (Fig. 3E). Similar results were observed using insulin-induced glycogen synthesis as an ortholog technique for Glut4 translocation. These results provide additional evidence highlighting the role of dysfunctional mitochondria in muscle cell glucose metabolism (Sup. Fig. 5K). Importantly, mtASAH1 overexpression did not rescue insulin sensitivity in cells depleted…”

We will add to the method section:

“L6 myotubes overexpressing ASAH were grown and differentiated in 12-well plates, as described in the Cell lines section, and stimulated for 16 h with palmitate-BSA or EtOH-BSA, as detailed in the Induction of insulin resistance section.

On day seven of differentiation, myotubes were serum starved in plain DMEM for 3 and a half hours. After incubation for 1 hour at 37C with 2 µCi/ml D-[U-14C]-glucose in the presence or absence of 100 nM insulin, glycogen synthesis assay was performed, as previously described (Zarini S. et al., J Lipid Res, 63(10): 100270, 2022).”

1. In addition, it would be critical to assess whether the increased mitochondrial ceramide and consequent lowering of energy levels affect all exocytic pathways in L6 myoblasts or just the GLUT4 trafficking. Is the secretory pathway also disrupted under these conditions?

Response: As the secretory pathway primarily involves the synthesis and transportation of soluble proteins that are secreted into the extracellular space, and given that the majority of cellular transmembrane proteins (excluding those of the mitochondria) use this pathway to arrive at their ultimate destination, we believe that the question posed by the reviewer is highly challenging and beyond the scope of our research. We will add this to the discussion:

“...the abundance of mPTP associated proteins suggesting a role of this pore in ceramide induced insulin resistance (Sup. Fig. 6E). In addition, it is yet to be determined whether the trafficking defect is specific to Glut4 or if it affects the exocytic-secretory pathway more broadly…”

Points from reviewer 2:

1. The mechanistic aspect of the work and conclusions put forth rely heavily on studies performed in cultured myocytes, which are highly glycolytic and generally viewed as a poor model for studying muscle metabolism and insulin action. Nonetheless, the findings provide a strong rationale for moving this line of investigation into mouse gain/loss of function models.

Response: The relative contribution of the anaerobic (glycolysis) and aerobic (mitochondria) contribution to the muscle metabolism can change in L6 depending on differentiation stage. For instance, Serrage et al (PMID30701682) demonstrated that L6-myotubes have a higher mitochondrial abundance and aerobic metabolism than L6-myoblasts. Others have used elegant transcriptomic analysis and metabolic characterisation comparing different skeletal muscle models for studying insulin sensitivity. For instance, Abdelmoez et al in 2020 (PMID31825657) reported that L6 myotubes exhibit greater insulin-stimulated glucose uptake and oxidative capacity compared with C2C12 and Human Mesenchymal Stem Cells (HMSC). Overall, L6 cells exhibit higher metabolic rates and primarily rely on aerobic metabolism, while C2C12 and HSMC cells rely on anaerobic glycolysis. It is worth noting that L6 myotubes are the cell line most closely related to adult human muscle when compared with other muscle cell lines (PMID31825657). Our presented results in Figure 6 H and I provide evidence for the similarities between L6 cells overexpressing SMPD5 and human muscle biopsies. Additionally, in Figure 3J-K, we demonstrate the reproducibility of the negative correlation between ceramide and Coenzyme Q observed in L6 cells in vivo, specifically in the skeletal muscle of mice in chow diet. Furthermore, we have supplemented these findings by demonstrating a comparable relationship in mice exposed to a high-fat diet, as shown in Supplementary Figure 4 H-I (refer to point 4). We will clarify these points in the Discussion:

“In this study, we mainly utilised L6-myotubes, which share many important characteristics with primary muscle fibres relevant to our research. Both types of cells exhibit high sensitivity to insulin and respond similarly to maximal doses of insulin, with Glut4 translocation stimulated between 2 to 4 times over basal levels in response to 100 nM insulin (as shown in Fig. 1-4 and (46,47)). Additionally, mitochondrial respiration in L6-myotubes have a similar sensitivity to mitochondrial poisons, as observed in primary muscle fibres (as shown in Fig. 5 (48)). Finally, inhibiting ceramide production increases CoQ levels in both L6-myotubes and adult muscle tissue (as shown in Fig. 2-3). Therefore, L6-myotubes possess the necessary metabolic features to investigate the role of mitochondria in insulin resistance, and this relationship is likely applicable to primary muscle fibres”.

We will also add additional data - in point 2 - from differentiated human myocytes that are consistent with our observations from the L6 models. Additional experiments are in progress to further extend these findings.

1. One caveat of the approach taken is that exposure of cells to palmitate alone is not reflective of in vivo physiology. It would be interesting to know if similar effects on CoQ are observed when cells are exposed to a more physiological mixture of fatty acids that includes a high ratio of palmitate, but better mimics in vivo nutrition.

Response: Palmitate is widely recognized as a trigger for insulin resistance and ceramide accumulation, which mimics the insulin resistance induced by a diet in rodents and humans. Previous studies have compared the effects of a lipid mixture versus palmitate on inducing insulin resistance in skeletal muscle, and have found that the strong disruption in insulin sensitivity caused by palmitate exposure was lessened with physiologic mixtures of fatty acids, even with a high proportion of saturated fatty acids. This was associated, in part, to the selective partitioning of fatty acids into neutral lipids (such as TAG) when muscle cells are exposed to physiologic lipid mixtures (Newsom et al PMID25793412). Hence, we think that using palmitate is a better strategy to study lipid-induced insulin resistance in vitro. We will add to results:

“In vitro, palmitate conjugated with BSA is the preferred strategy for inducing insulin resistance, as lipid mixtures tend to partition into triacylglycerides (33)”.

We are also performing additional in vivo experiments to add to the physiological relevance of the findings.

1. While the utility of targeting SMPD5 to the mitochondria is appreciated, the results in Figure 5 suggest that this manoeuvre caused a rather severe form of mitochondrial dysfunction. This could be more representative of toxicity rather than pathophysiology. It would be helpful to know if these same effects are observed with other manipulations that lower CoQ to a similar degree. If not, the discrepancies should be discussed.

Response: We conducted a staining procedure using the mitochondrial marker mitoDsRED to observe the effect of SMPD5 overexpression on cell toxicity. The resulting images, displayed in the figure below (Author response image 1), demonstrate that the overexpression of SMPD5 did not result in any significant changes in cell morphology or impact the differentiation potential of our myoblasts into myotubes.

Author response image 1.

In addition, we evaluated cell viability in HeLa cells following exposure to SACLAC (2 uM) to induce CoQ depletion (left panel). Specifically, we measured cell death by monitoring the uptake of Propidium iodide (PI) as shown in the right panel. Our results demonstrated that Saclac-induced CoQ depletion did not lead to cell death at the doses used for CoQ depletion (Author response image 2).

Author response image 2.

Therefore, we deemed it improbable that the observed effect is caused by cellular toxicity, but rather represents a pathological condition induced by elevated levels of ceramides. We will add to discussion:

“...downregulation of the respirasome induced by ceramides may lead to CoQ depletion. Despite the significant impact of ceramide on mitochondrial respiration, we did not observe any indications of cell damage in any of the treatments, suggesting that our models are not explained by toxic/cell death events.”

1. The conclusions could be strengthened by more extensive studies in mice to assess the interplay between mitochondrial ceramides, CoQ depletion and ETC/mitochondrial dysfunction in the context of a standard diet versus HF diet-induced insulin resistance. Does P053 affect mitochondrial ceramide, ETC protein abundance, mitochondrial function, and muscle insulin sensitivity in the predicted directions?

Response: We would like to note that the metabolic characterization and assessment of ETC/mitochondrial function in these mice (both fed a high-fat (HF) and chow diet, with or without P053) were previously published (Turner N, PMID30131496). In addition to this, we have conducted targeted metabolomic and lipidomic analyses to investigate the impact of P053 on ceramide and CoQ levels in HF-fed mice. As illustrated in the figures below (Author response image 3), the administration of P053 led to a reduction in ceramide levels (left panel) and an increase in CoQ levels (right panel) in HF-fed mice, which is consistent with our in vitro findings.

Author response image 3.

We will add to results:

“…similar effect was observed in mice exposed to a high fat diet for 5 wks (Supp. Fig. 4H-I further phenotypic and metabolic characterization of these animals can be found in (41))”

We will further perform more in-vivo studies to corroborate these findings.

Author Response

Reviewer #1 (Public Review):

We thank the Reviewer for their comments.

Reviewer #2 (Public Review):

1) In Figure 4, the authors injected a retrograde tracer in the NA and an anterograde tracer in DCN to find potential "nodes" of overlap. From this experiment, the authors identify the VTA and regions of the thalamus as potential areas of tracer overlap, but it is unclear how many other brain regions were examined. Did the authors jump straight to likely locations of overlap based on previous findings, or were large swaths of the brain examined systematically? If other brain regions were examined, which regions and how was this done? A table listing which brain regions were examined and the presence/intensity of ctb-Alexa568 and GFP fluorescence would be helpful.

We thank the Reviewer for their comments. Exhaustive characterizations of inputs into nucleus accumbens (NAc) as well as of direct outputs of the deep cerebellar nuclei (DCN) have appeared elsewhere (e.g, Ma et al., 2020 doi: 10.3389/fnsys.2020.00015; Novello et al., 2022 doi: 10.1007/s12311-022-01499-w). Our anatomical investigations with retrograde and anterograde tracers were focused on putative intermediary nodal regions with robust inputs from the DCN, clear outputs to NAc, and limbic functionality. Only a handful of brain regions fulfill these criteria, and from those, we chose to target the VTA and intralaminar thalamus based on the observation that cerebellar activation induces dopamine release in the NAc medial shell and core (Holloway et al., 2019 doi: 10.1007/s12311-019-01074-w; Low et al., 2021 10.1038/s41586-021-04143-5) and on our previous work on DCN projections to the midline thalamus (Jung et al., 2022 doi: 10.3389/fnsys.2022.879634).

2) In Figure 5, the authors inject AAV1-Cre in DCN and AAV-FLEX-tdTomato in VTA or thalamus. This is an interesting experiment, but controls are missing. An important control is to inject AAV-FLEX-tdTomato in the VTA or thalamus in the absence of AAV1-Cre injection in DCN. Cre-independent expression of tdTomato should be assessed in the VTA/thalamus and the NA.

We thank the reviewer for bringing up this important control. We routinely perform this control experiment to test for any “leakiness” of floxed vectors prior to use but we did not include it in the paper. In response to the Reviewer’s comment, we show results from this control below. Briefly, we performed a large injection (300 nl) of AAV-FLEX-tdTomato in the thalamus area together with AAV-EGFP (to visualize the injection). No Cre-expressing virus was injected anywhere in the brain. PFA-fixed brain slices were obtained at 3 weeks post-injection and imaged for EGFP and tdTomato. Author Response Figure 1 shows images of the injected thalamus area. No tdTomato expression (Fig. 1C, red) was observed despite abundant EGFP expression (Fig. 1B, green), which confirms that in the absence of Cre the floxed construct does not “leak”.

Author response image 1.

(related to Fig. 5 of manuscript). Control experiment for “leakiness” of floxed tdTomato. A, Epifluorescence image of thalamus region in brain slice with EGFP (green) and tdTomato (red) channels merged. Gain settings in the red channel were increased intentionally, to ensure detection of any “leaky” cells. B, Cellular EGFP expression marks successful viral injection. C, No cellular expression of tdTomato without Cre. Note diffuse red signal from background fluorescence.

Reviewer #3 (Public Review):

1) The novelty of this paper lies in the mapping of projections from the interposed and the lateral nuclei of the cerebellum, as the authors themselves mention. However, in some of the experiments the medial nucleus is also clearly injected (Fig. 4B and 6B). In those experiments, it is impossible to distinguish which nucleus these projections come from, and they could be the ones from the medial nucleus that were previously described (see above).

We thank the Reviewer for their comments. As stated in the Results and in the legend of Fig. 4, in addition to experiments with injections in all DCN (Fig. 4B-D), we also performed experiments with injections in only the lateral nucleus (Fig. 4E and F). The results of these experiments support the existence of lateral DCN projections that overlap with NAc-projecting neurons in VTA and thalamus. This finding is further corroborated by our transsynaptic experiments with lateral DCN-only injections (Fig. 5E,F). Regarding the optophysiological experiments, as mentioned in the Results, all DCN were injected (Fig. 6B) in order to maximize ChR2 expression and the chances of successful stimulation of projections.

2) A strength of the paper is the use of both electrical and optogenetic stimulation. However, the responses to the two in the NAc are very different - electrical stimulation results in both excitation and inhibition, whereas opto stimulation mostly results in only excitation.

We thank the Reviewer for this comment. At least two different explanations could account for the observed differences in the prevalence of inhibitory responses elicited by optogenetic vs electrical stimulation: i) inhibitory response prevalence is sensitive to stimulation intensity (Table 1 and Fig. 2B). Because of inherent differences between optogenetic and electrical stimulation, it is not possible to directly compare intensities between the two modalities in order to determine at which intensities, if at all, the prevalence of responses should match. The observation that, at least in the VTA, the prevalence of inhibitory responses elicited by 1 mW light intensity (the lowest intensity that we tested) was comparable to the prevalence of inhibitory responses elicited by 100 µA electrical stimulation is in line with this explanation; ii) DCN electrical stimulation is not node-specific. To our knowledge, there is currently no evidence to support mesoscale topographic organization in lateral and interposed DCN that is node-specific. Consequently, electrical stimulation of DCN could, in principle, result in NAc responses through various polysynaptic pathways and/or nodes. This possibility would still exist even if electrical stimulation had limited spread of a few hundred microns (as in our experiments) and is at least partly supported by the observed long latencies of electrically-evoked inhibitory responses (NAcCore: 268 ± 25 ms (10th percentile: 42 ms), NAcMed: 259 ± 14 ms (10th percentile: 60 ms). Our recognition of this intrinsic limitation of DCN electrical stimulation was the impetus behind the node-specific optogenetic experiments.

3) The stimulation frequency at which the electrical stimulation in Fig 1 is done to identify responses in the NAc is 200 Hz for 25 ms. Is this physiological? In addition, responses in the NAc are measured for 500 ms after, which is a very long response time.

Regarding stimulation frequency, DCN neurons readily reach firing rates between 100-200 Hz in vivo and ex vivo (e.g., Beekhof et al., 2021 doi.org/10.3390/cells10102686; Sarnaik & Raman, 2018 doi:10.7554/eLife.29546; Canto et al., 2016 doi:10.1371/journal.pone.0165887). Regarding the duration of the response window, at the outset of our experiments we were agnostic to the type of responses that our stimulation protocols would evoke in NAc. We therefore established a response time window that would allow us to capture both fast neurotransmitter-mediated responses as well as neuromodulatory (e.g., dopaminergic) responses, which are known to occur at hundred-millisecond latencies or longer (Wang et al., 2017 doi.org/10.1016/j.celrep.2017.02.062; Chuhma et al., 2014 doi:10.1016/j.neuron.2013.12.027; Gonon, 1997). A posteriori analysis indicated that even if we reduced the response time window by 50%, the ratio of DCN-evoked excitatory vs inhibitory responses in NAc would not change substantially (E/I500: 4.3 vs E/I250: 5).

4) Previous studies have described how different cell types within the DCN have different downstream projections (Fujita et al. 2020). However, the experiments here bundle together all this known heterogeneity.

We agree with the Reviewer that dissecting the contributions of specific DCN cell types to NAc circuits is an important next step, which we are excited to undertake in future studies. Here we have broken new ground by identifying for the first time nodes and functional properties of DCN-NAc connectivity. Performing these studies with DCN cell type-specificity was not justified or feasible, given that the molecular identity of participating DCN neurons is currently unknown.

5) Previous studies have also highlighted the importance of different cell types within the NAc and how input streams are differentially targeted to them. Here, that heterogeneity is also obscured.

Along the same lines as #4, we agree with the Reviewer about the importance of examining the cellular identity of NAc neurons that participate in DCN-NAc circuitry. We are excited to undertake these examinations using ex vivo approaches, which are well suited to dissect cellular and molecular mechanisms.

6) In Fig. 4C, E and F, the experiments on overlap between anterograde and retrograde tracers are not particularly convincing - it's hard to see the overlap.

We thank the reviewer for this comment and have included revised figure panels 4C5, E3, Author response image 1 and Figure 2 below. Single optical sections with altered color scheme and orthogonal confocal views are presented in order to show the overlap between DCN projections and NAc-projecting nodal neurons more clearly. The findings of these imaging experiments are corroborated by the results of our transsynaptic labeling approach (Fig. 5), which we have validated elsewhere (Jung et al., 2022 doi:10.3389/fnsys.2022.879634; and Author response image 1).

Author response image 2.

(related to Fig. 4 of manuscript). Co-localization of NAc-projecting neurons with DCN axonal projections in VTA and thalamus. A-D, Single optical sections and orthogonal views are shown. Green: EGFP-expressing DCN axons; white: ctb- Alexa 568; red: anti-TH (A-B; VTA) or NeuN (C-D; thalamus). White arrows in orthogonal views point to regions of overlap.

Author Response:

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

Reviewer #1 (Public Review):

[...] The experiments are well-designed and carefully conducted. The conclusions of this work are in general well supported by the data. There are a couple of points that need to be addressed or tested.

1) It is unclear how LC phasic stimulation used in this study gates cortical plasticity without altering cellular responses (at least at the calcium imaging level). As the authors mentioned that Polack et al 2013 showed a significant effect of NE blockers in membrane potential and firing rate in V1 layer2/3 neurons during locomotion, it would be useful to test the effect of LC silencing (coupled to mismatch training) on both cellular response and cortical plasticity or applying NE antagonists in V1 in addition to LC optical stimulation. The latter experiment will also address which neuromodulator mediates plasticity, given that LC could co-release other modulators such as dopamine (Takeuchi et al. 2016 and Kempadoo et al. 2016). LC silencing experiment would establish a causal effect more convincingly than the activation experiment.

Regarding the question of how phasic stimulation could alter plasticity without affecting the response sizes or activity in general, we believe there are possibilities supported by previous literature. It has been shown that catecholamines can gate plasticity by acting on eligibility traces at synapses (He et al., 2015; Hong et al., 2022). In addition, all catecholamine receptors are metabotropic and influence intracellular signaling cascades, e.g., via adenylyl cyclase and phospholipases. Catecholamines can gate LTP and LTD via these signaling pathways in vitro (Seol et al., 2007). Both of these influences on plasticity at the molecular level do not necessitate or predict an effect on calcium activity levels. We have now expanded on this in the discussion of the revised manuscript.

While a loss of function experiment could add additional corroborating evidence that LC output is required for the plasticity seen, we did not perform loss-of-function experiments for three reasons:

1. The effects of artificial activity changes around physiological set point are likely not linear for increases and decreases. The problem with a loss of function experiment here is that neuromodulators like noradrenaline affect general aspects of neuronal function. This is apparent in Polack et al., 2013: during the pharmacological blocking experiment, the membrane hyperpolarizes, membrane variance becomes very low, and the cells are effectively silenced (Figure 7 of (Polack et al., 2013)), demonstrating an immediate impact on neuronal function when noradrenaline receptor activation is presumably taken below physiological/waking levels. In light of this, if we reduce LC output/noradrenergic receptor activation and find that plasticity is prevented, this could be the result of a direct influence on the plasticity process, or, the result of a disruption of another aspect of neuronal function, like synaptic transmission or spiking. We would therefore challenge the reviewer’s statement that a loss-of-function experiment would establish a causal effect more convincingly than the gain- of-function experiment that we performed.

2. The loss-of-function experiment is technically more difficult both in implementation and interpretation. Control mice show no sign of plasticity in locomotion modulation index (LMI) on the 10-minute timescale (Figure 4J), thus we would not expect to see any effect when blocking plasticity in this experiment. We would need to use dark-rearing and coupled-training of mice in the VR across development to elicit the relevant plasticity ((Attinger et al., 2017); manuscript Figure 5). We would then need to silence LC activity across days of VR experience to prevent the expected physiological levels of plasticity. Applying NE antagonists in V1 over the entire period of development seems very difficult. This would leave optogenetically silencing axons locally, which in addition to the problems of doing this acutely (Mahn et al., 2016; Raimondo et al., 2012), has not been demonstrated to work chronically over the duration of weeks. Thus, a negative result in this experiment will be difficult to interpret, and likely uninformative: We will not be able to distinguish whether the experimental approach did not work, or whether local LC silencing does nothing to plasticity.

Note that pharmacologically blocking noradrenaline receptors during LC stimulation in the plasticity experiment is also particularly challenging: they would need to be blocked throughout the entire 15 minute duration of the experiment with no changes in concentration of antagonist between the ‘before’ and ‘after’ phases, since the block itself is likely to affect the response size, as seen in Polack et al., 2013, creating a confound for plasticity-related changes in response size. Thus, we make no claim about which particular neuromodulator released by the LC is causing the plasticity.

1. There are several loss-of-function experiments reported in the literature using different developmental plasticity paradigms alongside pharmacological or genetic knockout approaches. These experiments show that chronic suppression of noradrenergic receptor activity prevents ocular dominance plasticity and auditory plasticity (Kasamatsu and Pettigrew, 1976; Shepard et al., 2015). Almost absent from the literature, however, are convincing gain-of-function plasticity experiments.

Overall, we feel that loss-of-function experiments may be a possible direction for future work but, given the technical difficulty and – in our opinion – limited benefit that these experiments, would provide in light of the evidence already provided for the claims we make, we have chosen not to perform these experiments at this time. Note that we already discuss some of the problems with loss-of-function experiments in the discussion.

2) The cortical responses to NE often exhibit an inverted U-curve, with higher or lower doses of NE showing more inhibitory effects. It is unclear how responses induced by optical LC stimulation compare or interact with the physiological activation of the LC during the mismatch. Since the authors only used one frequency stimulation pattern, some discussion or additional tests with a frequency range would be helpful.

This is correct, we do not know how the artificial activation of LC axons relates to physiological activation, e.g. under mismatch. The stimulation strength is intrinsically consistent in our study in the sense that the stimulation level to test for changes in neuronal activity is similar to that used to probe for plasticity effects. We suspect that the artificial activation results in much stronger LC activity than seen during mismatch responses, given that no sign of the plasticity in LMI seen in high ChrimsonR occurs in low ChrimsonR or control mice (Figure 4J). Note, that our conclusions do not rely on the assumption that the stimulation is matched to physiological levels of activation during the visuomotor mismatches that we assayed. The hypothesis that we put forward is that increasing levels of activation of the LC (reflecting increasing rates or amplitude of prediction errors across the brain) will result in increased levels of plasticity. We know that LC axons can reach levels of activity far higher than that seen during visuomotor mismatches, for instance during air puff responses, which constitute a form of positive prediction error (unexpected tactile input) (Figures 2C and S1C). The visuomotor mismatches used in this study were only used to demonstrate that LC activity is consistent with prediction error signaling. We have now expanded on these points in the discussion as suggested.

Reviewer #1 (Recommendations For The Authors):

1) In Figure 3E, there is a rebound response of ChrimsonR at the offset of the mismatch. Is that common? If so, what does it mean? If not, maybe replace it with a more common example trace.

This trace in fact represents the population average, so this offset response (or ‘rebound’) reflects a significant component of the population response to visual flow onset (i.e., mismatch offset), only under conditions of LC stimulation. See our response to reviewer 2 concerning this element of the response.

2) It would be helpful to have some discussions on how a mismatch signal reaches and activates LC from cortical neurons.

We have now added a short segment on this to the discussion.

Reviewer #2 (Public Review):

[...] The study provides very compelling data on a timely and fascinating topic in neuroscience. The authors carefully designed experiments and corresponding controls to exclude any confounding factors in the interpretation of neuronal activity in LC axons and cortical neurons. The quality of the data and the rigor of the analysis are important strengths of the study. I believe this study will have an important contribution to the field of system neuroscience by shedding new light on the role of a key neuromodulator. The results provide strong support for the claims of the study. However, I also believe that some results could have been strengthened by providing additional analyses and experimental controls. These points are discussed below.

Calcium signals in LC axons tend to respond with pupil dilation, air puffs, and locomotion as the authors reported. A more quantitative analysis such as a GLM model could help understand the relative contribution (and temporal relationship) of these variables in explaining calcium signals. This could also help compare signals obtained in the sensory and motor cortical domains. Indeed, the comparison in Figure 2 seems a bit incomplete since only "posterior versus anterior" comparisons have been performed and not within-group comparisons. I believe it is hard to properly assess differences or similarities between calcium signal amplitude measured in different mice and cranial windows as they are subject to important variability (caused by different levels of viral expression for instance). The authors should at the very least provide a full statistical comparison between/within groups through a GLM model that would provide a more systematic quantification.

To provide a more detailed comparison of responses, we have expanded on the analysis in Figure 2 to include comparative heatmaps from anterior and posterior imaging sites, as well as statistical comparisons of the response curves as a function of time. This shows how similar the responses are in the two regions.

Beyond this, we are not sure how a regression analysis (GLM or otherwise) would help support the main point we aim to make here. The responses in anterior and posterior regions are similar, which supports a broadcast model of LC function in the cortex, rather than specialized routing of prediction error signals to cortical areas. Linear contributions of the signals are apparent from the stimulus triggered responses, and while non-linear interactions between the different variables are certainly an interesting question, they go beyond the point we aim to make and would also not be captured by a regression analysis. In addition, we have refined our language replacing descriptors of ‘the same’ or ‘indistinguishable’ between the two regions with ‘similar’, to highlight that while we find no evidence of a difference, our analysis does not cover all possible differences that might appear when looking at non-linear interactions.

Previous studies using stimulations of the locus coeruleus or local iontophoresis of norepinephrine in sensory cortices have shown robust responses modulations (see McBurney-Lin et al., 2019, https://doi.org/10.1016/j.neubiorev.2019.06.009 for a review). The weak modulations observed in this study seem at odds with these reports. Given that the density of ChrimsonR-expressing axons varies across mice and that there are no direct measurements of their activation (besides pupil dilation), it is difficult to appreciate how they impact the local network. How does the density of ChrimsonR-expressing axons compare to the actual density of LC axons in V1? The authors could further discuss this point.

In terms of estimating the percentage of cortical axons labelled based on our axon density measurements: we refer to cortical LC axonal immunostaining in the literature to make this comparison.

In motor cortex, an average axon density of 0.07 µm/µm2 has been reported (Yin et al., 2021), and 0.09 µm/µm2 in prefrontal cortex (Sakakibara et al., 2021). Density of LC axons varies by cortical area, with higher density in motor cortex and medial areas than sensory areas (Agster et al., 2013): V1 axon density is roughly 70% of that in cingulate cortex (adjacent to motor and prefrontal cortices) (Nomura et al., 2014). So, we approximate a maximum average axon density in V1 of approximately 0.056 µm/µm2.

Because these published measurements were made from images taken of tissue volumes with larger z-depth (~ 10 µm) than our reported measurements (~ 1 µm), they appear much larger than the ranges reported in our manuscript (0.002 to 0.007 µm/µm2). We repeated the measurements in our data using images of volumes with 10 µm z-depth, and find that the percentage axons labelled in our study in high ChrimsonR-expressing mice ranges between 0.012 to 0.039 µm/µm2. This corresponds to between 20% to 70% of the density we would expect based on previous work. Note that this is a potentially significant underestimate, and therefore should be used as a lower bound: analyses in the literature use images from immunostaining, where the signal to background ratio is very high. In contrast, we did not transcardially perfuse our mice leading to significant background (especially in the pia/L1, where axon density is high - (Agster et al., 2013; Nomura et al., 2014)), and the intensity of the tdTomato is not especially high. We therefore are likely missing some narrow, dim, and superficial fibers in our analysis.

We also can quantify how our variance in axonal labelling affects our results: For the dataset in Figure 3, there doesn’t appear to be any correlation between the level of expression and the effect of stimulating the axons on the mismatch or visual flow responses for each animal (Author response image 1), while there is a significant correlation between the level of expression and the pupil dilation, consistent with the dataset shown in Figure 4. Thus, even in the most highly expressing mice, there is no clear effect on average response size at the level of the population. We have added these correlations to the revised manuscript as a new Figure S3.

**Author response image 1. **

Correlations between axon density and average effect of laser stimulation on stimulus responses and pupil dilation (data from manuscript Figure 3). Grey points show control mice, blue points show low ChrimsonR-expressing mice, and purple points show high ChrimsonR- expressing mice.

To our knowledge, there has not yet been any similar experiment reported utilizing local LC axonal optogenetic stimulation while recording cortical responses, so when comparing our results to those in the literature, there are several important methodological differences to keep in mind. The vast majority of the work demonstrating an effect of LC output/noradrenaline on responses in the cortex has been done using unit recordings, and while results are mixed, these have most often demonstrated a suppressive effect on spontaneous and/or evoked activity in the cortex (McBurney-Lin et al., 2019). In contrast to these studies, we do not see a major effect of LC stimulation either on baseline or evoked calcium activity (Figure 3), and, if anything, we see a minor potentiation of transient visual flow onset responses (see also Author response image 2). There could be several reasons why our stimulation does not have the same effect as these older studies:

1. Recording location: Unit recordings are often very biased toward highly active neurons (Margrie et al., 2002) and deeper layers of the cortex, while we are imaging from layer 2/3 – a layer notorious for sparse activity. In one of the few papers to record from superficial layers, it was been demonstrated that deeper layers in V1 are affected differently by LC stimulation methods compared to more superficial ones (Sato et al., 1989), with suppression more common in superficial layers. Thus, some differences between our results and those in the majority of the literature could simply be due to recording depth and the sampling bias of unit recordings.

2. Stimulation method: Most previous studies have manipulated LC output/noradrenaline levels by either iontophoretically applying noradrenergic receptor agonists, or by electrically stimulating the LC. Arguably, even though our optogenetic stimulation is still artificial, it represents a more physiologically relevant activation compared to iontophoresis, since the LC releases a number of neuromodulators including dopamine, and these will be released in a more physiological manner in the spatial domain and in terms of neuromodulator concentration. Electrical stimulation of the LC as used by previous studies differs from our optogenetic method in that LC axons will be stimulated across much wider regions of the brain (affecting both the cortex and many of its inputs), and it is not clear whether the cause of cortical response changes is in cortex or subcortical. In addition, electrical LC stimulation is not cell type specific.

3. Temporal features of stimulation: Few previous studies had the same level of temporal control over manipulating LC output that we had using optogenetics. Given that electrical stimulation generates electrical artifacts, coincident stimulation during the stimulus was not used in previous studies. Instead, the LC is often repeatedly or tonically stimulated, sometimes for many seconds, prior to the stimulus being presented. Iontophoresis also does not have the same temporal specificity and will lead to tonically raised receptor activity over a time course determined by washout times.

4. State specificity: Most previous studies have been performed under anesthesia – which is known to impact noradrenaline levels and LC activity (Müller et al., 2011). Thus, the acute effects of LC stimulation are likely not comparable between anesthesia and in the awake animal.

Due to these differences, it is hard to infer why our results differ compared to other papers. The study with the most similar methodology to ours is (Vazey et al., 2018), which used optogenetic stimulation directly into the mouse LC while recording spiking in deep layers of the somatosensory cortex with extracellular electrodes. Like us, they found that phasic optogenetic stimulation alone did not alter baseline spiking activity (Figure 2F of Vazey et al., 2018), and they found that in layers 5 and 6, short latency transient responses to foot touch were potentiated and recruited by simultaneous LC stimulation. While this finding appears more overt than the small modulations we see, it is qualitatively not so dissimilar from our finding that transient responses appear to be slightly potentiated when visual flow begins (Author response image 2). Differences in the degree of the effect may be due to differences in the layers recorded, the proportion of the LC recruited, or the fact anesthesia was used in Vazey et al., 2018.

Note that we only used one set of stimulation parameters for optogenetic stimulation, and it is always possible that using different parameters would result in different effects. We have now added a discussion on the topic to the revised manuscript.

In the analysis performed in Figure 3, it seems that red light stimulations used to drive ChrimsonR also have an indirect impact on V1 neurons through the retina. Indeed, figure 3D shows a similar response profile for ChrimsonR and control with calcium signals increasing at laser onset (ON response) and offset (OFF response). With that in mind, it is hard to interpret the results shown in Figure 3E-F without seeing the average calcium time course for Control mice. Are the responses following visual flow caused by LC activation or additional visual inputs? The authors should provide additional information to clarify this result.

This is a good point. When we plot the average difference between the stimulus response alone and the optogenetic stimulation + stimulus response, we do indeed find that there is a transient increase in response at the visual flow onset (and the offset of mismatch, which is where visual flow resumes), and this is only seen in ChrimsonR-expressing mice (Author response image 2). We therefore believe that these enhanced transients at visual flow onset could be due to the effect of ChrimsonR stimulation, and indeed previous studies have shown that LC stimulation can reduce the onset latency and latency jitter of afferent-evoked activity (Devilbiss and Waterhouse, 2004; Lecas, 2004), an effect which could mediate the differences we see. We have added this analysis to the revised manuscript in Figure 3 and added discussion accordingly.

**Author response image 2. **

Difference in responses to visual stimuli caused by optogenetic stimulation, calculated by subtracting the average response when no laser was presented from the average response when the laser was presented concurrent with the visual stimulus. Pink traces show the response difference for ChrimsonR-expressing mice, and grey shows the same for control mice. Black blocks below indicate consecutive timepoints after stimulation showing a significant difference between ChrimsonR and control as determined by hierarchical bootstrapping (p<0.05).

Some aspects of the described plasticity process remained unanswered. It is not clear over which time scale the locomotion modulation index changes and how many optogenetic stimulations are necessary or sufficient to saturate this index. Some of these questions could be addressed with the dataset of Figure 3 by measuring this index over different epochs of the imaging session (from early to late) to estimate the dynamics of the ongoing plasticity process (in comparison to control mice). Also, is there any behavioural consequence of plasticity/update of functional representation in V1? If plasticity gated by repeated LC activations reproduced visuomotor responses observed in mice that were exposed to visual stimulation only in the virtual environment, then I would expect to see a change in the locomotion behaviour (such as a change in speed distribution) as a result of the repeated LC stimulation. This would provide more compelling evidence for changes in internal models for visuomotor coupling in relation to its behavioural relevance. An experiment that could confirm the existence of the LC-gated learning process would be to change the gain of the visuomotor coupling and see if mice adapt faster with LC optogenetic activation compared to control mice with no ChrimsonR expression. Authors should discuss how they imagine the behavioural manifestation of this artificially-induced learning process in V1.

Regarding the question of plasticity time course: Unfortunately, owing to the paradigm used in Figure 3, the time course of the plasticity will not be quantifiable from this experiment. This is because in the first 10 minutes, the mouse is in closed loop visuomotor VR experience, undergoing optogenetic stimulation (this is the time period in which we record mismatches). We then shift to the open loop session to quantify the effect of optogenetic stimulation on visual flow responses. Since the plasticity is presumably happening during the closed loop phase, and we have no read-out of the plasticity during this phase (we do not have uncoupled visual flow onsets to quantify LMI in closed loop), it is not possible to track the plasticity over time.

Regarding the behavioral relevance of the plasticity: The type of plasticity we describe here is consistent with predictive, visuomotor plasticity in the form of a learned suppression of responses to self-generated visual feedback during movement. Intuitive purposes of this type of plasticity would be 1) to enable better detection of external moving objects by suppressing the predictable (and therefore redundant) self-generated visual motion and 2) to better detect changes in the geometry of the world (near objects have a larger visuomotor gain that far objects). In our paradigm, we have no intuitive read-out of the mouse’s perception of these things, and it is not clear to us that they would be reflected in locomotion speed, which does not differ between groups (manuscript Figure S5). Instead, we would need to turn to other paradigms for a clear behavioral read-out of predictive forms of sensorimotor learning: for instance, sensorimotor learning paradigms in the VR (such as those used in (Heindorf et al., 2018; Leinweber et al., 2017)), or novel paradigms that reinforce the mouse for detecting changes in the gain of the VR, or moving objects in the VR, using LC stimulation during the learning phase to assess if this improves acquisition. This is certainly a direction for future work. In the case of a positive effect, however, the link between the precise form of plasticity we quantify in this manuscript and the effect on the behavior would remain indirect, so we see this as beyond the scope of the manuscript. We have added a discussion on this topic to the revised manuscript.

Finally, control mice used as a comparison to mice expressing ChrimsonR in Figure 3 were not injected with a control viral vector expressing a fluorescent protein alone. Although it is unlikely that the procedure of injection could cause the results observed, it would have been a better control for the interpretation of the results.

We agree that this indeed would have been a better control. However, we believe that this is fortunately not a major problem for the interpretation of our results for two reasons:

1. The control and ChrimsonR expressing mice do not show major differences in the effect of optogenetic LC stimulation at the level of the calcium responses for all results in Figure 3, with the exception of the locomotion modulation indices (Figure 3I). Therefore, in terms of response size, there is no major effect compared to control animals that could be caused by the injection procedure, apart from marginally increased transient responses to visual flow onset – and, as the reviewer notes, it is difficult to see how the injection procedure would cause this effect.

2. The effect on locomotion modulation index (Figure 3I) was replicated with another set of mice in Figure 4C, for which we did have a form of injected control (‘Low ChrimsonR’), which did not show the same plasticity in locomotion modulation index (Figure 4E). We therefore know that at least the injection itself is not resulting in the plasticity effect seen.

Reviewer #2 (Recommendations For The Authors):

In experiments where axonal imaging was performed on LC axons, the authors should indicate the number of mice used in addition to the number of Field of View (FoV). Indeed, samples (FoVs) are not guaranteed to be independent as LC axons can span large cortical areas and the same axon can end up in different FoVs. Please provide statistics across mice/cranial windows to confirm the robustness of the results.

All information requested regarding animal numbers in axonal imaging are provided in the statistical Table S1, as well as in the text and figures (e.g., Figure 2A). Samples will be independent in time (as different FoVs were imaged on different days), but it is indeed possible that axon segments from different FoVs within an animal come from the same axon.

Averaging across animals greatly reduces statistical power. We have therefore implemented hierarchical bootstrapping instead: bootstrapping first occurs at the level of animal and then at the level of FoV. All p-values that were reported as significant in manuscript remained significant with this test, with no major reduction in significance level, with the exception of Figure S2B, where statistical significance was lost (p = 0.04 with Rank sum, p = 0.07 with hierarchical Bootstrapping). We therefore conclude that sampling from the same animals across days is not responsible for the significance of results reported.

References

Agster, K.L., Mejias-Aponte, C.A., Clark, B.D., Waterhouse, B.D., 2013. Evidence for a regional specificityi n the density and distribution of noradrenergic varicosities in rat cortex. Journal of Comparative Neurology 521, 2195–2207. https://doi.org/10.1002/cne.23270

Attinger, A., Wang, B., Keller, G.B., 2017. Visuomotor Coupling Shapes the Functional Development of Mouse Visual Cortex. Cell 169, 1291-1302.e14. https://doi.org/10.1016/j.cell.2017.05.023

Devilbiss, D.M., Waterhouse, B.D., 2004. The Effects of Tonic Locus Ceruleus Output on Sensory-Evoked Responses of Ventral Posterior Medial Thalamic and Barrel Field Cortical Neurons in the Awake Rat. J. Neurosci. 24, 10773–10785. https://doi.org/10.1523/JNEUROSCI.1573-04.2004

He, K., Huertas, M., Hong, S.Z., Tie, X., Hell, J.W., Shouval, H., Kirkwood, A., 2015. Distinct Eligibility Traces for LTP and LTD in Cortical Synapses. Neuron 88, 528–538. https://doi.org/10.1016/j.neuron.2015.09.037

Heindorf, M., Arber, S., Keller, G.B., 2018. Mouse Motor Cortex Coordinates the Behavioral Response to Unpredicted Sensory Feedback. Neuron 0. https://doi.org/10.1016/j.neuron.2018.07.046

Hong, S.Z., Mesik, L., Grossman, C.D., Cohen, J.Y., Lee, B., Severin, D., Lee, H.-K., Hell, J.W., Kirkwood, A., 2022. Norepinephrine potentiates and serotonin depresses visual cortical responses by transforming eligibility traces. Nat Commun 13, 3202. https://doi.org/10.1038/s41467-022-30827-1

Kasamatsu, T., Pettigrew, J.D., 1976. Depletion of brain catecholamines: failure of ocular dominance shift after monocular occlusion in kittens. Science 194, 206–209. https://doi.org/10.1126/science.959850

Lecas, J.-C., 2004. Locus coeruleus activation shortens synaptic drive while decreasing spike latency and jitter in sensorimotor cortex. Implications for neuronal integration. European Journal of Neuroscience 19, 2519–2530. https://doi.org/10.1111/j.0953-816X.2004.03341.x

Leinweber, M., Ward, D.R., Sobczak, J.M., Attinger, A., Keller, G.B., 2017. A Sensorimotor Circuit in Mouse Cortex for Visual Flow Predictions. Neuron 95, 1420-1432.e5. https://doi.org/10.1016/j.neuron.2017.08.036

Mahn, M., Prigge, M., Ron, S., Levy, R., Yizhar, O., 2016. Biophysical constraints of optogenetic inhibition at presynaptic terminals. Nat Neurosci 19, 554–556. https://doi.org/10.1038/nn.4266

Margrie, T.W., Brecht, M., Sakmann, B., 2002. In vivo, low-resistance, whole-cell recordings from neurons in the anaesthetized and awake mammalian brain. Pflugers Arch. 444, 491–498. https://doi.org/10.1007/s00424-002-0831-z

McBurney-Lin, J., Lu, J., Zuo, Y., Yang, H., 2019. Locus coeruleus-norepinephrine modulation of sensory processing and perception: A focused review. Neurosci Biobehav Rev 105, 190–199. https://doi.org/10.1016/j.neubiorev.2019.06.009

Müller, C.P., Pum, M.E., Amato, D., Schüttler, J., Huston, J.P., De Souza Silva, M.A., 2011. The in vivo neurochemistry of the brain during general anesthesia. Journal of Neurochemistry 119, 419–446. https://doi.org/10.1111/j.1471-4159.2011.07445.x

Nomura, S., Bouhadana, M., Morel, C., Faure, P., Cauli, B., Lambolez, B., Hepp, R., 2014. Noradrenalin and dopamine receptors both control cAMP-PKA signaling throughout the cerebral cortex. Front Cell Neurosci 8. https://doi.org/10.3389/fncel.2014.00247

Polack, P.-O., Friedman, J., Golshani, P., 2013. Cellular mechanisms of brain-state-dependent gain modulation in visual cortex. Nat Neurosci 16, 1331–1339. https://doi.org/10.1038/nn.3464

Raimondo, J.V., Kay, L., Ellender, T.J., Akerman, C.J., 2012. Optogenetic silencing strategies differ in their effects on inhibitory synaptic transmission. Nat Neurosci 15, 1102–1104. https://doi.org/10.1038/nn.3143

Sakakibara, Y., Hirota, Y., Ibaraki, K., Takei, K., Chikamatsu, S., Tsubokawa, Y., Saito, T., Saido, T.C., Sekiya, M., Iijima, K.M., n.d. Widespread Reduced Density of Noradrenergic Locus Coeruleus Axons in the App Knock-In Mouse Model of Amyloid-β Amyloidosis. J Alzheimers Dis 82, 1513–1530. https://doi.org/10.3233/JAD-210385

Sato, H., Fox, K., Daw, N.W., 1989. Effect of electrical stimulation of locus coeruleus on the activity of neurons in the cat visual cortex. Journal of Neurophysiology. https://doi.org/10.1152/jn.1989.62.4.946

Seol, G.H., Ziburkus, J., Huang, S., Song, L., Kim, I.T., Takamiya, K., Huganir, R.L., Lee, H.-K., Kirkwood, A., 2007. Neuromodulators control the polarity of spike-timing-dependent synaptic plasticity. Neuron 55, 919–929. https://doi.org/10.1016/j.neuron.2007.08.013

Shepard, K.N., Liles, L.C., Weinshenker, D., Liu, R.C., 2015. Norepinephrine is necessary for experience-dependent plasticity in the developing mouse auditory cortex. J Neurosci 35, 2432–2437.https://doi.org/10.1523/JNEUROSCI.0532-14.2015

Vazey, E.M., Moorman, D.E., Aston-Jones, G., 2018. Phasic locus coeruleus activity regulates cortical encoding of salience information. Proceedings of the National Academy of Sciences 115, E9439– E9448. https://doi.org/10.1073/pnas.1803716115

Yin, X., Jones, N., Yang, J., Asraoui, N., Mathieu, M.-E., Cai, L., Chen, S.X., 2021. Delayed motor learning in a 16p11.2 deletion mouse model of autism is rescued by locus coeruleus activation. Nat Neurosci 24, 646–657. https://doi.org/10.1038/s41593-021-00815-7

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public review):

Hearing and balance rely on specialized ribbon synapses that transmit sensory stimuli between hair cells and afferent neurons. Synaptic adhesion molecules that form and regulate transsynaptic interactions between inner hair cells (IHCs) and spiral ganglion neurons (SGNs) are crucial for maintaining auditory synaptic integrity and, consequently, for auditory signaling. Synaptic adhesion molecules such as neurexin-3 and neuroligin-1 and -3 have recently been shown to play vital roles in establishing and maintaining these synaptic connections ( doi: 10.1242/dev.202723 and DOI: 10.1016/j.isci.2022.104803). However, the full set of molecules required for synapse assembly remains unclear.

Karagulan et al. highlight the critical role of the synaptic adhesion molecule RTN4RL2 in the development and function of auditory afferent synapses between IHCs and SGNs, particularly regarding how RTN4RL2 may influence synaptic integrity and receptor localization. Their study shows that deletion of RTN4RL2 in mice leads to enlarged presynaptic ribbons and smaller postsynaptic densities (PSDs) in SGNs, indicating that RTN4RL2 is vital for synaptic structure. Additionally, the presence of "orphan" PSDs-those not directly associated with IHCs-in RTN4RL2 knockout mice suggests a developmental defect in which some SGN neurites fail to form appropriate synaptic contacts, highlighting potential issues in synaptic pruning or guidance. The study also observed a depolarized shift in the activation of CaV1.3 calcium channels in IHCs, indicating altered presynaptic functionality that may lead to impaired neurotransmitter release. Furthermore, postsynaptic SGNs exhibited a deficiency in GluA2/3 AMPA receptor subunits, despite normal Gria2 mRNA levels, pointing to a disruption in receptor localization that could compromise synaptic transmission. Auditory brainstem responses showed increased sound thresholds in RTN4RL2 knockout mice, indicating impaired hearing related to these synaptic dysfunctions.

The findings reported here significantly enhance our understanding of synaptic organization in the auditory system, particularly concerning the molecular mechanisms underlying IHC-SGN connectivity. The implications are far-reaching, as they not only inform auditory neuroscience but also provide insights into potential therapeutic targets for hearing loss related to synaptic dysfunction.

We would like to thank the reviewer for appreciating the work and the advice that helped us to further improve the manuscript. We have carefully addressed all concerns, please see our point-per-point response below and the revised manuscript.

Reviewer #2 (Public review):

Summary:

Kargulyan et al. investigate the function of the transsynaptic adhesion molecule RTN4RL2 in the formation and function of ribbon synapses between type I spiral ganglion neurons (SGNs) and inner hair cells. For this purpose, they study constitutive RTN4RL2 knock-out mice. Using immunohistochemistry, they reveal defects in the recruitment of protein to ribbon synapses in the knockouts. Serial block phase EM reveals defects in SGN projections in mutants. Electrophysiological recordings suggest a small but statistically significant depolarized shift in the activation of Cav1.3 Ca²⁺ channels. Auditory thresholds are also elevated in the mutant mice. The authors conclude that RTN4RL2 contributes to the formation and function of auditory afferent synapses to regulate auditory function.

We would like to thank the reviewer for appreciating the work and the advice that helped us to further improve the manuscript. We have carefully addressed all concerns, please see our point-per-point response below and the revised manuscript.

Strengths:

The authors have excellent tools to analyze ribbon synapses.

Weaknesses:

However, there are several concerns that substantially reduce my enthusiasm for the study.

(1) The analysis of the expression pattern of RTN4RL2 in Figure 1 is incomplete. The authors should show a developmental time course of expression up into maturity to correlate gene expression with major developmental milestones such as axon outgrowth, innervation, and refinement. This would allow the development of models supporting roles in axon outgrowth versus innervation or both.

We agree that it would be valuable to show the developmental time course of RTN4RL2 expression. In response to the reviewer’s comment, we are providing RNAscope data from developmental ages E11.5, E12.5 and E16 in Figure 1. RTN4RL2 shows expression at E11.5/E12.5 both in the spiral ganglion and hair cell region, with first onset in the hair cells. We conclude that RTN4RL2 is expressed highest during fiber growth at embryonic stages and is downregulated during postnatal development maintaining low levels of expression during adulthood.

(2) It would be important to improve the RNAscope data. Controls should be provided for Figure 1B to show that no signal is observed in hair cells from knockouts. The authors apparently already have the sections because they analyzed gene expression in SGNs of the knock-outs (Figure 1C).

In Figure 1C gene expression in SGNs was assessed at p40, while the expression in hair cells is provided for p1 animals. Unfortunately, we do not have KO controls for p1 animals. However, as indicated in our manuscript, previously published RNA expression datasets do find RTN4RL2 expression in hair cells. Therefore, we think it is unlikely that our results are unspecific.

(3) It is unclear from the immunolocalization data in Figure 1D if all type I SGNs express RTN4RL2. Quantification would be important to properly document the presence of RTN4RL2 in all or a subset of type I SGNs. If only a subset of SGNs express RTN4RL2, it could significantly affect the interpretation of the data. For example, SGNs selectively projecting to the pillar or modiolar side of hair cells could be affected. These synapses significantly differ in their properties.

According to already published single cell RNAseq dataset from Shrestha et al., 2018, RTN4RL2 expression does not seem to show a clear type I SGN subtype specificity (Author response image 1). In response to the reviewer’s comment, we have further performed anti-Parvalbumin (PV) and anti-calretinin (CR) immunostainings in mid-modiolar cryosections of RTN4RL2^+/+ and RTN4RL2^-/- cochleae. Parvalbumin was chosen to label all SGNs and CALB2 was chosen primarily as a type Ia SGN marker (Sun et al., 2018). We present the data from all analyzed samples below (figure 2 of this rebuttal letter). Cell segmentation masks of PV positive cells were obtained using Cellpose 2.0 and the average CR intensity was calculated in those masks. While the distributions of CR intensity and the ratio of CR and PV intensities are slightly shifted in RTN4RL2^-/- cochleae, we take the data to suggest that the composition of the spiral ganglion by molecular type I SGN subtypes is largely unchanged in RTN4RL2^-/- mice.

Author response image 1.

Author response image 1 cites single cell RNAseq data of Brikha R Shrestha, Chester Chia, Lorna Wu, Sharon G Kujawa, M Charles Liberman, Lisa V Goodrich. Sensory neuron diversity in the inner ear is shaped by activity. Cell. 2018 Aug 23; 174(5):1229-1246.e17. doi: 10.1016/j.cell/2018.07.007

Author response image 2.

Calretinin intensity distribution in spiral ganglion of RTN4RL2^+/+ and RTN4RL2^-/- mice. (A) Mid-modiolar cochlear cryosections from RTN4RL2^+/+ (top) and RTN4RL2^-/- (bottom) mice immunolabeled against Parvalbumin (PV) and Calretinin (CR). Scale bar = 20 mm. (B) Distribution of CR intensity in PV positive cells (N = 3 for each genotype). (C) Distribution of the ratio of CR and PV intensities (N = 3 for each genotype).

(4) It is important to show proper controls for the RTN4RL2 immunolocalization data to show that no staining is observed in knockouts.

Unfortunately, our recent attempts to perform RTN4RL2 immunostainings on cryosections failed and therefore, we decided to remove the RTNr4RL2 immunostainings from Figure 1. We have adjusted the results section accordingly.

(5) The authors state in the discussion that no staining for RTN4RL2 was observed at synaptic sites. This is surprising. Did the authors stain multiple ages? Was there perhaps transient expression during development? Or in axons indicative of a role in outgrowth, not synapse formation?

We thank the reviewer for the comment. We have now tried RTN4RL2 immunostainings on cryosections at several developmental stages, but unfortunately this time did not succeed to obtain reproducible and reliable results. Therefore, we decided to also remove the previous immunostainings from Figure 1. We have adjusted the results section as well as removed our statement of not detecting RTN4RL2 near the synaptic regions from the discussion.

(6) In Figure 2 it seems that images in mutants are brighter compared to wildtypes. Are exposure times equivalent? Is this a consistent result?

Yes, the samples were prepared in parallel, imaged and analyzed in the same manner.

No, we did not observe consistent differences in brightness and also did not find it in the exemplary images of figure 2.

(7) The number of synaptic ribbons for wildtype in Figure 2 is at 10/IHCs, and in Figure 2 Supplementary Figure 2 at 20/IHCs (20 is more like what is normally reported in the literature). The value for mutant similarly drastically varies between the two figures. This is a significant concern, especially because most differences that are reported in synaptic parameters between wild-type and mutants are far below a 2-fold difference.

The key message is that there is no difference in the numbers of ribbons and synapses between the genotypes for the cochlear apex (~10 ribbons/IHCs, Figure 2 and Figure 2-figure supplement 2) and the mid- and base of the cochlea (more ribbons/IHCs, Figure 2-figure supplement 2). Figure 2-figure supplement 3 (now Figure 3) shows that there is a massive reduction of postsynaptic GluA2, while both Figure 2 and Figure 2-figure supplement 2 indicate that the number synapses is normal. These are two different data sets and while we closely collaborated and also shared the Moser lab protocols and analysis routines, we agree that there is a difference in the absolute synapse count, which most likely was an observer difference and different choice of tonotopic positions of analysis. In Figure 2 only the apical hair cells have been analyzed. The Moser lab, since establishing the immunofluorescence-based quantification of synapse number (Khimich et al., 2005) reported tonotopic differences in synapse counts (focus of Meyer et al., 2009 and reported by others: e.g. Kujawa and Liberman, 2009): apical and basal IHCs lower synapse numbers than mid-cochlear IHCs.

(8) The authors report differences in ribbon volume between wild-type and mutant. Was there a difference between the modiolar/pillar region of hair cells? It is known that synaptic size varies across the modiolar-pillar axis. Maybe smaller synapses are preferentially lost?

We thank the reviewer for the comment. Unfortunately, our already acquired datasets from 3-week-old mice did not allow us to check whether the previously described modiolar-pillar gradient of the ribbon size was collapsed in RTN4RL2^-/- mice due to the not so well-preserved morphology of the inner hair cells in our preparations. However, since the number of the ribbons is not changed in the RTN4RL2 KO mice, we do not think that the increase in the ribbon size is due to the loss of small ribbons. In response to the reviewers comment we have analyzed the modiolar-pillar gradient of the ribbon size in IHCs of middle turn of the cochlea form a newly acquired dataset of 14-week-old mice. We took the fluorescence intensity of Ctbp2 positive puncta as a proxy for the ribbon size. In these older mice we found a preserved modiolar-pillar gradient of the ribbon size (larger ribbons at the modiolar side). We summarized the results in the below Author response image 3.

Author response image 3.

The modiolar-pillar gradient of ribbon size is preserved in RTN4RL2^-/- IHCs. (A) Maximum intensity projections of approximately 2 IHCs stained against Vglut3 and Ctbp2 from 14-week-old RTN4RL2^+/+ (left) and RTN4RL2^-/- (right) mice. Scale bar = 5 mm. (B) Synaptic ribbons on the modiolar side show higher fluorescence intensity than the ones on the pillar side of mid-cochlear IHCs in both RTN4RL2^+/+ (left, N=2) RTN4RL2^-/- (right, N=2) mice. (C) Average fluorescence intensity of modiolar ribbons per IHC is higher than the average fluorescence intensity of pillar ribbons (paired t-test, p < 0.001).

(9) The authors show in Figure 2 - Supplement 3 that GluA2/3 staining is absent in the mutants. Are GluA4 receptors upregulated? Otherwise, synaptic transmission should be abolished, which would be a dramatic phenotype. Antibodies are available to analyze GluA4 expression, the experiment is thus feasible. Did the authors carry out recordings from SGNs?

In response to the reviewer’s comment, we have performed GluA4 stainings in RTN4LR2^-/- mice and did not detect any GluA4 positive signal in the mutants (new Figure 3-figure supplement 1). Unfortunately, our animal breeding license was expired at the time we received the reviews and that is why our results are from 14-week-old animals. To verify that the absence of GluA4 signal is not due to potential PSD loss in 14-week-old RTN4RL2^-/-, we have additionally performed anti-Ctbp2, anti-Homer1 and anti-Vglut3 stainings in 14-week-old animals. Despite the reduced number, we still observed juxtaposing pre- and postsynaptic puncta. We assume that the reviewer asks for patch-clamp recordings from SGNs, which are, as we are confident the reviewer is aware of, technically very challenging and beyond the scope of the present study but an important objective for future studies.  In response to the reviewers comment we have added a statement to the discussion pointing to these patch-clamp recordings from SGNs as important objective for future studies.

(10) The authors use SBEM to analyze SGN projections and synapses. The data suggest that a significant number of SGNs are not connected to IHCs. A reconstruction in Figure 3 shows hair cells and axons. It is not clear how the outline of hair cells was derived, but this should be indicated. Also, is this a defect in the formation of synapses and subsequent retraction of SGN projections? Or could RTN4RL2 mutants have a defect in axonal outgrowth and guidance that secondarily affects synapses? To address this question, it would be useful to sparsely label SGNs in mutants, for example with AAV vectors expression GFP, and to trace the axons during development. This would allow us to distinguish between models of RTN4RL2 function. As it stands, it is not clear that RTN4RL2 acts directly at synapses.

We agree with the reviewer on the value of a developmental study of afferent connectivity but consider this beyond the scope of the present study. In response to the reviewer's comment, we have replaced the IHC outlines with volume-reconstructed IHCs in Figure 3B (now Figure 4B). Moreover, as shown in Figure 3F (now Figure 4F), most if not all type-I SGNs (both with and without ribbon) were unbranched in the mutants just like in wildtype (also shown for a larger sample in Hua et al., 2021), arguing against morphological abnormality during development.

(11) The authors observe a tiny shift in the operation range of Ca²⁺ channels that has no effect on synaptic vesicle exocytosis. It seems very unlikely that this difference can explain the auditory phenotype of the mutant mice.

We assume that the statement refers to the normal exocytosis of mutant IHCs at the potential of maximal Ca²⁺ influx (Figure 3G and H, now Figure 4G and H). We would like to note that this experiment was performed to probe for a deficit of synapse function beyond that of the Ca²⁺ channel activation, but did not address the impact of the altered voltage—dependence of Ca²⁺ channel activation. In response to the reviewer’s comment, we have now added further discussion to more clearly communicate that for the range of receptor potentials achieved near sound threshold we expect impaired IHC exocytosis as the Ca²⁺ channels require slightly more depolarization for activation in the mutant IHCs.

(12) ABR recordings were conducted in whole-body knockouts. Effects on auditory thresholds could be a secondary consequence of perturbation along the auditory pathway. Conditional knockouts or precisely designed rescue experiments would go a long way to support the authors' hypothesis. I realize that this is a big ask and floxed mice might not be available to conduct the study.

Thanks for this helpful comment and, indeed, unfortunately, we do not have conditional KO mice at our disposal. We totally agree that this will be important also for clarifying the role of IHC vs. SGN expression of RTN4RL2. In response to the reviewer’s comment, we now discussed the shortcoming of using constitutive RTN4RL2^-/- mice and added this important experiment on IHC and SGN specific deletion of RTN4RL2 as an objective of future studies.

Reviewer #3 (Public review):

In this study, the authors used RNAscope and immunostaining to confirm the expression of RTN4RL2 RNA and protein in hair cells and spiral ganglia. Through RTN4RL2 gene knockout mice, they demonstrated that the absence of RTN4RL2 leads to an increase in the size of presynaptic ribbons and a depolarized shift in the activation of calcium channels in inner hair cells. Additionally, they observed a reduction in GluA2/3 AMPA receptors in postsynaptic neurons and identified additional "orphan PSDs" not paired with presynaptic ribbons. These synaptic alterations ultimately resulted in an increased hearing threshold in mice, confirming that the RTN4RL2 gene is essential for normal hearing. These data are intriguing as they suggest that RTN4RL2 contributes to the proper formation and function of auditory afferent synapses and is critical for normal hearing. However, a thorough understanding of the known or postulated roles of RTN4Rl2 is lacking.

We would like to thank the reviewer for appreciating the work and the advice that helped us to further improve the manuscript. We have carefully addressed all concerns, please see our point-per-point response below and the revised manuscript.

While the conclusions of this paper are generally well supported by the data, several aspects of the data analysis warrant further clarification and expansion.

(1) A quantitative assessment is necessary in Figure 1 when discussing RNA and protein expression. It would be beneficial to show that expression levels are quantitatively reduced in KO mice compared to wild-type mice. This suggestion also applies to Figure 2-supplement 3.D, which examines expression levels.

The processing of our control and KO samples for RNAscope was not strictly done in parallel and therefore we would like to refrain from quantitative comparison.

(2) In Figure 2, the authors present a morphological analysis of synapses and discuss the presence of "orphan PSDs." I agree that Homer1 not juxtaposed with Ctbp2 is increased in KO mice compared to the control group. However, in quantifying this, they opted to measure the number of Homer1 juxtaposed with Ctbp2 rather than directly quantifying the number of Homer1 not juxtaposed with Ctbp2. Quantifying the number of Homer1 not juxtaposed with Ctbp2 would more clearly represent "orphan PSDs" and provide stronger support for the discussion surrounding their presence.

We appreciate the reviewer’s comment. We did not perform this analysis primarily because “orphan” Homer1 puncta, as seen in our immunostainings, are distributed away from hair cells in diverse morphologies and sizes. This makes distinguishing them from unspecific immunofluorescent spots—also present in wild-type samples—challenging. In response to the reviewer’s request, we analyzed the number of “orphan” Homer1 puncta in our previously acquired RTN4RL2^+/+ and RTN4RL2^-/- samples. Using the surface algorithm in Imaris software, we applied identical parameters across all samples to create surfaces for Homer1-positive puncta (total Homer1 puncta). We quantified “orphan” Homer1 puncta as the difference between total and ribbon-juxtaposing Homer1 puncta and normalized this number to the IHC count. Our results showed 4.3 vs. 26.8 “orphan” Homer1 puncta per IHC in RTN4RL2^+/+ and RTN4RL2^-/- samples, respectively. We note that variations in acquired volumes between samples may introduce confounding effects.

(3) In Figure 2, Supplementary 3, the authors discuss GluA2/3 puncta reduction and note that Gria2 RNA expression remains unchanged. However, there is an issue with the lack of quantification for Gria2 RNA expression. Additionally, it is noted that RNA expression was measured at P4. While the timing for GluA2/3 puncta assessment is not specified, if it was assessed at 3 weeks old as in Figure 2's synaptic puncta analysis, it would be inappropriate to link Gria2 RNA expression with GluA2/3 protein expression at P4. If RNA and protein expression were assessed at P4, please indicate this timing for clarity.

GluA2/3 immunostainings were performed in 1 to 1.5-month-old animals. We apologize for not indicating this before and have now included it in Figure 3 legend. The processing of our control and KO samples for RNAscope was not strictly done in parallel and therefore we would like to refrain from quantitative comparison.

(4) In Figure 3, the authors indicate that RTN4RL2 deficiency reduces the number of type 1 SGNs connected to ribbons. Given that the number of ribbons remains unchanged (Figure 2), it is important to clearly explain the implications of this finding. It is already known that each type I SGN forms a single synaptic contact with a single IHC. The fact that the number of ribbons remains constant while additional "orphan PSDs" are present suggests that the overall number of SGNs might need to increase to account for these findings. An explanation addressing this would be helpful.

In Figure 3 (now Figure 4), we found additional type-1 SGNs that are unconnected to IHC, in good agreement with “orphan PSDs” observed under the light microscope. Indeed, we also confirmed monosynaptic, unbranched fiber morphology (Figure 3F, now Figure 4F). Together, these results imply about a 20% increase in the overall number of SGNs, which however we did not observe in SGN soma counting.

(5) In Figure 4F and 5Cii, could you clarify how voltage sensitivity (k) was calculated? Additionally, please provide an explanation for the values presented in millivolts (mV).

Voltage sensitivity (k) was calculated as the slope of the Boltzmann fit to the fractional activation curves: , Where G is conductance, G_max is the maximum conductance, V_m is the membrane potential, V_half is the voltage corresponding to the half maximal activation of Ca²⁺ channels and k (slope of the curve) is the voltage sensitivity of Ca²⁺ channel activation. We have now added this to our Materials and Methods section.

(6) In Figure 6, the author measured the threshold of ABR at 2-4 months old. Since previous figures confirming synaptic morphology and function were all conducted on 3-week-old mice, it would be better to measure ABR at 3 weeks of age if possible.

ABR measurements for comparisons in a cohort of age-matched mice require fully developed individuals. 3 weeks is the minimum age that is regarded for a mature ear. However, variation in developmental differences among one litter is very frequent that affects normal hearing thresholds. From our own experience we do not regard the ear fully functional before 6 weeks of age. Then hearing thresholds are lowest indicating full functionality. Since the C57BL/6 background strain has a genetic defect in the Cadherin 23-coding gene (Cdh23) at the ahl locus of mouse chromosome 10 these mice exhibit early onset and progression of age-related hearing loss starting at 5–8 months (Hunter & Willott, 1987). Therefore, we chose a “safe” time window for stable and unaffected ABR recordings of 2-4 months to provide most representative data.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

(1) Please include information on the validation of all the antibodies used in this study, or reference the relevant work where the antibodies were previously validated.

In response to the reviewer’s comment, we have now included a table listing all primary antibodies used in this study. Where possible, we provide references for knockout (KO) validation. Otherwise, we refer to the manufacturer’s information, as provided in the respective datasheets.

(2) Figure 2 illustrates the pre- and postsynaptic changes observed in RTN4RL2 knockout (KO) mice. Please specify the age of the mice and the cochlear region depicted and analyzed in Figure 2.

We thank the reviewer for the comment. The IHCs of apical cochlear region were analyzed in mice at 3 weeks of age. We have now added this to the figure legend.

(3) The discovery of orphan SGN neurites in RTN4RL2 KO mice is particularly intriguing. I wonder whether the additional Homer1-positive puncta illustrated in Figure 2 are present in these orphan SGN neurites, which would suggest that they may be functional. Conducting immunohistochemistry (IHC) labeling for type I SGN neurites using an anti-Tuj1 antibody, along with Homer1, would help localize the additional Homer1 puncta shown in Figure 2. Additionally, the "extra" Homer1 puncta appears less striking in the data presented in Figure 2-Supplement 2. Quantifying the number of Homer1 puncta in wild-type versus KO mice across different cochlear regions will help visualize the Figure 2-Supplement 2 data and relate the presence of extra neurites to the increased auditory brainstem response (ABR) thresholds observed at all frequencies.

We thank the reviewer for the comment and we agree that localizing orphan PSDs on the SGN neurites would be very useful. Unfortunately, the animal breeding license in the Göttingen lab had expired. At the time we received the reviews we only had access to 14-week-old animals and could not perform the stainings in animals which would have comparable age range to the rest of the study (3-4 weeks). The phenotype of extra Homer1 puncta was not as drastic in 14-week-old animals as it was in previously stained 3-week-old animals. Nevertheless, we still tried NF200, Homer1 and Vglut3 immunostainings in 14-week-old animals. We present representative single imaging planes of NF200, Homer1 and Vglut3 stainings in Author response image 4. Additionally, we provide exemplary images from 7-week-old RTN4RL2^-/-, where it looks like that the orphan Homer1 puncta are found on calretinin positive neurites.

Author response image 4.

Attempts to localize “orphan” Homer1 patches on type I SGN neurites. (A) Single exemplary imaging planes of apical IHC region from RTN4RL2^+/+ (left) and RTN4RL2^-/- (right) mice immunolabeled against NF200, Vglut3 and Homer1. White arrows show putative “orphan” Homer1 puncta on NF200 positive neurites. Scale bar = 5 mm. (B) Maximum intensity projections of representative confocal stacks of IHCs from RTN4RL2^-/- mice immunolabeled against Calretinin and Homer1. Scale bars = 5 mm. White arrows show possible “orphan” Homer1 puncta on Calretinin positive boutons.

(4) The authors noted a reduction in the number of GluA2/3-positive puncta in RTN4RL2 KOs, as shown in Figure 2-Supplement 3. However, in the Results section (page 5, line 124), it is unclear whether the authors refer to a reduction in fluorescence intensity or the number of puncta. Please clarify this.

We thank the reviewer for the comment. We refer to the number and have now added this to the manuscript.

(5) I find it particularly interesting that, despite the presence of smaller but synaptically engaged Homer1-positive SGN neurites, these appear to lack or present a reduction in the number of GluA2/3 puncta, and that GluA2/3 puncta are observed in non-ribbon juxtaposed neurites. Therefore, I suggest including GluA2/3 (Fig2 supplement 3) data in the main figure. It would be valuable to determine whether the orphan neurites express both Homer1 and GluA2/3, which could indicate that the defect is not solely due to reduced GluA2/3 expression at the formed synapses, but also to the presence of additional orphan synapses. I would also mention in the discussion how the phenotype of the RTN4L2 KO compares to the GluA2/3 KO and if the lack of GluA2/3 at the AZ could explain the increase in ABR threshold. Quantification of GluA2/3 puncta at the apical, middle, and basal region would also help understand the auditory phenotype of the KO mice.

We have changed Figure2-figure supplement 3 to become a main figure (Figure 3) based on the recommendation of the reviewer. We agree, that it would be valuable to perform immunohistochemistry combining anti-GluA2/3 and anti-Homer1 and anti-Ctbp2 antibodies to see if the “orphan” Homer1 patches house GluA2/3 not juxtaposing synaptic ribbons. Unfortunately, as mentioned above, due to the expiration of our animal breeding and experimentation licenses we did not manage to do those experiments. We have however performed stainings with anti-GluA4 antibodies and could not detect GluA4 signal in RTN4RL2^-/- mice (Figure 3-figure supplement 1). This potentially could explain the more drastic ABR threshold elevation in RTN4RL2^-/- mice compared to e.g. GluA3 KO mice. We have now made this clearer in our discussion.

(6) I suggest considering the use of color-blind friendly palettes for figures and graphs in this manuscript to enhance clarity and ensure that the findings are accessible to a wider audience and improve the overall effectiveness of the presentation. Please use color-blind-friendly schemes in Figure 1 and Figure 2 Supplement 3.

Done.

(7) Could you please explain what "XX {plus minus} Y, SD = W" means in the figure legends?

Mean ± SEM (standard error of the mean), SD (standard deviation) are indicated in the legends. In response to the reviewer comment we have now added an explanation in the Materials and Methods –> Data analysis and statistics section.

(8) Please include information about the ear tested (left or right or both).

Both ears were tested. Since there was no significant difference between right and left ear we did not further consider this factor. We will add this fact more precisely in the Material and methods section.

Reviewer #3 (Recommendations for the authors):

(1) Line 90: Why not show this control, it is a nice control.

Unfortunately, our recent attempts to perform RTN4RL2 immunostaining on cryosections were unsuccessful. Therefore, we decided to remove RTN4RL2 immunostaining from Figure 1 and have adjusted the results section accordingly.

(2) Line 94: Please provide a reference for these interactions.

Done.

Author Response:

Points from reviewer 1 (Public Review):

In this manuscript, Yong and colleagues link perturbations in lysosomal lipid metabolism with the generation of protein aggregates resulting from proteosome inhibition.

We apologize for any confusion in the explanation of the results. We found that both proteasome inhibition and, independently, perturbations to lysosomal lipid metabolism lead to accumulation of protein aggregates in the lysosome. There was no evidence of proteasome inhibition in the context of lysosomal lipid perturbations (Figure 4J).

Despite using various tools of lysosomal function, acidity, permeability, etc, the authors couldn't identify the link between lysosomal lipid metabolism and protein aggregate formation.

Indeed, despite testing numerous mechanistic hypotheses, we have yet to explain how perturbation of lysosomal lipid metabolism causes protein aggregates. However, we have demonstrated that lipids are both necessary (via epistasis and serum delipidation) and sufficient (media supplementation) to drive these phenotypes.

Although this work is interesting and thought-provoking, their approach to identify novel pathways involved in proteostasis is limited and this weakens the contribution of the paper in its current form.

We are glad the reviewer found the work to be thought-provoking. As a fundamental cellular process critical for longevity, we agree that the connections made here between lipids, lysosomes and protein aggregates are interesting and broaden the impact of cellular health on proteostasis. Though we have falsified multiple hypotheses for how perturbation of lysosomal lipid metabolism could influence protein aggregation, we agree that a major weakness of the current work is our limited mechanistic understanding of this process. We hope that by engaging the thoughtful and creative eLife readership, novel mechanistic hypotheses will emerge.

Points from reviewer 2 (Public Review):

This might be too much of an ask, but they should go further in excluding one very attractive alternative model: effects on proteasome activity. This explanation should be addressed definitively because the transcription factor that regulates proteasome subunit gene expression (Nrf1/NFE2L1) is processed in the ER and is therefore well placed to be influenced by membrane conditions, and because it is shown here that proteasome inhibition increase ProteoStat puncta.

We appreciate the constructive suggestion to examine loss of proteasome expression as a relevant mechanism linking cellular dyslipidemia with proteostasis impairment. We analyzed the genome-wide perturb-seq data from Replogle et al. [1], which was performed in K562 cells cultured under similar conditions to our screen. As expected, perturbation of Nrf1/NFE2L1 reduced expression of proteasome subunits, whereas perturbation of proteasome subunits that increased proteostat staining (e.g. PSMD2, PSMD13) homeostatically increased expression of multiple proteasome subunits. In contrast, other top hits, including those related to lipid-related perturbations (e.g. MYLIP, PSAP) did not reduce the expression of genes encoding the proteasome (Author response image 1).

Author response image 1.

The relative expression of genes encoding proteasomal subunits for representative genes was re-plotted from genome-wide perturb-seq data in K562 cells [1]. Shown are hit genes that increase Proteostat staining along with non-targeting controls and the positive control gene NFE2L1. Proteasome expression was induced by proteasome impairment (PSMD2 and PSMD13) and repressed by NFE2L1 knockdown. Other hit genes related to lipid metabolism and lysosome function did not consistently impact the expression of proteasome subunits.

The authors address proteasome activity only by using a dye that is not referenced. Here a much more solid answer is needed.

We thank Reviewer #2 for bringing to our attention the missing reference for the proteasome activity probe we used (Me4BodipyFL-Ahx3Leu3VS). Both this probe [2] and its close derivative [3], BodipyFL-Ahx3Leu3VS, were fully characterized previously. We’ll include these references in the revision. In our hands, this probe behaved as expected under MG132 and Bortezomib treatment when quantified by flow cytometry (Fig. 4I), and by in-blot fluorescence scan (data will be included as supplementary in the revision). We further observed that HMGCR KD increased proteasome activity, consistent with what’s suggested by current literature. This validated our use of this probe and strongly suggested that proteasome activity was not perturbed by impaired lipid homeostasis.

In general, most conclusions in the paper rely essentially solely on ProteoStat assays. The entire study would be greatly strengthened if the authors incorporated biochemical or other modalities to substantiate their results.

We agree that orthogonal characterization of proteostasis impairment would be valuable. We chose the ProteoStat stain as a reporter of proteostasis because it is capable of integrating the aggregation states of multiple endogenously expressed proteins, and in the absence of exogenous stressors such as the overexpression of aggregation-prone proteins. With aging, a context where ProteoStat staining increases, hundreds of proteins exhibit reduced solubility [4], thus motivating the focus on endogenously expressed proteins. Despite the biochemical limitations, we think our work is differentiated from published screens focused on specific metastable proteins by our focus on regulators of endogenous proteostasis.

The presentation would be improved greatly if the authors provided diagrams illustrating the pathways implicated in their results, as well as their models.

We thank Reviewer #2 for the helpful suggestion. We have provided the suggested diagrams below (Author response image 2).

Author response image 2.

Mechanistic models linking screen hits to accrual of lysosomal protein aggregates, related to Figure 4. Perturbations that increased cholesterol and sphingolipid levels were evaluated for effects on lysosomal pH, lysosomal proteolytic capacity, lysosomal membrane permeability, lipid peroxidation and proteasome activity. None of these mechanisms appear to play a causal role in protein aggregation in response to elevated lipids.

Author Response References

1. Replogle, J. M. et al. Mapping information-rich genotype-phenotype landscapes with genome-scale Perturb-seq. Cell 185, 2559-2575.e28 (2022).

2. Berkers, C. R. et al. Probing the Specificity and Activity Profiles of the Proteasome Inhibitors Bortezomib and Delanzomib. Mol Pharmaceut 9, 1126–1135 (2012).

3. Berkers, C. R. et al. Profiling Proteasome Activity in Tissue with Fluorescent Probes. Mol. Pharmaceutics 4, 739–748 (2007).

4. David, D. C. et al. Widespread Protein Aggregation as an Inherent Part of Aging in C. elegans. Plos Biol 8, e1000450 (2010).

eLife Assessment

This study provides important insights into the role of polyUbiquitination in neurodegenerative diseases, elucidating how pUb promotes neurodegeneration by affecting proteasomal function. The findings not only offer a new perspective on the pathophysiology of neurodegenerative diseases but also provide potential targets for developing new therapeutic strategies. The experiments in the revised submission provide solid evidence to support the conclusions.

Reviewer #1 (Public review):

Summary:

The manuscript discusses the role of phosphorylated ubiquitin (pUb) by PINK1 kinase in neurodegenerative diseases. It reveals that elevated levels of pUb are observed in aged human brains and those affected by Parkinson's disease (PD), as well as in Alzheimer's disease (AD), aging, and ischemic injury. The study shows that increased pUb impairs proteasomal degradation, leading to protein aggregation and neurodegeneration. The authors also demonstrate that PINK1 knockout can mitigate protein aggregation in aging and ischemic mouse brains, as well as in cells treated with a proteasome inhibitor. While this study provided some interesting data, several important points should be addressed before being further consideration.

Strengths:

(1) Reveals a novel pathological mechanism of neurodegeneration mediated by pUb, providing a new perspective on understanding neurodegenerative diseases.

(2) The study covers not only a single disease model but also various neurodegenerative diseases such as Alzheimer's disease, aging, and ischemic injury, enhancing the breadth and applicability of the research findings.

Comments on revisions:

This study, through a systematic experimental design, reveals the crucial role of pUb in forming a positive feedback loop by inhibiting proteasome activity in neurodegenerative diseases. The data are comprehensive and highly innovative. However, some of the results are not entirely convincing, particularly the staining results in Figure 1.

In Figure 1A, the density of DAPI staining differs significantly between the control patient and the AD patient, making it difficult to conclusively demonstrate a clear increase in PINK1 in AD patients. Quantitative analysis is needed. In Fig 1C, the PINK1 staining in the mouse brain appears to resemble non-specific staining.

Author response:

The following is the authors’ response to the original reviews

Reviewer #1 (Public review):

Summary:

The manuscript discusses the role of phosphorylated ubiquitin (pUb) by PINK1 kinase in neurodegenerative diseases. It reveals that elevated levels of pUb are observed in aged human brains and those affected by Parkinson's disease (PD), as well as in Alzheimer's disease (AD), aging, and ischemic injury. The study shows that increased pUb impairs proteasomal degradation, leading to protein aggregation and neurodegeneration. The authors also demonstrate that PINK1 knockout can mitigate protein aggregation in aging and ischemic mouse brains, as well as in cells treated with a proteasome inhibitor. While this study provided some interesting data, several important points should be addressed before being further considered.

Strengths:

(1) Reveals a novel pathological mechanism of neurodegeneration mediated by pUb, providing a new perspective on understanding neurodegenerative diseases.

(2) The study covers not only a single disease model but also various neurodegenerative diseases such as Alzheimer's disease, aging, and ischemic injury, enhancing the breadth and applicability of the research findings.

Weaknesses:

(1) PINK1 has been reported as a kinase capable of phosphorylating Ubiquitin, hence the expected outcome of increased p-Ub levels upon PINK1 overexpression. Figures 5E-F do not demonstrate a significant increase in Ub levels upon overexpression of PINK1 alone, whereas the evident increase in Ub expression upon overexpression of S65A is apparent. Therefore, the notion that increased Ub phosphorylation leads to protein aggregation in mouse hippocampal neurons is not yet convincingly supported.

Indeed, overexpression of sPINK1 alone resulted in minimal changes in Ub levels in the soluble fraction (Figure 5E), which is expected given that the soluble Ub pool remains relatively stable and buffered. However, sPINK1* overexpression led to a marked increase in Ub levels in the insoluble fraction, indicative of increased protein aggregation (Figure 5F). The molecular weight distribution of Ub in the insoluble fraction was predominantly below 70 kDa, suggesting that phosphorylation inhibits Ub chain elongation.

To further validate this mechanism, we utilized the Ub/S65A mutant to antagonize Ub phosphorylation and observed a significant reduction in the intensity of aggregated bands at low molecular weights, indicating restored proteasomal activity. The observed increase in Ub levels in the soluble fraction upon Ub/S65A overexpression is likely due to enhanced ubiquitination driven by elevated Ub-S65A, and notably, Ub/S65A was also detectable using an antibody against wild-type Ub.

Consistent with these findings, overexpression of Ub/S65E resulted in a further increase in Ub levels in the insoluble fraction, with intensified low molecular weight bands. The effect was even more pronounced than that observed with sPINK1 transfection, likely resulting from the complete phosphorylation mimicry achieved by Ub/S65E, compared to the relatively low levels of phosphorylation by PINK1.

These findings collectively support the conclusion that sPINK1 promotes protein aggregation via Ub phosphorylation. We have updated the Results and Discussion sections to more clearly present the data and explain the various controls.

(2) The specificity of PINK1 and p-Ub antibodies requires further validation, as a series of literature indicate that the expression of the PINK1 protein is relatively low and difficult to detect under physiological conditions.

We acknowledge the challenges in achieving high specificity with commercially available and customgenerated antibodies targeting PINK1 and pUb, particularly given their low endogenous expression under physiological conditions. However, in our study, we observed robust immunofluorescent staining for PINK1 (Figures 1A, 1C, and 1G) and pUb (Figures 1B, 1D, and 1G) in human brain samples from Alzheimer's disease (AD) patients, as well as in mouse models of AD and cerebral ischemia. The clear visualization can be partly attributed to the pathological upregulation of PINK1 and pUb under disease conditions. Importantly, the images from pink1^-/- mice exhibit much weaker staining.

Additionally, we detected a significant elevation in the pUb levels in aged mouse brains compared to younger ones (Figures 1E and 1F). In contrast, pink1^-/- mice showed no change in pUb levels with aging, despite some background signals, demonstrating that pUb accumulation during aging is PINK1dependent. Collectively, these results support the specificity of the antibodies used in detecting pathophysiological changes in PINK1 and pUb levels.

For cultured cells, pink1^-/- cells served as a negative control for both PINK1 (Figures 2B and 2C) and pUb (Figures 2D and 2E). While the pUb Western blot exhibited some nonspecific background, pUb levels in pink1^-/- cells remained unchanged across all MG132 treatment conditions (Figures 2D and 2E), further attesting the usability of the antibodies in conjunction with appropriated controls.

We have updated the manuscript with higher-resolution images; individual image files have been uploaded separately.

(3) In Figure 6, relying solely on Western blot staining and Golgi staining under high magnification is insufficient to prove the impact of PINK1 overexpression on neuronal integrity and cognitive function. The authors should supplement their findings with immunostaining results for MAP2 or NeuN to demonstrate whether neuronal cells are affected.

We included NeuN immunofluorescent staining at 10, 30, and 70 days post transfection in Figure 5— figure supplement 2. The results clearly demonstrate a significant loss of NeuN-positive cells in the hippocampus following Ub/S65E overexpression, while no apparent reduction was observed with sPINK1 transfection alone. 

We have also quantified MAP2 protein levels via Western blotting and examined morphology of neuronal dendrite and synaptic structure using Golgi staining. These analyses revealed a significant reduction in MAP2 levels and synaptic damage upon sPINK1 or Ub/S65E overexpression (Figures 6F and 6H), consistent with the proteomics analysis (Figure 5—figure supplementary 5). Notably, these detrimental effects could be rescued by co-expression of Ub/S65A, reinforcing the role of pUb in mediating these structural changes.

Together, our findings from NeuN immunostaining, MAP2 protein analysis, proteomics analysis, and Golgi staining provide strong evidence for the impact of PINK1 overexpression and pUb elevation on neuronal integrity and synaptic structure.

(4) The authors should provide more detailed figure captions to facilitate the understanding of the results depicted in the figures.

Figure captions have been updated with more details incorporated in the revised manuscript.

(5) While the study proposes that pUb promotes neurodegeneration by affecting proteasomal function, the specific molecular mechanisms and signaling pathways remain to be elucidated.

The molecular mechanisms and signaling pathways through which pUb promotes neurodegeneration are likely multifaceted and interconnected. Our findings suggest that mitochondrial dysfunction plays a central role following sPINK1* overexpression. This is supported by (1) an observed increase in full-length PINK1, indicative of impaired mitochondrial quality control, and (2) proteomic data showing enhanced mitophagy at 30 days post-transfection, followed by substantial mitochondrial injuries at 70 days post-transfection (Figure 5—figure supplement 5 and Supplementary Data). The progressive mitochondrial damage caused by protein aggregates would exacerbate neuronal injury and degeneration.

Additionally, reduced proteasomal activity may lead to the accumulation of inhibitory proteins that are normally degraded by the ubiquitin-proteasome system. Our proteomics analysis identified a >50fold increase in CamK2n1 (UniProt ID: Q6QWF9), an endogenous inhibitor of CaMKII activation, following sPINK1* overexpression. The accumulation of CamK2n1 suppresses CaMKII activation, thereby inhibiting the CREB signaling pathway (Figure 7), which is essential for synaptic plasticity and neuronal survival. This disruption can further contribute to neurodegenerative processes.

Thus, our findings underscore the complexity of pUb-mediated neurodegeneration and call for further investigation into downstream consequences.

Reviewer #1 (Recommendations for the authors):

Suggestions for improved or additional experiments, data or analyses.

We have performed additional experiments to investigate how the impairment of ubiquitinproteasomal activity contributes to neurodegeneration. Specifically, we investigated CamK2n1, an endogenous inhibitor of CaMKII, which is normally degraded by the proteasome to allow CaMKII activation. Our proteomics analysis revealed a significant (>50-fold) elevation of CamKI2n1 following sPINK1 overexpression (Figure 5—figure supplement 5 and Supplementary Data).

To validate this mechanism, we conducted immunofluorescence and Western blot analyses, demonstrating reduced levels of phosphorylated CaMKII (pCaMKII) and phosphorylated CREB (pCREB), as well as reduced levels of downstream proteins such as BDNF and ERK. These results have been incorporated into the revised manuscript (Figure 7).

As the proteasome is crucial in maintaining proteostasis, its dysregulation would trigger neurodegeneration through multiple pathways, contributing to a broad cascade of pathological events.

Reviewer #2 (Public review):

Summary:

The manuscript makes the claim that pUb is elevated in a number of degenerative conditions including Alzheimer's Disease and cerebral ischemia. Some of this is based on antibody staining which is poorly controlled and difficult to accept at this point. They confirm previous results that a cytosolic form of PINK1 accumulates following proteasome inhibition and that this can be active. Accumulation of pUb is proposed to interfere with proteostasis through inhibition of the proteasome. Much of the data relies on over-expression and there is little support for this reflecting physiological mechanisms.

Weaknesses:

The manuscript is poorly written. I appreciate this may be difficult in a non-native tongue, but felt that many of the problems are organizational. Less data of higher quality, better controls and incision would be preferable. Overall the referencing of past work is lamentable. Methods are also very poor and difficult to follow.

Until technical issues are addressed I think this would represent an unreliable contribution to the field.

(1) Antibody specificity and detection under pathological conditions

We recognize the limitations of commercially available antibodies for detecting PINK1 and pUb. Nevertheless, our findings reveal a significant elevation in PINK1 and pUb levels under pathological conditions, such as Alzheimer's disease (AD) and ischemia. Additionally, we observed an increase in pUb level during brain aging, further demonstrating its relevance and a potentially causative role for this special pathological condition. Similarly, elevated pUb levels were observed for cultured cells following pharmacological treatment or oxygen-glucose deprivation (OGD).

In contrast, in pink1^-/- mice and HEK293 cells used as negative controls, PINK1 and pUb levels remained consistently low. Therefore, the observed elevation of PINK1 and pUb are associated with special pathological conditions, rather than an antibody-detection anomaly.

(2) Overexpression as a model for pathological conditions

To investigate whether the inhibitory effects of sPINK1 on the ubiquitin-proteasome system (UPS) depend on its kinase activity, we employed a kinase-dead version of sPINK1* as a negative control. Given that PINK1 targets multiple substrates, we also investigated whether its effects on UPS inhibition were specifically mediated by ubiquitin phosphorylation. To this end, we used Ub/S65A (a phospho-null mutant) to block Ub phosphorylation by sPINK1, and Ub/S65E (a phospho-mimetic mutant) to mimic phosphorylated Ub. These well-defined controls ensured the robustness of our conclusions.

Although overexpression does not perfectly replicate physiological conditions, it provides a valuable model for studying pathological scenarios such as neurodegeneration and brain aging, where pUb levels are elevated. For example, we observed a 30.4% increase in pUb levels in aged mouse brains compared to young brains (Figure 1F). Similarly, in our sPINK1 overexpression model, pUb levels increased by 43.8% and 59.9% at 30- and 70-days post-transfection, respectively, compared to controls (Figures 5A and 5C). Notably, co-expression of sPINK1* with Ub/S65A almost entirely prevented sPINK1* accumulation (Figure 5B), indicating that an active UPS can efficiently degrade this otherwise stable variant of sPINK1.

Together, our findings demonstrate that sPINK1 accumulation inhibits UPS activity, an effect that can be reversed by the phospho-null Ub mutant. The overexpression model mimics pathological conditions and provides valuable insights into pUb-mediated proteasomal dysfunction.

(3) Organization of the manuscript

Following your suggestion, we have restructured the manuscript to present the key findings in a more logical and cohesive sequence:

(a) Evidence for elevated PINK1 and pUb levels across a broad spectrum of pathological and neurodegenerative conditions;

(b) The effects of pUb elevation in cultured cells, focusing on the proteasome;

(c) Mechanistic insights into how pUb elevation inhibits proteasomal activity;

(d) The absence of PINK1 and pUb alleviates protein aggregation;

(e) Evidence for the causative relationship between elevated pUb levels and proteasomal inhibition;

(f) Demonstration that pUb elevation directly contributes to neuronal degeneration;

(g) Give an additional evidence to explain the mechanism of neuronal degeneration post sPINK1* over-expression. The downstream effects of elevated CamK2n1, an inhibitor of CaMKII, resulting from proteasomal inhibition.

This reorganization should ensure a clear and progressive narrative, and enhance the overall coherence and impact of the revised manuscript.

(4) Revisions to writing, referencing, and methodology

We have made a great effort to enhance the clarity and flow of the manuscript, including the addition of references to appropriately acknowledge prior work. We have also expanded the Methods section with additional details to improve readability and ensure reproducibility. We believe these revisions effectively address the concerns raised and strengthen the overall quality of the manuscript.

Reviewer #2 (Recommendations for the authors):

Figure 1: PINK1 is a poorly expressed protein and difficult to detect by Western blot let alone by immunofluorescence. I have direct experience of the antibody used in this study and do not consider it reliable. There are much cleaner reagents out there, although they still have many challenges. The minimal requirement here is for the PINK1 antibody staining to be compared in wild-type and knockout mice. One would also expect to see a mitochondrial staining which would require higher magnification to be definitive, but it does not look like it to me. This is a key foundational figure and is unreliable. The pUb antibody also has a high background, see for example figure 2E.

Under physiological conditions, PINK1 and pUb levels are indeed low, making their detection challenging. However, under pathological conditions, their expression is significantly elevated, correlating with disease severity. Given the limitations of available reagents, using appropriate controls is a standard approach in biological research.

Nevertheless, we observed robust immunofluorescent staining for PINK1 (Figures 1A, 1C, and 1G) and pUb (Figures 1B, 1D, and 1G) in human brain samples from Alzheimer’s disease (AD) patients and mouse models of AD and cerebral ischemia. Compared to healthy controls, the significant elevation of PINK1 and pUb under these pathological conditions accounts for their clear visualization. To validate antibody specificity, we have included images from pink1^-/- mice as negative controls (Figure 1C and 1D, third panel).

Furthermore, we analyzed pUb levels in both young and aged mice, using pink1^-/- mice as controls.

Our results revealed a significant increase in pUb levels in aged wild-type mice (Figures 1E and 1F), In contrast, pink1^-/- mice exhibited relatively low pUb levels, with no notable change between young and aged groups. These findings reinforce the conclusion that pUb accumulation during aging is dependent on PINK1.Furthermore, we analyzed pUb levels in both young and aged mice, using pink1^-/- mice as controls.

For HEK293 cells, pink1^-/- cells were used as a negative control for assessing PINK1 (Figures 2B and 2C) and pUb levels (Figures 2D and 2E). While the pUb Western blot did show some nonspecific background, as you have noted, pUb levels significantly increased following MG132 treatment of the wildtype cells. In contrast, no such increase was observed in pink1^-/- cells (Figure 2D and 2E). These results further validate the reliability of our findings.

Regarding mitochondrial staining, we recognize that PINK1 localization can vary depending on the pathological context. For example, in Alzheimer’s disease, PINK1 exhibits relatively high nuclear staining, while in cerebral ischemia and brain aging, it is predominantly cytoplasmic and punctate. In contrast, in young, healthy mouse brains, PINK1 is more uniformly distributed. The observed elevation in pUb levels could arise from mitochondrial PINK1 or soluble sPINK1 in the cytoplasm, and it remains unclear whether nuclear PINK1 contributes to pUb accumulation. Investigating the role of PINK1 in different forms and subcellular localizations will be an important avenue for future research.

To enhance clarity, we have updated our images and replaced them with higher-resolution versions in the revised manuscript.

Please also confirm that the GAPDH loading controls represent the same gels, to my eye they do not match.

We have reviewed all the bands, and confirmed that the GAPDH loading controls correspond to the same gels. For different gels, we use separate GAPDH loading controls. There are two experimental scenarios to consider:

(1) When there is a large difference in molecular weight between target proteins, we cut the gel into sections and incubate each section with different antibodies separately.

(2) When the molecular weight difference is small and cutting is not feasible, we first probe the membrane with one antibody, strip it, and then re-incubate the membrane with a second antibody.

These approaches ensure accurate and reliable detection of target proteins with various molecular weights relative to GAPDH.

1H. Ponceau.

We have corrected the spelling.

Figure 2 many elements are confirmation of work already reported and this must be made clearer in the text. 

Indeed, the elevation of sPINK1 and pUb upon proteasomal inhibition has been previously reported, and these studies have been acknowledged (Gao, et al, 2016; Dantuma, et al, 2000). In the present study, we expand on these findings by conducting a detailed analysis of the time- and concentrationdependent effects of MG132 on sPINK1 and pUb levels, establishing a causative relationship between pUb accumulation and proteasomal inhibition. Furthermore, we demonstrate that sPINK1 overexpression and MG132-induced proteasomal inhibition exhibit no additive effect, indicating that both converge on the same pathway, resulting in the impairment of proteasomal activity.

It has been established that ubiquitin phosphorylation inhibits Ub chain elongation (Wauer, et al, 2015). However, our study provides novel insights by identifying an additional mechanism: phosphorylated Ub also interferes with the noncovalent interactions between Ub chain and Ub receptors in the proteasome, which further contributes to the impairment of UPS function.

The PINK1 kinase-dead mutant construction (Figure 2F) and the use of Ub-GFP as a proteasomal substrate were based on established methodologies, which have been appropriately cited in the manuscript (Beilina, etal 2005 for KD sPINK1; Yamano, et al for endogenous PINK1; Samant, et al, 2018 and Dantuma, et al, 2000 for Ub-GFP probe). Similarly, our use of puromycin and BALA treatments follows previously reported protocols (Gao, et al, 2016), which allowed us to dissect the relative contributions of sPINK1* overexpression to proteasomal vs. autophagic dysfunction.

As you have noted, our study has built upon prior findings while introducing new mechanistic insights into sPINK1 and pUb-mediated proteasomal dysfunction.

2C 24h MG132 not recommended, most cells are dead by then.

We used MG132 treatment for 24 hours to evaluate the time-course effects of proteasomal inhibition on PINK1 and pUb levels in HEK293 cells (Figures 2C and 2E). We did observe some decrease in both PINK1 and pUb levels at 24 hours compared to 12 hours, which may result from some extend of cell death at the longer treatment duration.

In SH-SY5Y cells, we collected cells at 24 hours after MG132 administration (Figure 5—figure supplementary 1). Though protein aggregation was evident in these cells, we did not observe pronounced cell death under these conditions, justifying our treatment.

Our findings are consistent with previous studies demonstrating that MG132 at 5 µM for 24 hours effectively induces proteasomal inhibition without substantial cytotoxicity. For example, studies using human esophageal squamous cancer cells have reported that this treatment condition inhibits cell proliferation while maintaining cell viability, with cell viability >70% after 24-hour treatment with 5 µM MG132 (Int J Mol Med 33: 1083-1088, 2014). 

MG132 has been commonly used at concentrations ranging from 5 to 50 µM for durations of 1 to 24 hours, as stated at the vendor’s website （https://www.cellsignal.com/products/activatorsinhibitors/mg-132/2194）.

2I what is BALA do they mean bafilomycin. This is a v-ATPase inhibitor, not just an autophagy inhibitor.

We appreciate the reviewer’s comment regarding the use of BALA in Figure 2I. To clarify, BALA refers to bafilomycin A1, a well-established v-ATPase inhibitor that blocks lysosomal acidification. While bafilomycin A1 is commonly used as an autophagy inhibitor, its primary mechanism involves inhibiting lysosomal function, which is critical for autophagosome-lysosome fusion and subsequent degradation of autophagic cargo.

In our study, we used bafilomycin A1 in conjunction with puromycin to dissect the relative contributions of sPINK1 overexpression on proteasomal and autophagic activities. Puromycin induces protein misfolding and aggregation, causing stress on both degradation pathways. By inhibiting lysosomal function with bafilomycin A1 and blocking the protein degradation load at various stages, we can tell the relative contributions of autophagy and UPS pathways.

We acknowledge that bafilomycin A1’s effects extend beyond autophagy, as it also inhibits v-ATPase activity. However, its inhibition of lysosomal degradation is integral to distinguishing autophagy’s contribution under the experimental conditions, and BALA treatment has been used in extensively in previous studies (Mauvezin and Neufeld, 2015). 

We have further clarified this treatment in the revised manuscript.

Figure 3. Legend or text needs to be more explicit about how chains have been produced. From what I can gather from methods only a single E2 has been trialed. Authors should use at least one of the criteria used by Wauer et al. (2014) to confirm the stoichiometry of phosphorylation. The concept that pUb can interfere with E2 discharging is not new, but not universal across E2s.

We have cited in the manuscript that PINK1-mediated ubiquitin phosphorylation can interfere with ubiquitin chain elongation for certain E2 enzymes (Wauer et al., 2015). 

To clarify, the focus of our current work is on how elevation of Ub phosphorylation impacts UPS activity, rather than exploring the broader effects of Ub phosphorylation on Ub chain elongation. For this reason, we have used the standard E2 that is well-established for generating K48-linked polyUb chain (Pickart CM, 2005). Moreover, our findings go further and by demonstrate that phosphorylated K48-linked polyubiquitin exhibits weaker non-covalent interactions with proteasomal ubiquitin receptors. This dual effect—on both covalent chain elongation and non-covalent interactions— contributes to the observed reduction in ubiquitin-proteasome activity, a novel aspect of our study.

To address the reviewer’s concerns, we have added details in the Methods section and figure legends regarding the generation of ubiquitin chains. Specifically, we used ubiquitin-activating enzyme E1 (UniProt ID: P22314) and ubiquitin-conjugating enzyme E2-25K (UniProt ID: P61086) to generate K48-linked ubiquitin chains. 

Our ESI-MS analysis showed that only 1–2 phosphoryl groups were incorporated into the K48-linked tetra-ubiquitin chains (Figure 3—figure supplement 2). This is consistent with our in vivo findings, where pUb levels increased by 30.4% in aged mouse brains compared to young brains (Figure 1F). Notably, even sub-stoichiometric phosphorylation onto the K48-linked ubiquitin chain significantly weakens the non-covalent interactions with the proteasome (Figures 3E and 3H).

Figure 4. I could find no definition of the insoluble fraction, nor details on how it is prepared.

The insoluble fraction primarily contains proteins that are aggregated or associated with hydrophobic interactions and cannot be solubilized by RIPA buffer. We have provided more details in the Methods of the revised manuscript about how the insoluble fraction was prepared. Our approach was based on established protocols for fractionating soluble and insoluble proteins from brain tissues (Wirths, 2017). Here is an outline of the procedure, which enables the separation and subsequent analysis of distinct protein populations:

• Lysis and preparation of soluble fraction: Cells and brain tissues were lysed using RIPA buffer (Beyotime Biotechnology, cat# P0013B) containing protease (P1005) and phosphatase inhibitors (P1081) on ice for 30 minutes, with gentle vortexing every 10 minutes. Brain samples were homogenized using a precooled TissuePrep instrument (TP-24, Gering Instrument Company). Lysates were centrifuged at 12,000 rpm for 30 minutes at 4°C. The supernatant was collected as the soluble protein fraction.

• Preparation of insoluble fraction: The pellet was resuspended in 20 µl of SDS buffer (2% SDS, 50 mM Tris-HCl, pH 7.5) and subjected to ultrasonic pyrolysis at 4°C for 8 cycles (10 seconds ultrasound, 30 seconds interval). The samples were then centrifuged at 12,000 rpm for 30 minutes at 4°C. The supernatant obtained after this step was designated as the insoluble protein fraction.

• Protein quantification: Protein concentrations for both soluble and insoluble fractions were determined using the BCA Protein Assay Kit (Beyotime Biotechnology, cat# P0009).

Figure 5. What is the transfection efficiency? How many folds is sPINK1 over-expressed? Typically, a neuron will have only a few hundred copies of PINK1 at the basal state. How much mutant ubiquitin is expressed relative to wild type, seeing the free ubiquitin signals on the gels might be helpful here, but they seem to have been cut off. 

We appreciate the reviewer's insightful comments regarding transfection efficiency, the extent of sPINK1 overexpression, and the expression levels of mutant ubiquitin relative to wild-type ubiquitin. Below, we provide detailed responses to each point:

Transfection Efficiency: Our immunofluorescent staining for NeuN, a neuronal marker, demonstrated that over 90% of NeuN-positive cells were co-localized with GFP (Figure 5—figure supplement 2), indicating a high transfection efficiency in our neuronal cultures.

Extent of sPINK1 Overexpression: Quantifying the exact fold increase of sPINK1 upon overexpression is inherently difficult due to its low basal expression under physiological conditions, making the relative increase difficult to measure (small denominator effect). However, our Western blot analysis shows that ischemic events can cause a substantial elevation of PINK1 levels, including both full-length and cleaved forms (Figure 1H). This suggests that our overexpression model recapitulates the pathological increase in PINK1, making it a relevant system for studying disease mechanisms.

From Figure 5B, it is evident that sPINK1 levels differ significantly between neurons overexpressing sPINK1 alone and those co-expressing sPINK1 + Ub/S65A (70 days post-transfection). Overexpression of sPINK1 alone results in multiple PINK1 bands, consistent with sPINK1, endogenous PINK1 (induced by mitochondrial damage), and ubiquitinated sPINK1. In comparison, co-expressing Ub/S65A leads to faint PINK1 bands, suggesting that in the presence of a functionally restored proteasome, overexpressed sPINK1 is rapidly degraded. Therefore, actual accumulation of sPINK1 depends on proteasomal activity, and the “over-expressed” PINK1 level can be comparable to levels observed under native, pathological conditions.

Expression Levels of Mutant Ubiquitin Relative to Wild-Type: Assessing the expression levels of mutant versus wild-type ubiquitin is indeed valuable. In Figure 5E, we observed a 38.9% increase in high-molecular-weight ubiquitin conjugates in the soluble fraction when comparing the sPINK1+Ub/S65A group to the control. This increase suggests that mutant ubiquitin is actively incorporated into polyubiquitin chains.

Regarding free monomeric ubiquitin, its low abundance and rapid incorporation into polyubiquitin chains make it difficult to visualize in Western blots. Additionally, its low molecular weight and lower antibody binding valency further reduce its visibility.

General: a number of effects are shown following over-expression but no case is made that these levels of pUb are ever attained physiologically. I am very unconvinced by these findings and think the manuscript needs to be improved at multiple levels before being added to the record.

We understand the reviewer’s concerns regarding the relevance of pUb levels observed in our overexpression model. To clarify, our study is not focused on physiological levels of pUb, but rather on pathologically elevated levels, which have been documented in various neurodegenerative conditions. While overexpression is not a perfect replication of pathological states, it provides a valuable tool to investigate mechanisms that become relevant under disease conditions. Moreover, we have taken steps to ensure the validity of our findings and to address potential limitations associated with overexpression models:

Pathological Relevance: Besides several reported literatures, we observed significant increases in PINK1 and pUb levels in human brain samples from Alzheimer's disease (AD) patients, as well as in mouse models of AD, cerebral ischemia (including mouse middle cerebral artery occlusion ischemic model and oxygen glucose deprivation cell model), and aging (e.g., Figures 1E, 1F, and 1H). All these data show that pUb levels are elevated under pathological conditions. Our overexpression model mimics these pathological scenarios by recreating the high levels of pUb, which lead to the impairment of proteasomal activity and subsequent disruption of proteostasis.

Use of Robust Controls: To ensure the reliability of our results and interpretations, we employed multiple controls for our experiments. We have used pink1^-/- mice and cells to confirm that pUb accumulation is PINK1-dependent (Figures 1C and 2C). We have also included kinase-dead sPINK1 mutant and Ub/S65A phospho-null mutants to negate/counteract the specific roles of PINK1 activity and pUb in proteasomal dysfunction. On the other hand, we have used Ub/S65E for phosphomimetic mutant, corresponding to a 100% Ub phosphorylation.

Importantly, we have compared sPINK1 overexpression with both baseline and disease-mimicking conditions, thus to ensure that the observed effects are consistent with pathological changes. Furthermore, our findings are supported by complementary evidences from human brain samples, model animals, cell cultures, and molecular assays. Integrating the different controls and various approaches, we have provided mechanistic insights into how elevated pUb levels causes proteasomal impairment and contributes to neurodegeneration.

Our findings elucidate how elevated pUb level contributes to the disruption of proteostasis in neurodegenerative conditions. While overexpression may have limitations, it remains a powerful tool for dissecting pathological mechanisms and testing hypotheses. Our results align with and expand upon previous studies suggesting pUb as a biomarker of neurodegeneration (Hou, et al, 2018; Fiesel, et al, 2015), and provide mechanistic insights into how elevated pUb and sPINK1 drive a viscous feedforward cycle, ultimately leading to proteasomal dysfunction and neurodegeneration. 

We hope these clarifications highlight the relevance and rigor of our study, and welcome additional suggestions to improve the manuscript.

Reviewer #3 (Public review):

Summary:

This study aims to explore the role of phosphorylated ubiquitin (pUb) in proteostasis and its impact on neurodegeneration. By employing a combination of molecular, cellular, and in vivo approaches, the authors demonstrate that elevated pUb levels contribute to both protective and neurotoxic effects, depending on the context. The research integrates proteasomal inhibition, mitochondrial dysfunction, and protein aggregation, providing new insights into the pathology of neurodegenerative diseases.

Strengths:

- The integration of proteomics, molecular biology, and animal models provides comprehensive insights.

- The use of phospho-null and phospho-mimetic ubiquitin mutants elegantly demonstrates the dual effects of pUb.

- Data on behavioral changes and cognitive impairments establish a clear link between cellular mechanisms and functional outcomes.

Weaknesses:

- While the study discusses the reciprocal relationship between proteasomal inhibition and pUb elevation, causality remains partially inferred.

It has been well-established that protein aggregates, particularly neurodegenerative fibrils, can impair proteasomal activity (McDade, et al., 2024; Kinger, et al., 2024; Tseng, et al., 2008). Other contributing factors, including ATP depletion, reduced proteasome component expression, and covalent modifications of proteasomal subunits, can also lead to declined proteasomal function. Additionally, mitochondrial injury serves as an important source of elevated PINK1 and pUb levels. Recent studies have demonstrated that efficient mitophagy is essential to prevent pUb accumulation, whereas partial mitophagy failure results in elevated PINK1 levels (Chin, et al, 2023; Pollock, et al. 2024).

While pathological conditions can impair proteasomal function and slow sPINK1 degradation, leading to its accumulation, our results demonstrate that overexpression of sPINK1 or PINK1 can initiate this cycle as well. Once this cycle is initiated, it becomes self-perpetuating, as sPINK1 and pUb accumulation progressively impair proteasomal function, leading to more protein aggregates and mitochondrial damages.

Importantly, we show that co-expression of Ub/S65A effectively rescues cells from this cycle, which further illustrates the pivotal role of pUb in driving proteasomal inhibition and the causality between pUb elevation and proteasomal inhibition. At the animal level, pink1 knockout prevents protein aggregation under aging and cerebral ischemia conditions (Figures 1E and 1G). 

Together, by controlling at protein, cell, and animal levels, our findings support this self-reinforcing and self-amplifying cycle of pUb elevation, proteasomal inhibition, protein aggregation, mitochondrial damage, and ultimately, neurodegeneration.

- The role of alternative pathways, such as autophagy, in compensating for proteasomal dysfunction is underexplored.

Indeed, previous studies have shown that elevated sPINK1 can enhance autophagy (Gao, et al., 2016,), potentially compensating for impaired UPS function. One mechanism involves PINK1mediated phosphorylation of p62, which enhances autophagic activity.

In our study, we observed increased autophagic activity upon sPINK1 overexpression, as shown in Figure 2I (middle panel, without BALA). This increase in autophagy may facilitate the degradation of ubiquitinated proteins induced by puromycin, partially mitigating proteasomal dysfunction. This compensation might also explain why protein aggregation, though statistically significant, increased only slightly at 70 days post-sPINK1 transfection (Figure 5F). Additionally, we detected a mild but statistically insignificant increase in LC3II levels in the hippocampus of mouse brains at 70 days postsPINK1 transfection (Figure 5—figure supplement 6), further supporting the notion of autophagy activation.

However, while autophagy may provide some compensation, its effect is likely limited. The UPS and autophagy serve distinct roles in protein degradation:

• Autophagy is a bulk degradation pathway, primarily targeting damaged organelles, intracellular pathogens, and protein aggregates, often in a non-selective manner.

• The UPS, in contrast, is highly selective, degrading short-lived regulatory proteins, misfolded proteins, and proteins tagged for degradation via ubiquitination.

Thus, while sPINK1 overexpression enhances autophagy-mediated degradation, it simultaneously impairs UPS-mediated degradation. This suggests that autophagy partially compensates for proteasomal dysfunction but is insufficient to counterbalance the UPS's selective degradation function. We have incorporated additional discussion in the revised manuscript.

- The immunofluorescence images in Figure 1A-D lack clarity and transparency. It is not clear whether the images represent human brain tissue, mouse brain tissue, or cultured cells. Additionally, the DAPI staining is not well-defined, making it difficult to discern cell nuclei or staging. To address these issues, lower-magnification images that clearly show the brain region should be provided, along with improved DAPI staining for better visualization. Furthermore, the Results section and Figure legends should explicitly indicate which brain region is being presented. These concerns raise questions about the reliability of the reported pUb levels in AD, which is a critical aspect of the study's findings.

We have taken steps to address the concerns regarding clarity and transparency in Figure 1A-D. We have already addressed the source of tissues at the left of each images. For example, we have written “human brain with AD” at the left side of Figure 1A, and “mouse brains with AD” at the left side of Figure 1C.

Briefly, the human brain samples in Figure 1 originate from the cingulate gyrus of Alzheimer’s disease (AD) patients. Our analysis revealed that PINK1 is primarily localized within cell bodies, whereas pUb is more abundant around Aβ plaques, likely in nerve terminals. For the mouse brain samples, we have now explicitly indicated in the figure legends and Results section that the images represent the neocortex of APP/PS1 mice, a mouse model relevant to AD pathology, as well as the corresponding regions in wild-type and pink1^-/- mice. We have ensured that the brain regions and sources are clearly stated throughout the manuscript.

Regarding image clarity, we have uploaded higher-resolution versions of the images in the revised manuscript to improve visualization of key features, including DAPI staining. We believe these revisions enhance the reliability and interpretability of our findings, particularly in relation to the reported pUb levels in AD. 

- Figure 4B should also indicate which brain region is being presented.

The images were taken for layer III-IV in the neocortex of mouse brains. We have included this information in the figure legend of the revised manuscript.

Reviewer #3 (Recommendations for the authors):

- Expand on the potential compensatory role of autophagy in response to proteasomal dysfunction.

Upon proteasomal inhibition, cells may activate autophagy as an alternative pathway of degradation to help clear damaged or misfolded proteins. Autophagy is a bulk degradation process that targets long-lived proteins, damaged organelles, and aggregated proteins for lysosomal degradation. While this pathway can provide some compensation, it is distinct from the ubiquitin-proteasome system (UPS), which specializes in the selective degradation of short-lived regulatory proteins and misfolded proteins.

In our study, we observed increased autophagic activity following sPINK1 overexpression (Figure 2J, middle panel, without BALA) and a slight, though statistically insignificant, increase in LC3II levels in the hippocampus of mouse brains at 70 days post-sPINK1 transfection (Figure 5—figure supplement 6). These findings suggest that autophagy is indeed upregulated as a compensatory response to proteasomal dysfunction, potentially facilitating the degradation of aggregated ubiquitinated proteins. Additionally, gene set enrichment analysis (GSEA) revealed similar enrichment of autophagy pathways at 30 and 70 days post-sPINK1 overexpression (Figure 5—figure supplement 5).

However, the compensatory capacity of autophagy is likely limited. While autophagy can reduce protein aggregation, it is an inherently non-selective process and cannot fully replace the targeted functions of the UPS. Moreover, as we illustrate in Figure 7 of the revised manuscript, UPS is essential for degrading specific regulatory and inhibitory proteins and plays a critical role in cellular proteostasis, particularly in signaling regulation, cell cycle control, and stress responses.

Together, while autophagy activation provides some degree of compensation, it cannot fully restore cellular proteostasis. The interplay between these two degradation pathways is an important area for future investigation. For the present study, our focus is on how pUb elevations impact proteasomal activity and elicits downstream effects.

We have incorporated these additional discussions on this topic in the revised manuscript.

- Simplify the discussion of complex mechanisms to improve accessibility for readers.

We have revised the Discussion to present the mechanisms in a more coherent and accessible manner, ensuring clarity for a broader readership. These revisions should make the discussion more intuitive while preserving the depth of our findings.

- Statistical analyses could benefit from clarifying how technical replicates and biological replicates were accounted for across experiments.

We have clarified our statistical analysis in the Methods section and figure legends, explicitly detailing how many biological replicates were accounted for across experiments. These revisions should enhance transparency and clarity, ensuring that our findings are robust and reproducible.

- The image in Figure 3D is too small to distinguish any signals. A larger and clearer image should be presented.

We have expanded the images in Figure 3D. Additionally, we have replaced figures with version of better resolutions throughout the manuscript.

- NeuN expression in Figure 4B differs between wildtype and pink-/- mice. Additional validation is needed to determine whether pink-/- enhances NeuN expression.

The difference in NeuN immunofluorescence intensity between wild-type and pink1^-/- mice in Figure 4B may simply result from variations in image acquisition rather than an actual difference in NeuN expression.

Our single nuclei RNA-seq analyses of wild-type and pink1^-/- mice at 3 and 18 months of age reveal no significant differences in NeuN expression at the transcript level (data provided below). This confirms that the observed variation in fluorescence intensity is unlikely to reflect an authentic upregulation of NeuN expression. Thus, factors like the concentration of antibody, image exposure and processing may contribute to differences in staining intensity.

Author response image 1.

eLife Assessment

This manuscript presents an in-depth analysis of gene expression across multiple brown algal species with differing life histories, providing convincing evidence for the conservation of life cycle-specific gene expression. While largely descriptive, the study is an important step forward in understanding the core cellular processes that differ between life cycle phases, and its findings will be of broad interest to developmental and evolutionary biologists.

Reviewer #2 (Public review):

Summary:

The manuscript by Ratchinski et al presents a comprehensive analysis of developmental and life history gene expression patterns in brown algal species. The manuscript shows that the degree of generation bias or generation-specific gene expression correlates with the degree of dimorphism. It also reports conservation of life cycle features within generations and marked changes in gene expression patterns in Ectocarpus in the transition between gamete and early sporophyte. The manuscript also reports considerable conservation of gene expression modules between two representative species, particularly in genes associated with conserved functional characteristics.

Strengths:

The manuscript represents a considerable "tour de force" dataset and analytical effort. While the data presented is largely descriptive, it is likely to provide a very useful resource for studies of brown algal development and for comparative studies with other developmental and life cycle systems.

Comments on revisions

The authors have provided in their response (point 1) a good clarification for their rationale in excluding fucoid algae from the study, based on the diploid nature of the fucoid life cycle. Similarly, they have noted (point 2) that "the relationship between changes in gene expression during very early sporophyte development and during alternation of life cycle generations could be investigated further using a highlydimorphic kelp model system such as Saccharina latissima." For the benefit of the reader who may not be too familiar with the different life cycles in brown algae, I would recommend that these clarifications are included in the Discussion.

Otherwise the authors have addressed my previous comments adequately.

Author response:

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

Reviewer #1 (Public review): 

Summary: 

The authors have examined gene expression between life cycle stages in a range of brown macroalgae to examine whether there are conserved aspects of biological features. 

Strengths: 

The manuscript incorporates large gene expression datasets from 10 different species and therefore enables a comprehensive assessment of the degree of conservation of different aspects of gene expression and underlying biology. 

The findings represent an important step forward in our understanding of the core aspects of cell biology that differ between life cycle phases and provide a substantial resource for further detailed studies in this area. Convincing evidence is provided for the conservation of lifecycle-specific gene expression between species, particularly in core housekeeping gene modules. 

Weaknesses: 

I found a few weaknesses in the methodology and experimental design. I think the manuscript could have been clearer when linking the findings to the biology of the brown algae. 

Reviewer #2 (Public review): 

Summary: 

The manuscript by Ratchinski et al presents a comprehensive analysis of developmental and life history gene expression patterns in brown algal species. The manuscript shows that the degree of generation bias or generation-specific gene expression correlates with the degree of dimorphism. It also reports conservation of life cycle features within generations and marked changes in gene expression patterns in Ectocarpus in the transition between gamete and early sporophyte. The manuscript also reports considerable conservation of gene expression modules between two representative species, particularly in genes associated with conserved functional characteristics. 

Strengths: 

The manuscript represents a considerable "tour de force" dataset and analytical effort. While the data presented is largely descriptive, it is likely to provide a very useful resource for studies of brown algal development and for comparative studies with other developmental and life cycle systems. 

Weaknesses: 

Notwithstanding the well-known issues associated with inferring function from transcriptomics-only studies, no major weaknesses were identified by this reviewer. 

Reviewing Editor Comments:

The overall assessment of the reviewers does not contain major aspects of concern. We nevertheless recommend that the authors carefully consider the constructive comments, as this will further improve their manuscript. 

Reviewer #1 (Recommendations for the authors): 

(1) Line 32: The abstract states 'considerable conservation of co-expressed gene modules', but the degree of conservation between Ectocarpus and D. dichotoma appeared limited to specific subsets of genes with highly conserved housekeeping functions, e.g., translation. I think the wording of the abstract should be rephrased to better reflect this. 

We agree that genes with housekeeping functions figure strongly in the gene modules that showed strong conservation between Ectocarpus species 7 and D. dichotoma (and we actually highlight this point in the manuscript) but we do not believe that this invalidates the conservation. In the analysis shown in Figure 6A, for example, high scores were obtained for both connectivity and density for about a third of the gene modules and these modules cover broad range of cellular functions. This is a significant result given the large phylogenetic distance and we feel that "considerable conservation" is appropriate as a description of the level of correlation. 

(2) Introduction - The Introduction needs a better explanation of the biology of the life cycle phases. Some of this information is present in the 1st paragraph of Materials and Methods, although it would be preferable to include this information within the main text, ideally within the Introduction before the Results are described. For example, when are flagella present? The presence of flagella could be indicated in Figure 3. The ecology of the life cycle is also not described. Are life cycles present in the same ecological niche? Do they co-exist or occupy distinct environments? It would be useful to understand how the observed genotypes could relate to this wider aspect of the brown algal biology. 

We have added a sentence to explain that zoids (gametes and spores) are the only flagellated stages of the life cycle (line 678). In addition, in the legend for Figure 3, we have indicated which of the life cycle stages analysed in panel 3A consisted entirely or partially of flagellated cells. We have also added information about phenology to the Introduction. 

(3) Line 127. 'The proportion of generation specific genes was positively correlated with the level of dimorphism'. The level of dimorphism between species was not clear to me. This needs to be clearly displayed in Figure 1B. 

We had attempted to illustrate the level of dimorphism, using the size of each generation as a measurable proxy, in Figure S1 but we agree that the information was not very clearly presented. To improve clarity, we now provide independent size scales for each generation of the life cycle in this figure and state in the legend that "Size bars indicate the approximate sizes of each generation of each life cycle, providing an indication of the degree of dimorphism between the two generations.". In the text, Figure S1 is cited earlier in the paragraph but we now repeat the citation of the figure at the end of the sentence "The proportion of generation-specific genes (...) was positively correlated with the level of dimorphism" so that the reader can specifically consult the supplementary figure for this phenotypic parameter. 

(4) Line 267. Are there known differences in cell wall composition between life cycle phases or within each generation as individual life cycle phases mature (e.g., differences between unicellular and multicellular stages)? 

Detailed comparative analyses of cell wall composition at different stages of the life cycle have not been carried out for brown algae. However, Congo red stains Ectocarpus gametophytes but not sporophytes (Coelho et al., 2011), indicating a difference in cell wall composition between the two generations. Zoids (spores and gametes) do not have a cell wall and calcofluor white staining of meio-spores has indicated that a cell wall only starts to be deposited 24-48 hours post-release (Arun et al., 2013).

(5) Line 388. The authors should comment on the accuracy of OrthoFinder for different gene types across this degree of divergence (250 MYA). The best conservation was found in genes with housekeeping characteristics (line 401). It may be that these gene modules show the highest degree of conservation in expression patterns, but I also wonder whether they pattern may also emerge because finding true orthologues is easier for highly conserved gene families. 

We do not believe that this is the case because, as mentioned above, the "housekeeping" modules cover quite a broad range of cellular functions. Note also that the modules were given functional labels based on their being clearly enriched in genes corresponding to a particular class of function but not all the genes in a module have a predicted function that corresponds to the functional classification. 

However, we have carried out an analysis to look for evidence of the bias proposed by the reviewer. For this, we used BLASTp identity scores as an approximate proxy for pairwise identity between Ectocarpus species 7 and D. dichotoma one-to-one orthologues in each module and plotted the mean identity score for each module against the Fischer test p-value of the contingency table in Figure 6C (Author response image 1).

Author response image 1.

Plot of estimations of the mean percent shared identity between the orthologues within each module (based on mean BLASTp identity scores) against log10(pvalue) values obtained with the Fisher's exact test applied in Figure 6C to determine whether pairs of modules shared a greater number of one-to-one orthologues than expected from a random distribution. Error bars indicate the standard deviation. 

This analysis did not detect any correlation between the degree of sequence conservation of orthologues in a module and the degree of conservation of the module between Ectocarpus species 7 and D. dichotoma.

Minor comments 

(1) Line 650 loose should be lose.

The error has been corrected.

(2) Line 695 filtered through a 1 μm filter to remove multicellular gametophyte fractions. Is this correct? It seems too small to allow gametes to pass through. 

Yes, the text is correct, a 1 μm filter was used. The gametes do pass through this filter, presumably because they do not have a rigid cell wall, allowing them to squeeze through the filter when a light pressure is applied. 

(3) Line 709 - DDT should be DTT 

The error has been corrected.

Reviewer #2 (Recommendations for the authors): 

(1) It is not clear why the chosen species for analysis do not include fucoid algae, which display a high degree of dimorphism between generations and which are relatively well studied with respect to gene expression patterns during early development. Indeed, it was recently shown that gene expression patterns in developing embryos of Fucus spp. obey the "hourglass" pattern whereby gene expression shows a minima of transcription age index (i.e., higher expression of evolutionarily older genes) associated with differentiation at the phylotypic stage. I am somewhat surprised that the manuscript does not consider this feature in the analysis or discussion. 

Brown algae of the order Fucales have diploid life cycles and therefore do not alternate between a sporophyte and gametophyte generation. It is for this reason that we thought that it was more interesting to compare Ectocarpus species 7 with D. dichotoma, which has a haploid-diploid life cycle.

(2) In Discussion, the comparison of maternal to zygote transition in animals and land plants, which show a high degree of dimorphism, with Ectocarpus would be strengthened by data/discussion from other brown algae that show a high degree of dimorphism. 

Animals have diploid life cycles and dimorphism in that lineage generally refers to sexual rather than generational dimorphism. Land plants do have highly dimorphic haploiddiploid life cycles but it is unclear how this characteristic relates to events that occur during the maternal to zygote transition. In Ectocarpus, the transition from gamete to the first stages of sporophyte development involved more marked changes in gene expression than we observed when comparing the mature sporophyte and gametophyte generations (Figure 3C). At present, there is no evidence that events during these two transitions are correlated. The relationship between changes in gene expression during very early sporophyte development and during alternation of life cycle generations could be investigated further using a highly dimorphic kelp model system such as Saccharina latissima but we are not aware of any studies that have specifically addressed this point. 

(3) Since marked changes were observed during the transition from gamete to early sporophyte in Ectocarpus, it would be interesting to know how gene expression patterns change during the transition from gamete to partheno-sporophyte. Would the same patterns of downregulation and upregulation be expected? 

The sporophyte individuals derived from gamete parthenogenesis (parthenosporophytes) are indistinguishable morphologically and functionally from diploid sporophytes derived from gamete fusions (see line 76). They also express generation marker genes in a comparable manner (Peters et al., 2008). Based on these observations, we have treated partheno-sporophytes and diploid sporophytes as equivalent in our experiments. For clarity, we have now distinguished partheno-sporophyte from diploid sporophyte samples in Table S1. 

(4) The authors show a correlation between the degree of dimorphism and generation-biased or generation-specific expression. How was the degree of dimorphism quantified? 

The degree of dimorphism is illustrated in Figure S1 using the relative size of the two generations as a proxy. Size estimations are approximate because the size of an individual of a particular species is quite variable but the ten species nonetheless represent a very clear gradient of dimorphism due to the extreme differences in size between generations of species at each end of the scale, with the sporophyte generation being several orders of magnitude larger than the gametophyte generation or visa versa. 

References

Arun A, Peters NT, Scornet D, Peters AF, Cock JM, Coelho SM. 2013. Non-cell autonomous regulation of life cycle transitions in the model brown alga Ectocarpus. New Phytol 197:503– 510. doi:10.1111/nph.12007

Coelho SM, Godfroy O, Arun A, Le Corguillé G, Peters AF, Cock JM. 2011. OUROBOROS is a master regulator of the gametophyte to sporophyte life cycle transition in the brown alga Ectocarpus. Proc Natl Acad Sci USA 108:11518–11523. doi:10.1073/pnas.1102274108

Peters AF, Scornet D, Ratin M, Charrier B, Monnier A, Merrien Y, Corre E, Coelho SM, Cock JM. 2008. Life-cycle-generation-specific developmental processes are modified in the immediate upright mutant of the brown alga Ectocarpus siliculosus. Development 135:1503–1512.doi:10.1242/dev.016303

eLife Assessment

In this preregistered study, Kunkel and colleagues set out to compare the magnitude and duration of placebo versus nocebo effects in healthy volunteers, and also to examine the different factors contributing to these effects. The authors follow a rigorous methodology in a within-subjects design, taking into consideration standard conventions for manipulation of expectations, and using an appropriate sham condition. They present compelling evidence of long-lasting placebo and nocebo effects, with nocebo responses demonstrating consistently greater strength. These valuable results have the potential for a great impact in the field of experimental and clinical pain.

Reviewer #1 (Public review):

Summary:

The study aimed to: (1) assess the magnitude of placebo and nocebo effects immediately after induction through verbal instructions and conditioning, (2) examine the persistence of these effects one week later, and (3) identify predictors of sustained placebo and nocebo responses over time.

Strengths:

An innovation was to use sham TENS stimulation as the expectation manipulation. This expectation manipulation was reinforced not only by the change in pain stimulus intensity, but also by delivery of non-painful electrical stimulation, labelled as TENS stimulation.

Questionnaire-based treatment expectation ratings were collected before conditioning and after conditioning, and after the test session, which provided an explicit measure of participant's expectations about the manipulation.

The finding that placebo and nocebo effects are influenced by recent experience provides a novel insight into a potential moderator of individual placebo effects.

Weaknesses:

There are a limited number of trials per test condition (10) which means that the trajectory of responses to the manipulation may not be explored, which would be an interesting future study.

The differences between the nocebo and control condition in pain ratings during conditioning could be explained by differing physiological effects of the different stimulus intensities, so it is difficult to make any claims about the expectation effects here. A a randomisation error meant that 25 participants received an unbalanced number 448 of trials per condition (i.e., 10 x VAS 40, 14 x VAS 60, 12 x VAS 80), although the authors accounted for this during analysis so it is not of major concern.

This manuscript presents a study on expectation manipulation to induce placebo and nocebo effects in healthy participants. The study follows standard placebo experiment conventions with use of TENS stimulation as the placebo manipulation. The authors were able to achieve their aims. A key finding is that placebo and nocebo effects were predicted by recent experience, which is a novel contribution to the literature. The findings provide insights into the differences between placebo and nocebo effects and the potential moderators of these effects.

Comments on revisions:

I am satisfied with the author's revisions to the manuscript and have no further comments.

Reviewer #2 (Public review):

Summary:

Kunkel et al aim to answer a fundamental question: Do placebo and nocebo effects differ in magnitude or longevity? To address this question, they used a powerful within-participants design, with a very large sample size (n=104), in which they compared placebo and nocebo effects - within the same individuals - across verbal expectations, conditioning, testing phase, and a 1-week follow-up. With elegant analyses, they establish that different mechanisms underlie the learning of placebo vs nocebo effects, with the latter being acquired faster and extinguished slower. This is an important finding for both the basic understanding of learning mechanisms in humans and for potential clinical applications to improve human health.

Strengths:

Beyond the above - the paper is well-written and very clear. It lays out nicely the need for the current investigation and what implications it holds. The design is elegant, and the analyses are rich, thoughtful, and interesting. The sample size is large which is highly appreciated, considering the longitudinal, in-lab study design. The question is super important and well-investigated, and the entire manuscript is very thoughtful with analyses closely examining the underlying mechanisms of placebo versus nocebo effects.

Comments on revisions:

The authors have addressed all of my concerns and comments - one point for them to verify is that indeed analyses that have not been preregistered will be flagged as such. The provided pre-registration link doesn't specify much about the analysis plans and specific tests used.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public review):

Summary:

This manuscript presents a study on expectation manipulation to induce placebo and nocebo effects in healthy participants. The study follows standard placebo experiment conventions with the use of TENS stimulation as the placebo manipulation. The authors were able to achieve their aims. A key finding is that placebo and nocebo effects were predicted by recent experience, which is a novel contribution to the literature. The findings provide insights into the differences between placebo and nocebo effects and the potential moderators of these effects.

Specifically, the study aimed to:

(1) assess the magnitude of placebo and nocebo effects immediately after induction through verbal instructions and conditioning

(2) examine the persistence of these effects one week later, and

(3) identify predictors of sustained placebo and nocebo responses over time.

Strengths:

An innovation was to use sham TENS stimulation as the expectation manipulation. This expectation manipulation was reinforced not only by the change in pain stimulus intensity, but also by delivery of non-painful electrical stimulation, labelled as TENS stimulation.

Questionnaire-based treatment expectation ratings were collected before conditioning and after conditioning, and after the test session, which provided an explicit measure of participants' expectations about the manipulation.

The finding that placebo and nocebo effects are influenced by recent experience provides a novel insight into a potential moderator of individual placebo effects.

We thank the reviewer for their thorough evaluation of our manuscript and for highlighting the novelty and originality of our study.

Weaknesses:

There are a limited number of trials per test condition (10), which means that the trajectory of responses to the manipulation may not be adequately explored.

We appreciate the reviewer’s comment regarding the number of trials in the test phase. The trial number was chosen to ensure comparability with previous studies addressing similar research questions with similar designs (e.g. Colloca et al., 2010). Our primary objective was to directly compare placebo and nocebo effects within a within-subject design and to examine their persistence one week after the first test session. While we did not specifically aim to investigate the trajectory of responses within a single testing session, we fully agree that a comprehensive analysis of the trajectories of expectation effects on pain would be a valuable extension of our work. We have now acknowledged this limitation and future direction in the revised manuscript.

The paragraph reads as follows: “It is important to note that our study was designed in alignment with previous studies addressing similar questions (e.g., Colloca et al., 2010). Our primary aim was to directly compare placebo and nocebo effects in a within-subject design and assess their persistence of these effects one week following the first test session. One limitation of our approach is the relatively short duration of each session, which may have limited our ability to examine the trajectory of responses within a single session. Future studies could address this limitation by increasing the number of trials for a more comprehensive analysis.”

On day 8, one stimulus per stimulation intensity (i.e., VAS 40, 60, and 80) was applied before the start of the test session to re-familiarise participants with the thermal stimulation. There is a potential risk of revealing the manipulation to participants during the re-familiarization process, as they were not previously briefed to expect the painful stimulus intensity to vary without the application of sham TENS stimulation.

We thank the reviewer for the opportunity to clarify this point. Participants were informed at the beginning of the experiment that we would use different stimulation intensities to re-familiarize them with the stimuli before the second test session. We are therefore confident that participants perceived this step as part of a recalibration rather than associating it with the experimental manipulation. We have added this information to the revised version of the manuscript.

The paragraph now reads as follows: “On day 8, one stimulus per stimulation intensity (i.e., VAS 40, 60 and 80) was applied before the start of the test session to re-familiarise participants with the thermal stimulation. Note that participants were informed that these pre-test stimuli were part of the recalibration and refamiliarization procedure conducted prior to the second test session.”

The differences between the nocebo and control conditions in pain ratings during conditioning could be explained by the differing physiological effects of the different stimulus intensities, so it is difficult to make any claims about expectation effects here.

We appreciate the reviewer’s comment and agree that, despite the careful calibration of the three pain stimuli, we cannot entirely rule out the possibility that temporal dynamics during the conditioning session were influenced by differential physiological effects of the varying stimulus intensities (e.g., intensity-dependent habituation or sensitization). We have addressed this in the revision of the manuscript, but we would like to emphasize that the stronger nocebo effects during the test phase are statistically controlled for any differences in the conditioning session.

The paragraph now reads: “This asymmetry is noteworthy in and of itself because it occurred despite the equidistant stimulus calibration relative to the control condition prior to conditioning. It may be the result of different physiological effects of the stimuli over time or amplified learning in the nocebo condition, consistent with its heightened biological relevance, but it could also be a stronger effect of the verbal instructions in this condition.”

A randomisation error meant that 25 participants received an unbalanced number of 448 trials per condition (i.e., 10 x VAS 40, 14 x VAS 60, 12 x VAS 80).

We agree that this is indeed unfortunate. However, we would like to point out that all analyses reported in the manuscript have been controlled for the VAS ratings in the conditioning session, i.e., potential effects of the conditioned placebo and nocebo stimuli. Moreover, we have now conducted additional analyses, presented here in our response to the reviewers, to demonstrate that this imbalance did not systematically bias the results. Importantly, the key findings observed during the test phase remain robust despite this issue.

Specifically, when excluding these 25 participants from the analyses, the reported stronger nocebo compared to placebo effects in the test session on day 1 remain unchanged. Likewise, the comparison of placebo and nocebo effects between days 1 and 8 shows the same pattern when excluding the participants in question. The only exception is the interaction between effect (placebo vs nocebo) x session (day 1 vs day 8), which changed from a borderline significant result (p = .049) to insignificant (p = .24). However, post hoc tests continued to show the same pattern as originally reported: a significant reduction in the nocebo effect from day 1 to day 8 and no significant change in the placebo effect.

Reviewer #2 (Public review):

Summary:

Kunkel et al aim to answer a fundamental question: Do placebo and nocebo effects differ in magnitude or longevity? To address this question, they used a powerful within-participants design, with a very large sample size (n=104), in which they compared placebo and nocebo effects - within the same individuals - across verbal expectations, conditioning, testing phase, and a 1-week follow-up. With elegant analyses, they establish that different mechanisms underlie the learning of placebo vs nocebo effects, with the latter being acquired faster and extinguished slower. This is an important finding for both the basic understanding of learning mechanisms in humans and for potential clinical applications to improve human health.

Strengths:

Beyond the above - the paper is well-written and very clear. It lays out nicely the need for the current investigation and what implications it holds. The design is elegant, and the analyses are rich, thoughtful, and interesting. The sample size is large which is highly appreciated, considering the longitudinal, in-lab study design. The question is super important and well-investigated, and the entire manuscript is very thoughtful with analyses closely examining the underlying mechanisms of placebo versus nocebo effects.

We thank the reviewer for their positive evaluation of our manuscript and for acknowledging the methodological rigor and the significant implications for clinical applications and the broader research field.

Weaknesses:

There were two highly addressable weaknesses in my opinion:

(1) I could not find the preregistration - this is crucial to verify what analyses the authors have committed to prior to writing the manuscript. Please provide a link leading directly to the preregistration - searching for the specified number in the suggested website yielded no results.

We thank the reviewer for pointing this out. We included a link to the preregistration in the revised manuscript. This study was pre-registered with the German Clinical Trial Register (registration number: DRKS00029228; https://drks.de/search/de/trial/DRKS00029228).

(2) There is a recurring issue which is easy to address: because the Methods are located after the Results, many of the constructs used, analyses conducted, and even the main placebo and nocebo inductions are unclear, making it hard to appreciate the results in full. I recommend finding a way to detail at the beginning of the results section how placebo and nocebo effects have been induced. While my background means I am familiar with these methods, other readers will lack that knowledge. Even a short paragraph or a figure (like Figure 4) could help clarify the results substantially. For example, a significant portion of the results is devoted to the conditioning part of the experiment, while it is unknown which part was involved (e.g., were temperatures lowered/increased in all trials or only in the beginning).

We thank the reviewer for their helpful comment and agree that the Results section requires additional information that would typically be provided by the Methods section if it directly followed the Introduction. In response, we have moved the former Figure 4 from the Methods section to the beginning of the Results section as a new Figure 1, to improve clarity. Further, we have revised the Methods section to explicitly state that all trials during the conditioning phase were manipulated in the same way.

Recommendations for the Authors:

Reviewer #1 (Recommendations for the authors):

(1) Given that the authors are claiming (correctly) that there is only limited work comparing placebo/nocebo effects, there are some papers missing from their citations:

Nocebo responses are stronger than placebo responses after subliminal pain conditioning - - Jensen, K., Kirsch, I., Odmalm, S., Kaptchuk, T. J. & Ingvar, M. Classical conditioning of analgesic and hyperalgesic pain responses without conscious awareness. Proc. Natl. Acad. Sci. USA 112, 7863-7 (2015)

We thank the reviewer and have now included this relevant publication into the introduction of the revised manuscript.

Hird, E.J., Charalambous, C., El-Deredy, W. et al. Boundary effects of expectation in human pain perception. Sci Rep 9, 9443 (2019). https://doi.org/10.1038/s41598-019-45811-x

We thank the reviewer for suggesting this relevant publication. We have now included it into the discussion of the revised manuscript by adding the following paragraph:

“Recent work using a predictive coding framework further suggests that nocebo effects may be less susceptible to prediction error than placebo effects (Hird et al., 2019), which could contribute to their greater persistence and strength in our study.”

(2) The trial-by-trial pain ratings could have been usefully modelled with a computational model, such as a Bayesian model (this is especially pertinent given the reference to Bayesian processing in the discussion). A multilevel model could also be used to increase the power of the analysis. This is a tentative suggestion, as I appreciate it would require a significant investment of time and work - alternatively, the authors could acknowledge it in the Discussion as a useful future avenue for investigation, if this is preferred.

We thank the reviewer for this thoughtful suggestion. While we agree that computational modelling approaches could provide valuable insights into individual learning, our study was not designed with this in mind and the relatively small number of trials per condition and the absence of trial-by-trial expectancy ratings limit the applicability of such models. We have therefore chosen not to pursue such analysis but highlight it in the discussion as a promising direction for future research.

“Notably, the most recent experience was the most predictive in all three analyses; for instance, the placebo effect on day 8 was predicted by the placebo effect on day 1, not by the initial conditioning. This finding supports the Bayesian inference framework, where recent experiences are weighted more heavily in the process of model updating because they are more likely to reflect the current state of the environment, providing the most relevant and immediate information needed to guide future actions and predictions24. Interestingly, while a change in pain predicted subsequent nocebo effects, it seemed less influential than for placebo effects. This aligns with findings that longer conditioning enhanced placebo effects, while it did not affect nocebo responses10 and the conclusion that nocebo instruction may be sufficient to trigger nocebo responses. Using Bayesian modeling, future studies could identify individual differences in the development of placebo and nocebo effects by integrating prior experiences and sensory inputs, providing a probabilistic framework for understanding the underlying mechanisms.”

(3) The paper is missing any justification of sample size, i.e. power analysis - please include this.

We apologize for the missing information on our a priori power analysis. As there is a lack of prior studies investigating within-subjects comparisons of placebo and nocebo effects that could inform precise effect size estimates for our research question, we based our calculation on the ability detect small effects. Specifically, the study was powered to detect effect sizes in the range of d = 0.2 - 0.25 with α = .05 and power = .9, yielding a required sample size of N = 83-129. We have now added this information to the methods section of the revised manuscript.

(4) "On day 8, one stimulus per stimulation intensity (i.e., VAS 40, 60 and 80) was applied before the start of the test session to re-familiarise participants with the thermal stimulation."

What were the instructions about this? Was it before the electrode was applied? This runs the risk of unblinding participants, as they only expect to feel changes in stimulus intensity due to the TENS stimulation.

We thank the reviewer for pointing out the potential risk of unblinding participants due to the re-familiarization process prior to the second test session. We would like to clarify that we followed specific procedures to prevent participants from associating this process with the experimental manipulation. The re-familiarisation with the thermal stimuli was conducted after the electrode had been applied and re-tested to ensure that both stimulus modalities were re-introduced in a consistent and neutral context. Participants were explicitly informed that both procedures were standard checks prior to the actual test session (“We will check both once again before we begin the actual measurement.”). For the thermal stimuli, we informed participants that they would experience three different intensities to allow the skin to acclimate (e.g., “...we will test the heat stimuli in 3 trials with different temperatures, allowing your skin to acclimate to the stimuli. …”), without implying any connection to the experimental conditions.

Importantly, this re-familiarization procedure mirrored what participants had already experienced during the initial calibration session on day 1. We therefore assume that participants interpreted as a routine technical step rather than part of the experimental manipulation. We have now clarified this procedure in the methods section of the revised manuscript.

(5) "For a comparison of pain intensity ratings between time-points, an ANOVA with the within-subject factors Condition (placebo, nocebo, control) and Session (day 1, day 8) was carried out. For the comparison of placebo and nocebo effects between the two test days, an ANOVA with the with-subject factors Effect (placebo effect, nocebo effect) and Session (day 1, day 8) was used."

It seems that one ANOVA is looking at raw pain scores and one is looking at difference scores, but this is a bit confusing - please rephrase/clarify this, and explain why it is useful to include both.

We thank the reviewer for highlighting this point. Our primary analyses focus on placebo and nocebo effects, which we define as the difference in pain intensity ratings between the control and the placebo condition (placebo effect) and the nocebo and the control condition (nocebo effect), respectively.

To examine whether condition effects were present at each time-point, we first conducted two separate repeated measures ANOVAs - one for day 1 and one for day 8 - with the within-subject factor CONDITION (placebo, nocebo, control).

To compare the magnitude and persistence of placebo and nocebo effects over time, we then calculated the above-mentioned difference scores and submitted these to a second ANOVA with within-subject factors EFFECT (placebo vs. nocebo effect) and SESSION (day 1 vs. day 8). We have now clarified this approach on page 19 of the revised manuscript. To avoid confusion, the Condition x Session ANOVA has been removed from the manuscript.

(6) Please can the authors provide a figure illustrating trial-by-trial ratings during test trials as well as during conditioning trials?

In response to the reviewer’s point, we now provide the trial-by-trial ratings of the test phases on days 1 and 8 as an additional figure in the Supplement (Figure S1) and would like to clarify that trial-by-trial pain intensity ratings of the conditioning phase are displayed in Figure 2C of the manuscript,

(7) "Separate multiple linear regression analyses were performed to examine the influence of expectations (GEEE ratings) and experienced effects (VAS ratings) on subsequent placebo and nocebo effects. For day 1, the placebo effect was entered as the dependent variable and the following variables as potential predictors: (i) expected improvement with placebo before conditioning, (ii) placebo effect during conditioning and (iii) the expected improvement with placebo before the test session at day 1"

The term "placebo effect during conditioning" is a bit confusing - I believe this is just the effect of varying stimulus intensities - please could the authors be more explicit on the terminology they use to describe this? NB changes in pain rating during the conditioning trials do not count as a placebo/nocebo effect, as most of the change in rating will reflect differences in stimulation intensity.

We agree with the reviewer that the cited paragraph refers to the actual application of lower or higher pain stimuli during the conditioning session, rather than genuinely induced placebo or nocebo effect. We thank the reviewer for this helpful observation and have revised the terminology, accordingly, now referring to these as “pain relief during conditioning” and “pain worsening during conditioning”.

(8) Supplementary materials: "The three temperature levels were perceived as significantly different (VAS ratings; placebo condition: M= 32.90, SD= 16.17; nocebo condition: M= 56.62, SD= 17.09; control condition: M= 80.84, SD= 12.18"

This suggests that the VAS rating for the control condition was higher than for the nocebo condition. Please could the authors clarify/correct this?

We thank the reviewer for spotting this error. The values for the control and the nocebo condition had accidentally been swapped. This has now been corrected in the manuscript: control condition: M= 56.62, SD= 17.09; nocebo condition: M= 80.84, SD= 12.18.

(9) "To predict placebo responses a week later (VAScontrol - VASplacebo at day 8), the same independent variables were entered as for day 1 but with the following additional variables (i) the placebo effect at day 1 and (ii) the expected improvement with placebo before the test session at day 8."

Here it would be much clearer to say 'pain ratings during test trials at day 1".

We agree with the reviewer and have revised the manuscript as suggested.

(10) For completeness, please present the pain intensity ratings during conditioning as well as calibration/test trials in the figure.

Please see our answer to comment (6).

(11) In Figure 1a, it looks like some participants had rated the control condition as zero by day 8. If so, it's inappropriate to include these participants in the analysis if they are not responding to the stimulus. Were these the participants who were excluded due to pain insensitivity?

On day 8, the lowest pain intensity ratings observed were VAS 3 in the placebo condition and VAS 2 in the control condition, both from the same participant. All other participants reported minimum values of VAS 11 or higher (all on a scale from 0-100). Thus, no participant provided a pain rating of VAS 0, and all ratings indicated some level of pain perception in response to the stimulus. We did not define an exclusion criterion based on day 8 pain ratings in our preregistration, and we did not observe any technical issues with the stimulation procedure. To avoid post-hoc exclusions and maintain consistency with our preregistered analysis plan, we therefore decided to include all participants in the analysis.

(12) "Comparison of day 1 and day 8. A direct comparison of placebo and nocebo effects on day 1 and day 8 pain intensity ratings showed a main effect of Effect with a stronger nocebo effect (F(1,97)= 53.93, 131 p< .001, η2= .36) but no main effect of Day (F(1,97)= 2.94, p= .089, η2 = .029). The significant Effect x Session interaction indicated that the placebo effect and the nocebo effect developed differently over time (F(1,97)= 3.98, p= .049, η2 = .039)"

This is confusing as it talks about a main effect of "day" and then interaction with "session" - are they two different models? The authors need to clarify.

We thank the reviewer for pointing this out. In our analysis, “Session” is the correct term for the experimental factor, which has two factor levels, “day 1” and “day 8”. This has now been corrected in the revised manuscript.

Reviewer #2 (Recommendations for the authors):

(1) More information on how "size of the effect" in Figures 1b and 2b was calculated is needed; this can be in the legend. If these are differences between control and each condition, then they were reversed for one condition (nocebo?), which is ok - but this should be clearly explained.

We agree with the reviewer and have now revised the figure legends to improve clarity. The legends now read:

1b: “Figure 1. Pain intensity ratings and placebo and nocebo effects during calibration and test sessions. (A) Mean pain intensity ratings in the placebo, nocebo and control condition during calibration, and during the test sessions at day 1 and day 8. (B) Placebo effect (control condition - placebo condition, i.e., positive value of difference) and nocebo effect (nocebo condition - control condition, i.e., positive value of difference) on day 1 and day 8. Error bars indicate the standard error of the mean, circles indicate mean ratings of individual participants. *: p < .001, : p < .01, n.s.: non-significant.”

2b: “Figure 2. Mean and trial-by-trial pain intensity ratings, placebo and nocebo effects during conditioning. (A) Mean pain intensity ratings of the placebo, nocebo and control condition during conditioning. (B) Placebo effect (control condition - placebo condition, i.e., positive value of difference) and nocebo effect (nocebo condition - control condition, i.e., positive value of difference) during conditioning. (C) Trial-by-trial pain intensity ratings (with confidence intervals) during conditioning. Error bars indicate the standard error of the mean, circles indicate mean ratings of individual participants. ***: p < .001.”

(2) In the methods, I was missing a clear understanding of how many trials there were in the conditioning phase, and then how many in the other testing phases. Also, how long did the experiment last in total?

We apologize that the exact number of trials in the testing phases was not clear in the original manuscript. We now indicate on page 18 of the revised manuscript that we used 10 trials per condition in the test sessions. We have also added information on the duration of each test day (i.e., three hours on day 1 and one hour on day 8) on page 15.

(3) In expectancy ratings, line 186 - are improvement and worsening expectations different from expected pain relief? It is implied that these are two different constructs - it would be helpful to clarify that.

We agree that this is indeed confusing and would like to clarify that both refer to the same construct. We used the Generic rating scale for previous treatment experiences, treatment expectations, and treatment effects (GEEE questionnaire, Rief et al. 2021) that discriminates between expected symptom improvement, expected symptom worsening, and expected side effects due to a treatment. We now use the terms “expected pain relief” and “expected pain worsening” throughout the whole manuscript.

(4) In the last section of the Results, somatosensory amplification comes out of nowhere - and could be better introduced (see point 2 above).

We agree with the reviewer that introducing the concept of somatosensory amplification and its potential link to placebo/nocebo effects only in the Methods is unhelpful, given that this section appears at the end of the manuscript. We therefore now introduce the relevant publication (Doering et al., 2015) before reporting our findings on this concept.

(5) In line 169, if the authors want to specify what portion of the variance was explained by expectancy, they could conduct a hierarchical regression, where they first look at R2 without the expectancy entered, and only then enter it to obtain the R2 change.

We fully agree that hierarchical regression can be a useful approach for isolating the contribution of variables. However, in our case, expectancy was assessed at different time points (e.g., before conditioning and before the test session on day 1), and there was no principled rationale for determining the order in which these different expectancy-related variables should be entered into a hierarchical model.

That said, in response to the reviewer’s suggestion, we have now conducted hierarchical regression analyses in which all expectancy-related variables were entered together as a single block (see below). These analyses largely confirmed the findings reported so far and are provided here in the response to the reviewers below. Given the exploratory nature of this grouping and the lack of an a priori hierarchy, we feel that the standard multiple regression models remain the most appropriate for addressing our research question because it allows us to evaluate the total contribution of expectancy-related predictors while also examining the individual contribution of each variable within the block. We would therefore prefer to retain these as the primary analyses in the manuscript.

Results of the hierarchical regression analyses:

Day 1 - Placebo response: In step 1, we entered the difference in pain intensity ratings between the control and the placebo condition during conditioning as a predictor. In step 2, we added the two variables reflecting expectations (i.e., expected improvement with placebo (i) before conditioning and (ii) before the test session on day 1). This allowed us to assess whether expectation-related variables explained additional variance beyond the effect of conditioning.

The overall regression model at step 1 was significant, F(1, 102) = 13.42, p < .001, explaining 11.6% of the variance in the dependent variable (R² = .116). Adding the expectancy-related predictors in step 2 did not lead to a significant increase in explained variance, ΔR² = .007, F(2, 100) = 0.384, p = .682. Thus, the conditioning response significantly predicted placebo-related pain reduction on day 1, but additional information on expectations did not account for further variance.

Day 1 - Nocebo response: The equivalent analysis was run for the nocebo response on day 1. In step 1, the pain intensity difference between the nocebo and the control condition was entered as a predictor before adding the two expectancy ratings (i.e., expected worsening with nocebo (i) before conditioning and (ii) before the test session on day 1).

In step 1, the regression model was not statistically significant, F(1, 102) = 2.63, p = .108, and explained only 2.5% of the variance in nocebo response (R² = .025). Adding the expectation-related predictors in Step 2 slightly increased the explained variance by ΔR² = .027, but this change was also non-significant, F(2, 100) = 1.41, p = .250. The overall variance explained by the full model remained low (R² = .052). These results suggest that neither conditioning nor expectation-related variables reliably predicted nocebo-related pain increases on day 1.

Day 8 - Placebo response: For the prediction of the placebo effect on day 8, the following variables reflecting perceived effects were entered as predictors in step 1: the difference in pain intensity ratings between the control and the placebo condition (i) during conditioning and (ii) on day 1. In step 2, the variables reflecting expectations were added: the expected improvement with placebo (i) before conditioning, (ii) before the test session on day 1 and (iii) before the test session on day 8.

In step 1, the model was statistically significant, F(3, 95) = 14.86, p < .001, explaining 23.8% of the variance in the placebo response (R² = .238, Adjusted R² = .222). In step 2, the addition of the expectation-related predictors resulted in a non-significant improvement in model fit, ΔR² = .051, F(3, 92) = 2.21, p = .092. The overall variance explained by the full model increased modestly to 29.0%.

Day 8 - Nocebo response: For the equivalent analyses of nocebo responses on day 8, the following variables were included in step 1: the difference in pain intensity ratings between the nocebo and the control condition (i) during conditioning and (ii) on day 1. In step 2, we entered the variables reflecting nocebo expectations including expected worsening with nocebo (i) before conditioning, (ii) before the test session on day 1 and (iii) before the test session on day 8. In step 1, the model significantly predicted the day 8 nocebo response, F(3, 95) = 6.04, p = .003, accounting for 11.3% of the variance (R² = .113, Adjusted R² = .094). However, the addition of expectation-related predictors in Step 2 resulted in only a negligible and non-significant improvement, ΔR² = .006, F(3, 92) = 0.215, p = .886. The full model explained just 11.9% of the variance (R² = .119).

Typos:

(6) Abstract - 104 heathy xxx (word missing).

(7) Line 61 - reduce or decrease - I think you meant increase.

Thank you, we have now corrected both sentences.

References

Colloca L, Petrovic P, Wager TD, Ingvar M, Benedetti F. How the number of learning trials affects placebo and nocebo responses. Pain. 2010

Doering BK, Nestoriuc Y, Barsky AJ, Glaesmer H, Brähler E, Rief W. Is somatosensory amplification a risk factor for an increased report of side effects? Reference data from the German general population. J Psychosom Res. 2015

eLife Assessment

This work describes a highly complex automated algorithm for analyzing vascular imaging data from two-photon microscopy. This tool has the potential to be extremely valuable to the field and to fill gaps in knowledge of hemodynamic activity across a regional network. The solid biological application provides a demonstration of their pipeline's capabilities and suggests intriguing hypotheses around prolonged vascular tone changes, but will need to be followed up by further experiments to be conclusively demonstrated.

Reviewer #1 (Public Review):

Summary:

In this manuscript, the authors describe a new pipeline to measure changes in vasculature diameter upon opt-genetic stimulation of neurons.

The work is interesting and the topic is quite relevant to better understand the hemodynamic response on the graph/network level.

Strengths:

The manuscript provides a pipeline that allows for the detection of changes in the vessel diameter as well as simultaneously allowing for the location of the neurons driven by stimulation.

The resulting data could provide interesting insights into the graph-level mechanisms of regulating activity-dependent blood flow.

The interesting findings include that vessel radius changes depend on depth from the cortical surface and that dilations on average happen closer to the activated neurons.

Reviewer #2 (Public Review):

Summary:

The authors develop a highly detailed pipeline to analyze hemodynamic signals from in vivo two-photon fluorescence microscopy. This includes motion correction, segmentation of the vascular network, diameter measurements across time, mapping neuronal position relative to the vascular network, and analyzing vascular network properties (interactions between different vascular segments). For the segmentation, the authors use a Convolution Neural Network to identify vessel (or neural) and background pixels and train it using ground truth images based on semi-automated mapping followed by human correction/annotation. Considerable processing was done on the segmented images to improve accuracy, extract vessel center lines, and compute frame-by-frame diameters. The model was tested with artificial diameter increases and Gaussian noise and proved robust to these manipulations.

Network-level properties include Assortativity - a measure of how similar a vessel's response is to nearby vessels - and Efficiency - the ease of flow through the network (essentially, the combined resistance of a path based on diameter and vessel length between two points).

Strengths:

This is a very powerful tool for cerebral vascular biologists as many of these tasks are labor intensive, prone to subjectivity, and often not performed due to the complexity of collecting and managing volumes of vascular signals. Modelling is not my specialty so I cannot speak too specifically, but the model appears to be well-designed and robust to perturbations. It has many clever features for processing the data.

The authors rightly point out that there is a real lack in the field of knowledge of vascular network activity at single-vessel resolution. Network anatomy has been studied, but hemodynamics are typically studied either with coarse resolution or in only one or a few vessels at a time. This pipeline has the potential to change that.

[Editors' note: this work has been through three rounds of revisions, and most recently the authors have added caveats to the discussion. This version of the paper has been assessed by the editors and the weaknesses identified previously remain with earlier versions of the work.]

Author response:

The following is the authors’ response to the previous reviews

Reviewer #1 (Public review):

Summary:

In the manuscript the authors describe a new pipeline to measure changes in vasculature diameter upon optogenetic stimulation of neurons. The work is useful to better understand the hemodynamic response on a network /graph level.

Strengths:

The manuscript provides a pipeline that allows to detect changes in the vessel diameter as well as simultaneously allows to locate the neurons driven by stimulation.

The resulting data could provide interesting insights into the graph level mechanisms of regulating activity dependent blood flow.

Weaknesses:

(1) The manuscript contains (new) wrong statements and (still) wrong mathematical formulas.

The symbols in these formulas have been updated to disambiguate them, and the accompanying statements have been adjusted for clarity.

(2) The manuscript does not compare results to existing pipelines for vasculature segmentation (opensource or commercial). Comparing performance of the pipeline to a random forest classifier (illastik) on images that are not preprocessed (i.e. corrected for background etc.) seems not a particularly useful comparison.

We’ve now included comparisons to Imaris (a commercial) for segmentation and VesselVio (open-source) for graph extraction software.

For the ilastik comparison, the images were preprocessed prior to ilastik segmentation, specifically by doing intensity normalization.

Example segmentations utilizing Imaris have now been included. Imaris leaves gaps and discontinuities in the segmentation masks, as shown in Supplementary Figure 10. The Imaris segmentation masks also tend to be more circular in cross-section despite irregularities on the surface of the vessels observable in the raw data and identified in manual segmentation. This approach also requires days to months to generate per image stack.

A comparison to VesselVio has now also been generated, and results are visualized in Supplementary Figure 11. VesselVio generates individual graphs for each time point, resulting in potential discrepancies in the structure of the graphs from different time points. Furthermore, Vesselvio uses distance transformation to estimate the vascular radius, which renders the vessel radius estimates highly susceptible to variation in the user selected methodology used to obtain segmentation results; while our approach uses intensity gradient-based boundary detection from centerlines in the image instead mitigating this bias. We have added the following paragraph to the Discussion section on the comparisons with the two methods:

“Comparison with commercial and open-source vascular analysis pipelines

To compare our results with those achievable on these data with other pipelines for segmentation and graph network extraction, we compared segmentation results qualitatively with Imaris version 9.2.1 (Bitplane) and vascular graph extraction with VesselVio [1]. For the Imaris comparison, three small volumes were annotated by hand to label vessels. Example slices of the segmentation results are shown in Supplementary Figure 10. Imaris tended to either over- or under-segment vessels, disregard fine details of the vascular boundaries, and produce jagged edges in the vascular segmentation masks. In addition to these issues with segmentation mask quality, manual segmentation of a single volume took days for a rater to annotate. To compare to VesselVio, binary segmentation masks (one before and one after photostimulation) generated with our deep learning models were loaded into VesselVio for graph extraction, as VesselVio does not have its own method for generating segmentation masks. This also facilitates a direct comparison of the benefits of our graph extraction pipeline to VesselVio. Visualizations of the two graphs are shown in Supplementary Figure 11. Vesselvio produced many hairs at both time points, and the total number of segments varied considerably between the two sequential stacks: while the baseline scan resulted in 546 vessel segments, the second scan had 642 vessel segments. These discrepancies are difficult to resolve in post-processing and preclude a direct comparison of individual vessel segments across time. As the segmentation masks we used in graph extraction derive from the union of multiple time points, we could better trace the vasculature and identify more connections in our extracted graph. Furthermore, VesselVio relies on the distance transform of the user supplied segmentation mask to estimate vascular radii; consequently, these estimates are highly susceptible to variations in the input segmentation masks.We repeatedly saw slight variations between boundary placements of all of the models we utilized (ilastik, UNet, and UNETR) and those produced by raters. Our pipeline mitigates this segmentation method bias by using intensity gradient-based boundary detection from centerlines in the image (as opposed to using the distance transform of the segmentation mask, as in VesselVio).”

(3) The manuscript does not clearly visualize performance of the segmentation pipeline (e.g. via 2d sections, highlighting also errors etc.). Thus, it is unclear how good the pipeline is, under what conditions it fails or what kind of errors to expect.

On reviewer’s comment, 2D slices have been added in the Supplementary Figure 4.

(4) The pipeline is not fully open-source due to use of matlab. Also, the pipeline code was not made available during review contrary to the authors claims (the provided link did not lead to a repository). Thus, the utility of the pipeline was difficult to judge.

All code has been uploaded to Github and is available at the following location: https://github.com/AICONSlab/novas3d

The Matlab code for skeletonization is better at preserving centerline integrity during the pruning of hairs from centerlines than the currently available open-source methods.

- Generalizability: The authors addressed the point of generalizability by applying the pipeline to other data sets. This demonstrates that their pipeline can be applied to other data sets and makes it more useful.  However, from the visualizations it's unclear to see the performance of the pipeline, where the pipelines fails etc. The 3d visualizations are not particularly helpful in this respect . In addition, the dice measure seems quite low, indicating roughly 20-40% of voxels do not overlap between inferred and ground truth. I did not notice this high discrepancy earlier. A thorough discussion of the errors appearing in the segmentation pipeline would be necessary in my view to better assess the quality of the pipeline.

2D slices from the additional datasets have been added in the Supplementary Figure 13 to aid in visualizing the models’ ability to generalize to other datasets.

The dice range we report on (0.7-0.8) is good when compared to those (0.56-86) of 3D segmentations of large datasets in microscopy [2], [3], [4], [5], [6]. Furthermore, we had two additional raters segment three images from the original training set. We found that the raters had a mean inter class correlation  of 0.73 [7]. Our model outperformed this Dice score on unseen data: Dice scores from our generalizability tests on C57 mice and Fischer rats on par or higher than this baseline.

Reviewer #2 (Public review):

The authors have addressed most of my concerns sufficiently. There are still a few serious concerns I have. Primarily, the temporal resolution of the technique still makes me dubious about nearly all of the biological results. It is good that the authors have added some vessel diameter time courses generated by their model. But I still maintain that data sampling every 42 seconds - or even 21 seconds - is problematic. First, the evidence for long vascular responses is lacking. The authors cite several papers:

Alarcon-Martinez et al. 2020 show and explicitly state that their responses (stimulus-evoked) returned to baseline within 30 seconds. The responses to ischemia are long lasting but this is irrelevant to the current study using activated local neurons to drive vessel signals.

Mester et al. 2019 show responses that all seem to return to baseline by around 50 seconds post-stimulus.

In Mester et al. 2019, diffuse stimulations with blue light showed a return to baseline around 50 seconds post-stimulus (cf. Figure 1E,2C,2D). However, focal stimulations where the stimulation light is raster scanned over a small region focused in the field of view show longer-lasting responses (cf. Figure 4) that have not returned to baseline by 70 seconds post-stimulus [8]. Alarcon-Martinez et al. do report that their responses return baseline within 30 seconds; however, their physiological stimulation may lead to different neuronal and vessel response kinetics than those elicited by the optogenetic stimulations as in current work.

O'Herron et al. 2022 and Hartmann et al. 2021 use opsins expressed in vessel walls (not neurons as in the current study) and directly constrict vessels with light. So this is unrelated to neuronal activity-induced vascular signals in the current study.

We agree that optogenetic activation of vessel-associated cells is distinct from optogenetic activation of neurons, but we do expect the effects of such perturbations on the vasculature to have some commonalities.

There are other papers including Vazquez et al 2014 (PMID: 23761666) and Uhlirova et al 2016 (PMID: 27244241) and many others showing optogenetically-evoked neural activity drives vascular responses that return back to baseline within 30 seconds. The stimulation time and the cell types labeled may be different across these studies which can make a difference. But vascular responses lasting 300 seconds or more after a stimulus of a few seconds are just not common in the literature and so are very suspect - likely at least in part due to the limitations of the algorithm.

The photostimulation in Vazquez et al. 2014 used diffuse photostimulation with a fiberoptic probe similar to Mester et al. 2019 as opposed to raster scanning focal stimulation we used in this study and in the study by Mester et al. 2019  where we observed the focal photostimulation to elicited longer than a minute vascular responses. Uhlirova et al. 2016 used photostimulation powers between 0.7 and 2.8 mW, likely lower than our 4.3 mW/mm² photostimulation. Further, even with focal photostimulation, we do see light intensity dependence of the duration of the vascular responses. Indeed, in Supplementary Figure 2, 1.1 mW/mm² photostimulation leads to briefer dilations/constrictions than does 4.3 mW/mm²; the 1.1 mW/mm² responses are in line, duration wise, with those in Uhlirova et al. 2016.

Critically, as per Supplementary Figure 2, the analysis of the experimental recordings acquired at 3-second temporal resolution did likewise show responses in many vessels lasting for tens of seconds and even hundreds of seconds in some vessels.

Another major issue is that the time courses provided show that the same vessel constricts at certain points and dilates later. So where in the time course the data is sampled will have a major effect on the direction and amplitude of the vascular response. In fact, I could not find how the "response" window is calculated. Is it from the first volume collected after the stimulation - or an average of some number of volumes? But clearly down-sampling the provided data to 42 or even 21 second sampling will lead to problems. If the major benefit to the field is the full volume over large regions that the model can capture and describe, there needs to be a better way to capture the vessel diameter in a meaningful way.

In the main experiment (i.e. excluding the additional experiments presented in the Supplementary Figure 2 that were collected over a limited FOV at 3s per stack), we have collected one stack every 42 seconds. The first slice of the volume starts following the photostimulation, and the last slice finishes at 42 seconds. Each slice takes ~0.44 seconds to acquire. The data analysis pipeline (as demonstrated by the Supplementary Figure 2) is not in any way limited to data acquired at this temporal resolution and - provided reasonable signal-to-noise ratio (cf. Figure 5) - is applicable, as is, to data acquired at much higher sampling rates.

It still seems possible that if responses are bi-phasic, then depth dependencies of constrictors vs dilators may just be due to where in the response the data are being captured - maybe the constriction phase is captured in deeper planes of the volume and the dilation phase more superficially. This may also explain why nearly a third of vessels are not consistent across trials - if the direction the volume was acquired is different across trials, different phases of the response might be captured.

Alternatively, like neuronal responses to physiological stimuli, the vascular responses elicited by increases in neuronal activity may themselves be variable in both space and time.

I still have concerns about other aspects of the responses but these are less strong. Particularly, these bi-phasic responses are not something typically seen and I still maintain that constrictions are not common. The authors are right that some papers do show constriction. Leaving out the direct optogenetic constriction of vessels (O'Herron 2022 & Hartmann 2021), the Alarcon-Martinez et al. 2020 paper and others such as Gonzales et al 2020 (PMID: 33051294) show different capillary branches dilating and constricting. However, these are typically found either with spontaneous fluctuations or due to highly localized application of vasoactive compounds. I am not familiar with data showing activation of a large region of tissue - as in the current study - coupled with vessel constrictions in the same region. But as the authors point out, typically only a few vessels at a time are monitored so it is possible - even if this reviewer thinks it unlikely - that this effect is real and just hasn't been seen.

Uhlirova et al. 2016 (PMID: 27244241) observed biphasic responses in the same vessel with optogenetic stimulation in anesthetized and unanesthetized animals (cf Fig 1b and Fig 2, and section “OG stimulation of INs reproduces the biphasic arteriolar response”). Devor et al. (2007) and Lindvere et al. (2013) also reported on constrictions and dilations being elicited by sensory stimuli.

I also have concerns about the spatial resolution of the data. It looks like the data in Figure 7 and Supplementary Figure 7 have a resolution of about 1 micron/pixel. It isn't stated so I may be wrong. But detecting changes of less than 1 micron, especially given the noise of an in vivo prep (brain movement and so on), might just be noise in the model. This could also explain constrictions as just spurious outputs in the model's diameter estimation. The high variability in adjacent vessel segments seen in Figure 6C could also be explained the same way, since these also seem biologically and even physically unlikely.

Thank you for your comment. To address this important issue, we performed an additional validation experiment where we placed a special order of fluorescent beads with a known diameter of 7.32 ± 0.27um, imaged them following our imaging protocol, and subsequently used our pipeline to estimate their diameter. Our analysis converged on the manufacturer-specified diameters, estimating the diameter to be 7.34 ± 0.32. The manuscript has been updated to detail this experiment, as below:

Methods section insert

“Second, our boundary detection algorithm was used to estimate the diameters of fluorescent beads of a known radius imaged under similar acquisition parameters. Polystyrene microspheres labelled with Flash Red (Bangs Laboratories, inc, CAT# FSFR007) with a nominal diameter of 7.32um and a specified range of 7.32 ± 0.27um as determined by the manufacturer using a Coulter counter were imaged on the same multiphoton fluorescence microscope set-up used in the experiment (identical light path, resonant scanner, objective, detector, excitation wavelength and nominal lateral and axial resolutions, with 5x averaging). The images of the beads had a higher SNR than our images of the vasculature, so Gaussian noise was added to the images to degrade the SNR to the same level of that of the blood vessels. The images of the beads were segmented with a threshold, centroids calculated for individual spheres, and planes with a random normal vector extracted from each bead and used to estimate the diameter of the beads. The same smoothing and PSF deconvolution steps were applied in this task. We then reported the mean and standard deviation of the distribution of the diameter estimates. A variety of planes were used to estimate the diameters.”

Results Section Insert

“Our boundary detection algorithm successfully estimated the radius of precisely specified fluorescent beads. The bead images had a signal-to-noise ratio of 6.79 ± 0.16 (about 35% higher than our in vivo images): to match their SNR to that of in vivo vessel data, following deconvolution, we added Gaussian noise with a standard deviation of 85 SU to the images, bringing the SNR down to 5.05 ± 0.15. The data processing pipeline was kept unaltered except for the bead segmentation, performed via image thresholding instead of our deep learning model (trained on vessel data). The bead boundary was computed following the same algorithm used on vessel data: i.e., by the average of the minimum intensity gradients computed along 36 radial spokes emanating from the centreline vertex in the orthogonal plane. To demonstrate an averaging-induced decrease in the uncertainty of the bead radius estimates on a scale that is finer than the nominal resolution of the imaging configuration, we tested four averaging levels in 289 beads. Three of these averaging levels were lower than that used on the vessels, and one matched that used on the vessels (36 spokes per orthogonal plane and a minimum of 10 orthogonal planes per vessel). As the amount of averaging increased, the uncertainty on the diameter of the beads decreased, and our estimate of the bead's diameter converged upon the manufacturer's Coulter counter-based specifications (7.32 ± 0.27um), as tabulated below in Table 1.”

Bibliography

(1) J. R. Bumgarner and R. J. Nelson, “Open-source analysis and visualization of segmented vasculature datasets with VesselVio,” Cell Rep. Methods, vol. 2, no. 4, Apr. 2022, doi: 10.1016/j.crmeth.2022.100189.

(2) G. Tetteh et al., “DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes,” Front. Neurosci., vol. 14, Dec. 2020, doi: 10.3389/fnins.2020.592352.

(3) N. Holroyd, Z. Li, C. Walsh, E. Brown, R. Shipley, and S. Walker-Samuel, “tUbe net: a generalisable deep learning tool for 3D vessel segmentation,” Jul. 24, 2023, bioRxiv. doi: 10.1101/2023.07.24.550334.

(4) W. Tahir et al., “Anatomical Modeling of Brain Vasculature in Two-Photon Microscopy by Generalizable Deep Learning,” BME Front., vol. 2020, p. 8620932, Dec. 2020, doi: 10.34133/2020/8620932.

(5) R. Damseh, P. Delafontaine-Martel, P. Pouliot, F. Cheriet, and F. Lesage, “Laplacian Flow Dynamics on Geometric Graphs for Anatomical Modeling of Cerebrovascular Networks,” ArXiv191210003 Cs Eess Q-Bio, Dec. 2019, Accessed: Dec. 09, 2020. (Online). Available: http://arxiv.org/abs/1912.10003

(6) T. Jerman, F. Pernuš, B. Likar, and Ž. Špiclin, “Enhancement of Vascular Structures in 3D and 2D Angiographic Images,” IEEE Trans. Med. Imaging, vol. 35, no. 9, pp. 2107–2118, Sep. 2016, doi: 10.1109/TMI.2016.2550102.

(7) T. B. Smith and N. Smith, “Agreement and reliability statistics for shapes,” PLOS ONE, vol. 13, no. 8, p. e0202087, Aug. 2018, doi: 10.1371/journal.pone.0202087.

(8) J. R. Mester et al., “In vivo neurovascular response to focused photoactivation of Channelrhodopsin-2,” NeuroImage, vol. 192, pp. 135–144, May 2019, doi: 10.1016/j.neuroimage.2019.01.036.

eLife Assessment

This paper provides important insight into how early life experience shapes adult behavior in fruit bats. The authors raised juvenile bats either in an impoverished or enriched environment and studied their foraging behaviors. The evidence is convincing that bats raised in enriched environments are more active, bold, and exploratory. The work will be of interest to ethologists and developmental psychologists.

Reviewer #1 (Public review):

Summary:

The authors show that early life experience of juvenile bats shape their outdoor foraging behaviors. They achieve this by raising juvenile bats either in an impoverished or enriched environment. They subsequently test the behavior of bats indoors and outdoors. The authors show that behavioral measures outdoors were more reliable in delineating the effect of early life experiences as the bats raised in enriched environments were more bold, active and exhibit higher exploratory tendencies.

Strengths:

The major strength of the study is providing a quantitative study of animal "personality" and how it is likely shaped by innate and environmental conditions. The other major strength is the ability to do reliable long term recording of bats in the outdoors giving researchers the opportunity to study bats in their natural habitat. To this point, the study also shows that the behavioral variables measured indoors do not correlate to that measured outdoor, thus providing a key insight into the importance of test animal behaviors in their natural habitat.

Weaknesses were in the first round of review:

It is not clear from the analysis presented in the paper how persistent those environmentally induced changes, do they remain with the bats till end of their lives.

Comments on revisions:

The authors have addressed those weaknesses and the paper is much stronger.

Author response:

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

Reviewer #1 (Public Reviewer):

It is not clear from the analysis presented in the paper how persistent those environmentally induced changes, do they remain with the bats till the end of their lives.

Currently, the long-term effects of enrichment on the bats remain uncertain. Preliminary results suggest that these differences may persist throughout the bats’ lifetimes; however, further data analysis is ongoing to determine the extent of these effects. We also addressed now at the manuscript discussion

Reviewer #2 (Public Reviewer):

(1) Assessing personality metrics and the indoor paradigm: While I applaud this effort and think the metrics used are justified, I see a few issues in the results as they are currently presented:

(a) [Major] I am somewhat concerned that here, the foraging box paradigm is being used for two somewhat conflicting purposes: (1) assessing innate personality and (2) measuring changes in personality as a result of experience. If the indoor foraging task is indeed meant to measure and reflect both at the same time, then perhaps this can be made more explicit throughout the manuscript. In this circumstance, I think the authors could place more emphasis on the fact that the task, at later trials/measurements, begins to take on the character of a "composite" measure of personality and experience.

Personality traits should generally be stable over time, but personality can also somewhat change with experience. We used the foraging box to assess individual personality, but we also examined the assumption that what we are measuring is a proxy of personality and hence is stable over time. We now clarify this in the manuscript. 

(b) [Major] Although you only refer to results obtained in trials 1 and 2 when trying to estimate "innate personality" effects, I am a little worried that the paradigm used to measure personality, i.e. the stable components of behavior, is itself affected by other factors such as age (in the case of activity, Fig. 1C3, S1C1-2), the environment (see data re trial 3), and experience outdoors (see data re trials 4/5).

We found that boldness was the most consistent trait, showing persistence between trials 1 to 5, i.e., 144 days apart on average. We thus also used Boldness as the primary parameter for assessing the effects of personality on the outdoors behavior. While we evaluated other traits for completeness, boldness was the only one that consistently met the criteria for personality, which is why we focused on it in our analyses. The other traits which were not stable over time could be used to assess the effects of experience on behavior

Ideally, a study that aims to disentangle the role of predisposition from early-life experience would have a metric for predisposition that is relatively unchanging for individuals, which can stand as a baseline against a separate metric that reflects behavioral differences accumulated as a result of experience.

I would find it more convincing that the foraging box paradigm can be used to measure personality if it could be shown that young bats' behavior was consistent across retests in the box paradigm prior to any environmental exposure across many baseline trials (i.e. more than 2), and that these "initial settings" were constant for individuals. I think it would be important to show that personality is consistent across baseline trials 1 and 2. This could be done, for example, by reproducing the plots in Fig. 1C1-3 while plotting trial 1 against trial 2. (I would note here that if a significant, positive correlation were to be found (as I would expect) between the measures across trial 1 and 2, it is likely that we would see the "habituation effect" the authors refer to expressed as a steep positive slope on the correlation line (indicating that bold individuals on trial 1 are much bolder on trial 2).)

We agree and thus used boldness which was found to be stable over five trials (three of which were without external experience). We note that if Boldness as we measured it increased over time, the differences between individuals remained similar and this is what is expected from personality traits measured in the same paradigm several times (after the animal acquires experience).  

(c) Related to the previous point, it was not clear to me why the data from trial 2 (the second baseline trial) was not presented in the main body of the paper, and only data from trial 1 was used as a baseline.

We added a main figure, showing the correlation between the two baseline trials

In the supplementary figure and table, you show that the bats tended to exhibit more boldness and exploratory behavior, but fewer actions, in trial 2 as compared with trial 1. You explain that this may be due to habituation to the experimental setup, however, the precise motivation for excluding data from trial 2 from the primary analyses is not stated. I would strongly encourage the authors to include a comparison of the data between the baseline trials in their primary analysis (see above), combine the information from these trials to form a composite baseline against which further analyses are performed, or further justify the exclusion of data as a baseline.

We had no intention of excluding data from baseline 2. As we have shown several times before (e.g., Harten, 2021) bats’ boldness as we measure it in the box experiment increases over sessions performed nearby in time. This means that trial 2’s boldness was higher than that of trial 1 and trial 3 which made the data less suitable for a Linear model. Moreover, our measurement of boldness is capped (with a maximum of 1) again making it less suitable for a Linear model. However, following the reviewer’s question we now ran all analyses with trial 2’s data included and not only that the results remained the same, some of the models fit better (based on the AIC criterion). We added this information to the revised manuscript.  

(2) Comparison of indoor behavioral measures and outdoor behavioral measures Regarding the final point in the results, correlation between indoor personality on Trial 4 and outdoor foraging behavior: It is not entirely clear to me what is being tested (neither the details of the tests nor the data or a figure are plotted). Given some of the strong trends in the data - namely, (1) how strongly early environment seems to affect outdoor behavior, (2) how strongly outdoor experience affects boldness, measured on indoor behavior (Fig. 1D) - I am not convinced that there is no relationship, as is stated here, between indoor and outdoor behavior. If this conclusion is made purely on the basis of a p-value, I would suggest revisiting this analysis.

We agree that the relationship between indoor personality measures and outdoor foraging behavior is of great interest and had expected to find some correspondence between the two. To test this, we conducted multiple GLM analyses using the different indoor behavioral traits as predictors of outdoor behaviors. These analyses did not reveal any significant correlations. We also performed a separate analysis using PC1 (derived from the indoor behavioral variables) as a predictor, and again found no significant associations with outdoor behavior.

We were indeed surprised by this outcome. It is possible that the behavioral traits we assessed indoors (boldness, exploration, and activity) do not fully capture the dimensions of behavior that are most relevant to foraging in the wild. For example, traits such as neophobia or decisionmaking under risk, which we did not assess directly, may have had stronger predictive value for outdoor behavior. We now highlight this point more clearly in the Discussion and acknowledge the possibility that alternative or additional personality traits might have revealed meaningful relationships.

(3) Use of statistics/points regarding the generalized linear models While I think the implementation of the GLMM models is correct, I am not certain that the interpretation of the GLMM results is entirely correct for cases where multivariate regression has been performed (Tables 4s and S1, and possibly Table 3). (You do not present the exact equation they used for each model (this would be a helpful addition to the methods), therefore it is somewhat difficult to evaluate if the following critique properly applies, however...)

The "estimate" for a fixed effect in a regression table gives the difference in the outcome variable for a 1 unit increase in the predictor variable (in the case of numeric predictors) or for each successive "level" or treatment (in the case of categorical variables), compared to the baseline, the intercept, which reflects the value of the outcome variable given by the combination of the first value/level of all predictors. Therefore, for example, in Table 4a - Time spend outside: the estimate for Bat sex: male indicates (I believe) the difference in time spent outside for an enriched male vs. an enriched female, not, as the authors seem to aim to explain, the effect of sex overall. Note that the interpretation of the first entry, Environmental condition: impoverished, is correct. I refer the authors to the section "Multiple treatments and interactions" on p. 11 of this guide to evaluating contrasts in G/LMMS: https://bbolker.github.io/mixedmodelsmisc/notes/contrasts.pdf

We are not certain we fully understand the comment; however, if our understanding is correct, we respectfully disagree. A GLM analysis without interaction terms—as conducted in our study—functions as a multiple linear regression, wherein each factor's estimate reflects its individual effect on the dependent variable. For example in the case of sex, it examines he effect of sex on the tie spent out independently of enrichment. An interaction term would be needed to test sex*enrichment. We have added the models’ formula, and we hope this clarifies our approach

Reviewer #1 (Recommendations for the authors):

I would recommend the following:

(1) As video tracking and behavioral analysis softwares are wide spread, it would be great to see this applied to the bat behavior indoor to answer questions like how does the bat velocity or heading or acceleration correlate with the behavioral measures boldness , activity or exploration? In the same gist, can one infer boldness, activity or exploration from measured bat velocity or other parameters? I think this will further make the indoor behavior more quantitative.

In a tent of the size used in our study, bats’ flight behavior tends to be highly stereotypical: they typically perch on the wall, take off, circle the tent—sometimes multiple times—and then either land or not, and enter or not. Flight velocity is largely determined by individual maneuverability and the physical constraints of the space; thus, precise tracking is unlikely to provide further insight into boldness. In contrast, decision-making behaviors—such as whether to land or enter—more accurately reflect personality traits, as we have shown previously (Harten et al., 2018). Moreover, accurate 3D tracking in such an environment is possible but definitely not easy due to the many blind-spots resulting from the cameras being inside the 3D volume.  Nonetheless, we quantified flight activity and assessed its correlation with the other behavioral axes. As it was highly correlated with general activity, we did not include it as an independent parameter in the main analysis. However, in response to the reviewer’s suggestion, we now present this analysis in the Supplementary Materials.

(2) It is not clear whether the bats come from the same genetic background. they might be but it is not mentioned in the methods under the experimental subjects.

We have shown in the past that there is no familial relations in a randomly caught sample of bats in the colony where we usually work (Harten et al., 2018). The bats were caught in three, not related wild colonies. The text referring to the table was clarified in the revised manuscript

(3) It will be great to include the author's thoughts about mechanisms underlying those environmentally induced changes in behavior in the discussion section along with how this will affect the bats' social foraging abilities. Another question that comes to mind is whether growing up with a large number of bats constitute an enriched environment in itself.

We agree that this could count as an enrichment, and we thus ensured similar group sizes in both groups for this reason. We clarify this in the revised manuscript. 

We have elaborated on the underlying mechanisms in the discussion, focusing on how they contribute to behavioral changes.

Reviewer #2 (Recommendations for the authors):

(1) Outdoor foraging behavior

If I understand correctly, the data you display in Fig. 3A is only from the 2nd to 3rd weeks of exploration, i.e. just before the first post-exploration trial.

What does the data look like for the second outdoor exploration data, i.e. before the final trial?

Is there a specific reason why these measures were only computed on the GPS data from the 3rd week outside? If so, can this sampling of the data be motivated or briefly addressed (in the methods and wherever else necessary)?

In order to allow a comparison between individuals, we had to restrict ourself to a period we had data from many individuals (some dissapeared later on).

Following the reviewer suggestion – we added a supplemenry figure including days 21-26

I would find it important and of great interest to see movement maps for more animals, as these give very rich information that is not entirely captured by the three proxies of outdoor activity.

Are these four exemplary animals sampled from both seasons?

Did you check to see if there were any overall differences in outdoor foraging behavior as a function of the season in which the bats were captured?

Yes, the samples represent individuals from both tested years. This was clarified, and additional examples were included in a supplementary figure.

Variable of time spent outdoors: You mention that you did not include the nights that the bat spent in the colony in these calculations. Did you also look to see if 'the number of nights when the bats left the colony' predicted the bat's earlier enrichment treatment? This could also be interesting to consider.

In response to the reviewer’s comment, we conducted an additional analysis to test whether the proportion of nights each bat spent foraging outside the roost was predicted by its earlier environmental condition (enriched vs. impoverished). We also examined whether sex or age influenced this variable. This analysis showed no significant effect of environmental condition, sex, or age on the proportion of nights spent foraging outside the roost

[Following on point 3 in public review...]

When wishing to discuss the effect/significance of predictors overall, it is common to present the modelling results as an analysis of variance table. See, for example, the two-way anova section (p. 182) in the book Practical Regression and ANOVA using R: https://cran.r-project.org/doc/contrib/Faraway-PRA.pdf

I think the output of passing the model object to an "anova" yields the table that you may be looking for, where the variance accounted for by a predictor is given overall, and not just relative to the first level of all predictors. Naturally, this information can be used in combination with the information provided by the raw model output presented in the paper.

I assume you have done this analysis in R, but am not sure, as the statistical software used is not mentioned. There are several packages in R that allow users to quickly plot the graphical interaction of the parameters they use in models, which aids in interpreting results. It would be good to check results of model fitting in this manner.

Relatedly, I was unable to locate the data and code for this paper using the DOI provided. Neither searching the internet using the doi nor entering the doi on the Mendeley Data website returned the right results. I tried searching Mendeley Data using the senior author's last name, but the most recent entry does not appear to be from this paper. https://data.mendeley.com/datasets/fr48bmnhxj/1

We thank the reviewer for the helpful comment. The analysis was indeed conducted in MATLAB, and this has now been clarified in the manuscript. We have also revised the result tables to improve clarity and included the exact formulas used for each model. Regarding the data availability, the reviewer is correct — the dataset had not yet been published at the time of submission. It is now available at the provided DOI link.

### Suggestions and questions for the present paper, grouped thematically:

[Major] Expansion and development of results: I thought there were many interesting and suggestive points in this data that could be expanded upon. I mention some of these here. While the authors of course do not need to implement all of these suggestions, I think the paper would benefit from a more substantial presentation of this rich data set:

(a) Individual differences as such are not emphasized in the paper so much, as the analyses, particularly those expressed as boxplots, are grouped. The scatter plots in Figure 1 give the richest insight into how individual behavior changes throughout the course of the experiment. I would advocate for the authors to show additional comparisons using such scatter plots (perhaps in the supplementary, if needed).

We thank the reviewer and added scatter plots to figure 2

(b) In the second paragraph of the results, the authors introduce the concept of a pareto front and that of personality archetypes (lines 101-107). I found this very interesting, but these concepts were never reiterated upon later in the results or in the discussion. In fact, at many points, I found myself curious as to how the three indoor measures of personality might be combined to form a composite measure of personality (and likewise for outdoor measures). Have you tried to combine measures into a composite and tried to measure whether this composite metric provides any additional insight into these phenomena? For example, what if you mapped the starting position of each bat as a point in a three-dimensional space, given by the three personality measures, and then evaluated their trajectory through this space with measurements taken at later trials. Could innate personality be interpreted as the starting vector in this space (measured across the two baseline trials)? 

Following the reviewer’s (justified) curiosity we ran a PCA analysis on the behavioral data from trials 1 and 5 and found that there is a significant correlation between the individual scores on PC1. This can be thought of as a measurement that takes both boldness and exploration into account (the weight of activity was very low). We added this information to the revised manuscript and also use this new behavioral parameter as a predisposition in the models (instead of exploration and activity). 

Could environmental exposure be quantified as a warping of the trajectory through this space? Finally, could outdoor experience also be incorporated to evaluate how an individual arrives at its final measurement of personality combined with experience (trial 5)?

The paper currently tries to explain outdoors behavior given personality and not vice versa. While this is a very interesting suggestion, we feel that adding this analysis would make the premise of the paper less clear and since the paper is already somewhat complex, we prefer to leave this analysis for a future study. 

Examining the 3D trajectories of the individuals through the personality space did not reveal any immediate clear pattern (triangles mark the first trial and colours depict the environmental treatment) – 

Author response image 1.

Related to this point: I think the strongest part of the paper is the result showing that bats exposed to enriched environments explore farther, more often, and over larger distances than bats that were raised in an impoverished environment.

We completely agree and tried to further emphasize this  

(c) While these results of the outdoor GPS tracking are very clear, I wish that more information were extracted from the tracking data, which is incredibly rich and certainly can be used to derive many interest parameters beyond those that the authors have shown here. Examples might include: distance travelled (as opposed to estimated km2 or farthest point), a metric of navigational ability (how much "dead reckoning" the animal engages in). I even wonder if the areas or landmarks visited by the enriched bats might be found to be more complex, challenging, or richer by some measure.

This study was a first step, aiming to establish a connection between early exposure and outdoors foraging

We agree that there are many more analyses that can be done and indeed that ones related to navigation capabilities are missing. We are still collecting data on these bats and hope to present a more advanced analysis with a time span of years. 

(d) Related to the above point: I find it very interesting that in 3 of the 4 bats for which you show exemplary movement data (Fig. 3, panels B and C), they appear to travel to the farthest distances and cover the most ground early on, and become more "conservative" in their flight paths on later evenings. This point is not explored in the discussion, nor related to earlier measurements.

During the first months of exploration, bats will occasionally perform long exploratory flights in between bouts of shorter flights where they return to nearby familiar trees. This behavior can be seen in more detail in Harten et al Science 2020. We are currently quantifying this more carefully for another study. 

(e) Finally, my points about the possible strength of a composite measure of the three personality metrics is related to my concern about one of the conclusions, which is that innate personality does not have an effect on outdoor foraging behavior. I think the manner in which this was tested statistically is likely to bias the results against finding such a result given that personality metrics are used to predict outdoor behaviors in an individual manner (6 models in total, each examining a single comparison of predisposition to outdoor behavior), while both indoor personality metrics (Fig 1B) and outdoor behaviors appear to be correlated with each other (Table 5).

Are there other analyses you have performed that are not presented in the paper and that have led you to conclude that there is no relationship here?

We agree with the reviewer, that our findings do not exclude an effect of innate personality on foraging but only suggest no such affect for the parameter we measured. That said, we did expect to find an effect of boldness because this parameter has been shown to differentiate much between groups (Harten et al., 2018), and to correlate with other parameters of behavior. We were therefore surprised to find no significant effects, as we had anticipated observing some differences.

Following the reviewer’s previous comment we now also tested another predisposition parameter – the PC1 score and also found that it did not explain foraging. 

(f) Personality measured before and after early environmental exposure (related to point (a) above): I find it interesting that the positive correlation in boldness between baseline and post-enrichment or baseline and post-release suggests that the individuals that were the most bold remained bold (and likewise for less adventurous individuals). The correlation for activity, too, still suggests that more active individuals early in life are likely to remain very active after enrichment, even accounting for the fact that activity is confounded with age.

Perhaps you could place some emphasis on the fact that the initial variation between individuals also appears to be relatively stable over repeated trials. You might also consider measuring this directly (population variance over successive trials; relationship of population variance on indoor measures vs. outdoor measures...)

Yes – this is a main point of interest. We further emphasize that in the revised manuscript 

(g) Effect of indoor behavior following early experience on outdoor behavior: You evaluate the effect of predisposition (measured on baseline trial 1) and environmental condition on measures of outdoor activity (Table 4). I wonder if you also tried using indoor behavioral measures measured on the post-enrichment trial 3 to predict outdoor foraging behavior.

Assuming that these measures are in fact reflecting a combination of predisposition and accumulated experience, then measurements at this closer time point may tell you how the combination of innate traits and early acquired experience affect behavior in the wild.

We appreciate the reviewer’s insightful suggestion to test whether indoor behavior from post-enrichment Trial 3, reflecting both innate traits and experience, predicts outdoor foraging behavior. We conducted this analysis, but found that the boldness in Trial 3 did not significantly predict any of the outdoor activity measures.

(2) [Minor] Age/development: While the authors discuss the effect of their manipulations on behavioral measures, they do not much discuss the effect of age.

I think it would be important to include at some point a mention of the developmental stages of Rousettus, giving labels to certain age ranges, e.g. pup, juvenile, adult, and to provide more context about the stages at which bats were tested in the discussion. Presently, age is only really mentioned as an explanation for declining activity levels, but I wonder if it might also have an influence on boldness.

It would also be very elegant for figures where age is given in days, to additional label then with these stages.

All bats were juveniles during the trials (approximately 4 to 8 months old), so they could not be divided into distinct age groups. To assess the effect of age, it was included as a predictor (in days) in the GLM analysis.

(3) [Major] Effect of early experience and outdoor experience on the indoor task: In the paragraph on lines 278-285, you argue that the effect of seeing earlyenriched bats exhibit more boldness in trial 5 was likely due to post-sampling bias...

I tend to disagree with this conclusion. I actually find this result both interesting and intuitive - that bats that were exposed to an enriched environment and have had experience in the wild, show much bolder activity on a familiar indoor foraging test (i.e. outside experience has made the animals bolder than before) (Fig 1, lines 159-161, Fig. S1). I did not notice this possibility mentioned in the discussion of the results.

I also do not fully understand this argument. Could you please explain further?

We accept the reviewer's comment and updated the manuscript (lines 336346) explaining the two hypotheses more clearly and arguing that it is difficult to tell them apart with the current data.

[Minor] You also say that "this difference... can be seen in Figure 2 when examining only the bats that had remained until the last trial (Figure 2A2)." Do you mean supplementary Figure S1 A2? In fact, I am entirely unclear on what data is plotted in the supplementary Figure S1 and what differentiates the two columns of figures and the two models presented in the supplementary table. Did you plot data similar to that in Figure 2, with only bats that were present for all trials, but not show this data?

There was a mistake: what was previously referred to as 2A2 is actually S2 A2.

On the right side—only among the individuals with GPS data—the change is already evident at Baseline 2, where only the bolder individuals remain. If you have suggestions for a better analysis approach, we would be happy to hear them.

### Minor points

General points regarding figures:

For Figures 2 and 3A1-3 (as well as Fig. S1): Authors must show the raw data points over the box plots. It is very difficult to interpret the data and conclusions without being able to see the true distribution.

Done

For all figures showing grouped individual data, please annotate all panels or sets of boxplots with the number of bats whose data entered into each, as it is a little difficult to keep track of the changing sample sizes across experimental stages.

To enhance transparency, we have added individual data points to all boxplots, allowing visual estimation of sample sizes across experimental stages. While numerical annotations are not included on the figures, the exact number of bats contributing to each group is provided in the Methods section (Table 8), ensuring this information is readily accessible to readers.In response to the reviewer’s request, we have updated all relevant figures to display individual data points within each boxplot. This addition makes it easier to track changes in sample size across different experimental stages.

Unless I've missed the reason behind differences in axis labelling across the figures, it seems that trials are not always referred to consistently. E.g. Fig. 1 labels say "Trial 1 (baseline)" and fig. 2 labels say "Baseline 1 0 days." I'm not entirely sure if these correspond to exactly the same data. If so, perhaps the labels can be made uniform. I think the descriptive ones (Baseline 1, Postenrichment...) may be more helpful to the reader than providing the trial number (Trial 1, etc....).

Done

Figure 1:

Very good Fig. 1A and 1B.

For panels C1-3 & D, I think it would make it easier for the reader if the personality measure labels were placed at the top of each panel, e.g. "Boldness (entrance proportion)". The double axis labels are not only harder to read, they are also redundant, as the personality measure label repeats on both axes.

Done

Panel C1: For the first panel in this sequence, I think it would be elegant to include an annotation in the figure that indicates what the datapoints lying on either side of the dashed line means, i.e. "bolder after enrichment treatment" in the upper left corner, and "bolder before enrichment treatment" in the bottom right corner.

Panel C2: It appears as though many of the data points in this panel overlap, and it appears to me that the blue data points in particular are overlaid by the orange ones. I am guessing this happens because proportion values based on entrances to only 6 boxes end up giving a more "discrete" looking distribution. I wonder if you can find a way to allow all the data to be visible by, e.g., jittering the data slightly; if there is rounding being done to the proportions, perhaps don't round them so that minute differences will allow them to escape the overlap; or possibly split the panel by enrichment treatment.

Caption for C1-3: it may be helpful to mention the correlation line color scheme: "enriched (blue lines), the impoverished (orange lines)". The caption also says positive correlations were found for "both environments together," but this correlation line is not shown. Perhaps mention "(not shown)" or show line. Please rephrase the sentence "Dashed line represents the Y=X line." for more transparency and clarity. I understand you mean an "equality" or "unity" line, but perhaps you can explicitly state the information that this line provides, something like e.g. "Dashed line indicates equal values measured on both trials."

We added the line for a reference, the caption was corrected

Figure 3:

Panels B1-C2: I would suggest giving these panels supertitles that indicate that B panels are enriched, C panels are impoverished, and that each panel is data from a different individual.

The legend was corrected to be more clear about the figure

General points regarding tables:

Please revisit tables for formatting and typos, particularly in Table 4. Please also revise table captions for clarity. E.g. "first exploration as predisposition" to "Exploration (Baseline 1)" or similar

Done

Supplementary Tables and Figure: these are missing captions and explanations.

The missing parts were adddad and corrected

Points of clarification/style:

It would seem to me more logical to present the results shown in Table 3 before those in Table 2, given that the primary in-lab manipulation is discussed with relation to Table 3, and the analysis in Table 2 is discussed rather as a limitation (though I believe this result can be expanded upon further, see above).

For the activity metric, I would suggest showing this data as actions/hour instead of actions/minute. I think it is much more intuitive to consider, for example, that a bat makes 2 actions every hour, than that it makes 0.002 actions per minute.

Done

eLife Assessment

This revised manuscript presents an important characterization of mouse auditory cortex receptive field organization, utilizing two-photon imaging of specific subpopulations. They demonstrate a degradation of tonotopic organization from the input to the output neurons. The strength of the evidence is convincing.

Reviewer #1 (Public review):

Summary:

In this study, Gu et al., employed novel viral strategies, combined with in vivo two-photon imaging, to map the tone response properties of two groups of cortical neurons in A1 - The thalamocortical recipient (TR neurons) and the corticothalamic (CT neurons). They observed a clear tonotopic gradient among TR neurons but not in CT neurons. Moreover, CT neurons exhibited high heterogeneity of their frequency tuning and broader bandwidth, suggesting increased synaptic integration in these neurons. By parsing out different projecting-specific neurons within A1, this study provides insight into how neurons with different connectivity can exhibit different frequency response-related topographic organization.

Strengths:

This study reveals the importance of studying neurons with projection specificity rather than layer specificity since neurons within the same layer have very diverse molecular, morphological, physiological, and connectional features. By utilizing a newly developed rabies virus CSN-N2c GCaMP-expressing vector, the authors can label and image specifically the neurons (CT neurons) in A1 that project to the MGB. To compare, they used an anterograde trans-synaptic tracing strategy to label and image neurons in A1 that receive input from MGB (TR neurons).

Weaknesses:

- Perhaps as cited in the introduction, it is well known that tonotopic gradient is well preserved across all layers within A1, but I feel if the authors want to highlight the specificity of their virus tracing strategy and the populations that they imaged in L2/3 (TR neurons) and L6 (CT neurons), they should perform control groups where they image general excitatory neurons in the two depths and compare to TR and CT neurons, respectively. This will show that it's not their imaging/analysis or behavioral paradigms that are different from other labs.  

- Fig 1D and G, the y-axis is Distance from pia (%). I'm not exactly sure what this means. How does % translate to real cortical thickness? 

- For Fig. 2G and H, is each circle a neuron or an animal? Why are they staggered on top of each other on the x-axis? If x-axis is thedistance from caudal to rostral, each neuron should have a different distance? Also,it seems like it's because Fig. 2H has more circles, that's why it has morevariation thus not significant (for example, at 600 or 900um, 2G seems to haveless circles than 2H).  

- Similar in Fig 2J and L, why are the circles staggered onthe y-axis now? And is each circle now a neuron or a trial? It seems they havemuch more circles than Fig 2G and 2H. Also I don't think doing a correlation isthe proper stats for this type of plot (this point applies to Fig. 3H and 3J)

- What does inter-quartile range of BF (IQRBF, in octaves) imply? What's the interpretation of this analysis? I am confused why TR neurons showhigh IQR in HF areas compared to LF areas mean homogeneity among TR neurons (line 213 - 216). On the same note, how is this different from the BF variability?  Isn't higher IQR = tohigher variability?

- Fig. 4A-B, there's no clear critieria on how the authors categorize V, I, and O Shape. The descriptions in the Methods (line 721 - 725) are also very vague.  

Comments on revisions:

The authors have addressed all my questions in the previous round.

Reviewer #2 (Public review):

Summary:

Gu and Liang et. al investigated how auditory information is mapped and transformed as it enters and exits a auditory cortex. They use anterograde transsynaptic tracers to label and perform calcium imaging of thalamorecipient neurons in A1 and retrograde tracers to label and perform calcium imaging of corticothalamic output neurons. They demonstrate a degradation of tonotopic organization from the input to output neurons.

Strengths:

The experiments appear well executed, well described, and analyzed.

Weaknesses:

(1) Given that the CT and TR neurons were imaged at different depths, the question as to whether not these differences could otherwise be explained by layer-specific differences is still not 100% resolved. Control measurements would be needed either by recording 1) CT neurons upper layers 2) TR in deeper layers 3) non-CT in deeper layers and/or 4) non-TR in upper layers.

(2) What percent of the neurons at the depths being are CT neurons? Similar questions for TR neurons?

(3) V-shaped, I-shaped, or O-shaped is not an intuitively understood nomenclature, consider changing. Further, the x/y axis for Figure 4a is not labeled, so it's not clear what the heat maps are supposed to represent.

(4). Many references about projection neurons and cortical circuits are based on studies from visual or somatosensory cortex. Auditory cortex organization is not necessarily the same as other sensory areas. Auditory cortex references should be used specifically, and not sources reporting on S1, V1.

Comments on revisions:

The authors have fully addressed my concerns.

Reviewer #3 (Public review):

Summary:

The authors performed wide-field and 2-photon imaging in vivo in awake head-fixed mice, to compare receptive fields and tonotopic organization in thalamocortical recipient (TR) neurons vs corticothalamic (CT) neurons of mouse auditory cortex. TR neurons were found in all cortical layers while CT neurons were restricted to layer 6. The TR neurons at nominal depths of 200-400 microns have a remarkable degree of tonotopy (as good if not better than tonotopic maps reported by multiunit recordings). In contrast, CT neurons were very heterogenous in terms of their best frequency (BF), even when focusing on the low vs high frequency regions of primary auditory cortex. CT neurons also had wider tuning.

Strengths:

This is a thorough examination using modern methods, helping to resolve a question in the field with projection-specific mapping.

Weaknesses:

There are some limitations due to the methods, and it's unclear what the importance of these responses are outside of behavioral context or measured at single timepoints given the plasticity, context-dependence, and receptive field 'drift' that can occur in cortex.

(1) Probably the biggest conceptual difficulty I have with the paper is comparing these results to past studies mapping auditory cortex topography, mainly due to differences in methods. Conventionally, tonotopic organization is observed for characteristic frequency maps (not best frequency maps), as tuning precision degrades and best frequency can shift as sound intensity increases. The authors used six attenuation levels (30-80 dB SPL) and report that the background noise of the 2-photon scope is <30 dB SPL, which seems very quiet. The authors should at least describe the sound-proofing they used to get the noise level that low, and some sense of noise across the 2-40 kHz frequency range would be nice as a supplementary figure. It also remains unclear just what the 2-photon dF/F response represents in terms of spikes. Classic mapping using single-unit or multi-unit electrodes might be sensitive to single spikes (as might be emitted at characteristic frequency), but this might not be as obvious for Ca2+ imaging. This isn't a concern for the internal comparison here between TR and CT cells as conditions are similar, but is a concern for relating the tonotopy or lack thereof reported here to other studies.

(2) It seems a bit peculiar that while 2721 CT neurons (N=10 mice) were imaged, less than half as many TR cells were imaged (n=1041 cells from N=5 mice). I would have expected there to be many more TR neurons even mouse for mouse (normalizing by number of neurons per mouse), but perhaps the authors were just interested in a comparison data set and not being as thorough or complete with the TR imaging?

(3) The authors definitions of neuronal response type in the methods needs more quantitative detail. The authors state: ""Irregular" neurons exhibited spontaneous activity with highly variable responses to sound stimulation. "Tuned" neurons were responsive neurons that demonstrated significant selectivity for certain stimuli. "Silent" neurons were defined as those that remained completely inactive during our recording period (> 30 min). For tuned neurons, the best frequency (BF) was defined as the sound frequency associated with the highest response averaged across all sound levels." The authors need to define what their thresholds are for 'highly variable', 'significant', and 'completely inactive'. Is best frequency the most significant response, the global max (even if another stimulus evokes a very close amplitude response), etc.

Comments on revisions:

I think the authors misunderstood my point about sound level and characteristic frequency vs best frequency tonotopic maps. Yes, many studies of cortical responses present stimuli at higher intensities than the characteristic frequencies, but as tuning curves widen with sound level, the macroscopic tonotopic organization of primary auditory cortex breaks down at higher intensities. This is why most of the classic studies of tonotopy e.g., from the Merzenich lab) generated maps of characteristic frequency. As I mentioned before, this isn't so much of an issue for the authors' comparisons of TR vs CT organization in their own study, but in general, this makes it difficult to compare aspects of spatially-organized tonotopy from imaging studies with the older electrophysiological 'truer' tonotopic maps. That said, this means that CT cells also might be tonotopically organized if the authors had been able to look at lower intensity tuning properties.

Author response:

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

Reviewer #1 (Public Review):

Summary:

In this study, Gu et al. employed novel viral strategies, combined with in vivo two-photon imaging, to map the tone response properties of two groups of cortical neurons in A1. The thalamocortical recipient (TR neurons) and the corticothalamic (CT neurons). They observed a clear tonotopic gradient among TR neurons but not in CT neurons. Moreover, CT neurons exhibited high heterogeneity of their frequency tuning and broader bandwidth, suggesting increased synaptic integration in these neurons. By parsing out different projecting-specific neurons within A1, this study provides insight into how neurons with different connectivity can exhibit different frequency response-related topographic organization.

Strengths:

This study reveals the importance of studying neurons with projection specificity rather than layer specificity since neurons within the same layer have very diverse molecular, morphological, physiological, and connectional features. By utilizing a newly developed rabies virus CSN-N2c GCaMP-expressing vector, the authors can label and image specifically the neurons (CT neurons) in A1 that project to the MGB. To compare, they used an anterograde trans-synaptic tracing strategy to label and image neurons in A1 that receive input from MGB (TR neurons).

Weaknesses:

Perhaps as cited in the introduction, it is well known that tonotopic gradient is well preserved across all layers within A1, but I feel if the authors want to highlight the specificity of their virus tracing strategy and the populations that they imaged in L2/3 (TR neurons) and L6 (CT neurons), they should perform control groups where they image general excitatory neurons in the two depths and compare to TR and CT neurons, respectively. This will show that it's not their imaging/analysis or behavioral paradigms that are different from other labs. 

We thank the reviewer for these constructive suggestions. As recommended, we have performed control experiments that imaged the general excitatory neurons in superficial layers (shown below), and the results showed a clear tonotopic gradient, which was consistent with previous findings (Bandyopadhyay et al., 2010; Romero et al., 2020; Rothschild et al., 2010; Tischbirek et al., 2019), thereby validating the reliability of our imaging/analysis approach. The results are presented in a new supplemental figure (Figure 2- figure supplementary 3).

Related publications:

(1) Gu M, Li X, Liang S, Zhu J, Sun P, He Y, Yu H, Li R, Zhou Z, Lyu J, Li SC, Budinger E, Zhou Y, Jia H, Zhang J, Chen X. 2023. Rabies virus-based labeling of layer 6 corticothalamic neurons for two-photon imaging in vivo. iScience 26: 106625. DIO: https://doi.org/10.1016/j.isci.2023.106625, PMID: 37250327

(2) Bandyopadhyay S, Shamma SA, Kanold PO. 2010. Dichotomy of functional organization in the mouse auditory cortex. Nat Neurosci 13: 361-8. DIO: https://doi.org/10.1038/nn.2490, PMID: 20118924

(3) Romero S, Hight AE, Clayton KK, Resnik J, Williamson RS, Hancock KE, Polley DB. 2020. Cellular and Widefield Imaging of Sound Frequency Organization in Primary and Higher Order Fields of the Mouse Auditory Cortex. Cerebral Cortex 30: 1603-1622. DIO: https://doi.org/10.1093/cercor/bhz190, PMID: 31667491

(4) Rothschild G, Nelken I, Mizrahi A. 2010. Functional organization and population dynamics in the mouse primary auditory cortex. Nat Neurosci 13: 353-60. DIO: https://doi.org/10.1038/nn.2484, PMID: 20118927

(5) Tischbirek CH, Noda T, Tohmi M, Birkner A, Nelken I, Konnerth A. 2019. In Vivo Functional Mapping of a Cortical Column at Single-Neuron Resolution. Cell Rep 27: 1319-1326 e5. DIO: https://doi.org/10.1016/j.celrep.2019.04.007, PMID: 31042460

Figures 1D and G, the y-axis is Distance from pia (%). I'm not exactly sure what this means. How does % translate to real cortical thickness?

We thank the reviewer for this question. The distance of labeled cells from pia was normalized to the entire distance from pia to L6/WM border for each mouse, according to the previous study (Chang and Kawai, 2018). For all mice tested, the entire distance from pia to L6/WM border was 826.5 ± 23.4 mm (in the range of 752.9 to 886.1).

Related publications:

Chang M, Kawai HD. 2018. A characterization of laminar architecture in mouse primary auditory cortex. Brain Structure and Function 223: 4187-4209. DIO: https://doi.org/10.1007/s00429-018-1744-8, PMID: 30187193

For Figure 2G and H, is each circle a neuron or an animal? Why are they staggered on top of each other on the x-axis? If the x-axis is the distance from caudal to rostral, each neuron should have a different distance? Also, it seems like it's because Figure 2H has more circles, which is why it has more variation, thus not significant (for example, at 600 or 900um, 2G seems to have fewer circles than 2H). 

We sincerely appreciate the reviewer’s careful attention to the details of our figures. Each circle in the Figure 2G and H represents an individual imaging focal plane from different animals, and the median BF of some focal planes may be similar, leading to partial overlap. In the regions where overlap occurs, the brightness of the circle will be additive.

Since fewer CT neurons, compared to TR neurons, responded to pure tones within each focal plane, as shown in Figure 2- figure supplementary 2, a larger number of focal planes were imaged to ensure a consistent and robust analysis of the pure tone response characteristics. The higher variance and lack of correlation in CT neurons is a key biological finding, not an artifact of sample size. The data clearly show a wide spread of median BFs at any given location for CT neurons, a feature absent in the TR population.

Similarly, in Figures 2J and L, why are the circles staggered on the y-axis now? And is each circle now a neuron or a trial? It seems they have many more circles than Figure 2G and 2H. Also, I don't think doing a correlation is the proper stats for this type of plot (this point applies to Figures 3H and 3J).

We regret any confusion have caused. In fact, Figure 2 illustrates the tonotopic gradient of CT and TR neurons at different scales. Specifically, Figures 2E-H present the imaging from the focal plane perspective (23 focal planes in Figures 2G, 40 focal planes in Figures 2H), whereas Figures 2I-L provide a more detailed view at the single-cell level (481 neurons in Figures 2J, 491 neurons in Figures 2L). So, Figures 2J and L do indeed have more circles than Figures 2G and H. The analysis at these varying scales consistently reveals the presence of a tonotopic gradient in TR neurons, whereas such a gradient is absent in CT neurons.

We used Pearson correlation as a standard and direct method to quantify the linear relationship between a neuron's anatomical position and its frequency preference, which is widely used in the field to provide a quantitative measure (R-value) and a significance level (p-value) for the strength of a tonotopic gradient. The same statistical logic applies to testing for spatial gradients in local heterogeneity in Figure 3. We are confident that this is an appropriate and informative statistical approach for these data.

What does the inter-quartile range of BF (IQRBF, in octaves) imply? What's the interpretation of this analysis? I am confused as to why TR neurons show high IQR in HF areas compared to LF areas, which means homogeneity among TR neurons (lines 213 - 216). On the same note, how is this different from the BF variability?  Isn't higher IQR equal to higher variability?

We thank the reviewer for raising this important point. IQRBF, is a measure of local tuning heterogeneity. It quantifies the diversity of BFs among neighboring neurons. A small IQRBF means neighbors are similarly tuned (an orderly, homogeneous map), while a large IQRBF means neighbors have very different BFs (a disordered, heterogeneous map). (Winkowski and Kanold, 2013; Zeng et al., 2019).

From the BF position reconstruction of all TR neurons (Figures 2I), most TR neurons respond to high-frequency sounds in the high-frequency (HF) region, but some neurons respond to low frequencies such as 2 kHz, which contributes to high IQR in HF areas. This does not contradict our main conclusion, that the TR neurons is significantly more homogeneous than the CT neurons. BF variability represents the stability of a neuron's BF over time, while IQR represents the variability of BF among different neurons within a certain range. (Chambers et al., 2023).

Related publications:

(1) Chambers AR, Aschauer DF, Eppler JB, Kaschube M, Rumpel S. 2023. A stable sensory map emerges from a dynamic equilibrium of neurons with unstable tuning properties. Cerebral Cortex 33: 5597-5612. DIO: https://doi.org/10.1093/cercor/bhac445, PMID: 36418925

(2) Winkowski DE, Kanold PO. 2013. Laminar transformation of frequency organization in auditory cortex. Journal of Neuroscience 33: 1498-508. DIO: https://doi.org/10.1523/JNEUROSCI.3101-12.2013, PMID: 23345224

(3) Zeng HH, Huang JF, Chen M, Wen YQ, Shen ZM, Poo MM. 2019. Local homogeneity of tonotopic organization in the primary auditory cortex of marmosets. Proceedings of the National Academy of Sciences of the United States of America 116: 3239-3244. DIO: https://doi.org/10.1073/pnas.1816653116, PMID: 30718428

Figure 4A-B, there are no clear criteria on how the authors categorize V, I, and O shapes. The descriptions in the Methods (lines 721 - 725) are also very vague.

We apologize for the initial vagueness and have replaced the descriptions in the Methods section. “V-shaped”: Neurons whose FRAs show decreasing frequency selectivity with increasing intensity. “I-shaped”: Neurons whose FRAs show constant frequency selectivity with increasing intensity. “O-shaped”: Neurons responsive to a small range of intensities and frequencies, with the peak response not occurring at the highest intensity level.

To provide better visual intuition, we show multiple representative examples of each FRA type for both TR and CT neurons below. We are confident that these provide the necessary clarity and reproducibility for our analysis of receptive field properties.

Author response image 1.

Different FRA types within the dataset of TR and CT neurons. Each row shows 6 representative FRAs from a specific type. Types are V-shaped (‘V'), I-shaped (‘I’), and O-shaped (‘O’). The X-axis represents 11 pure tone frequencies, and the Y-axis represents 6 sound intensities.

Reviewer #2 (Public Review):

Summary:

Gu and Liang et. al investigated how auditory information is mapped and transformed as it enters and exits an auditory cortex. They use anterograde transsynaptic tracers to label and perform calcium imaging of thalamorecipient neurons in A1 and retrograde tracers to label and perform calcium imaging of corticothalamic output neurons. They demonstrate a degradation of tonotopic organization from the input to output neurons.

Strengths:

The experiments appear well executed, well described, and analyzed.

Weaknesses:

(1) Given that the CT and TR neurons were imaged at different depths, the question as to whether or not these differences could otherwise be explained by layer-specific differences is still not 100% resolved. Control measurements would be needed either by recording (1) CT neurons in upper layers, (2) TR in deeper layers, (3) non-CT in deeper layers and/or (4) non-TR in upper layers.

We appreciate these constructive suggestions. To address this, we performed new experiments and analyses.

Comparison of TR neurons across superficial layers: we analyzed our existing TR neuron dataset to see if response properties varied by depth within the superficial layers. We found no significant differences in the fraction of tuned neurons, field IQR, or maximum bandwidth (BWmax) between TR neurons in L2/3 and L4. This suggests a degree of functional homogeneity within the thalamorecipient population across these layers. The results are presented in new supplemental figures (Figure 2- figure supplementary 4).

Necessary control experiments.

(1) CT neurons in upper layers. CT neurons are thalamic projection neurons that only exist in the deeper cortex, so CT neurons do not exist in upper layers (Antunes and Malmierca, 2021).

(2) TR neurons in deeper layers. As we mentioned in the manuscript, due to high-titer AAV1-Cre virus labeling controversy (anterograde and retrograde labelling both exist), it is challenging to identify TR neurons in deeper layers.

(3) non-CT in deeper layers and/or (4) non-TR in upper layers.

To directly test if projection identity confers distinct functional properties within the same cortical layers, we performed the crucial control of comparing TR neurons to their neighboring non-TR neurons. We injected AAV1-Cre in MGB and a Cre-dependent mCherry into A1 to label TR neurons red. We then co-injected AAV-CaMKII-GCaMP6s to label the general excitatory population green.  In merged images, this allowed us to functionally image and directly compare TR neurons (yellow) and adjacent non-TR neurons (green). We separately recorded the responses of these neurons to pure tones using two-photon imaging. The results show that TR neurons are significantly more likely to be tuned to pure tones than their neighboring non-TR excitatory neurons. This finding provides direct evidence that a neuron's long-range connectivity, and not just its laminar location, is a key determinant of its response properties. The results are presented in new supplemental figures (Figure 2- figure supplementary 5).

Related publications:

Antunes FM, Malmierca MS. 2021. Corticothalamic Pathways in Auditory Processing: Recent Advances and Insights From Other Sensory Systems. Front Neural Circuits 15: 721186. DIO: https://doi.org/10.3389/fncir.2021.721186, PMID: 34489648

(2) What percent of the neurons at the depths are CT neurons? Similar questions for TR neurons?

We thank the reviewer for the comments. We performed histological analysis on brain slices from our experimental animals to quantify the density of these projection-specific populations. Our analysis reveals that CT neurons constitute approximately 25.47%\22.99%–36.50% of all neurons in Layer 6 of A1. In the superficial layers（L2/3 and L4）, TR neurons comprise approximately 10.66%\10.53%–11.37% of the total neuronal population.

Author response image 2.

The fraction of CT and TR neurons. (A) Boxplots showing the fraction of CT neurons. N = 11 slices from 4 mice. (B) Boxplots showing the fraction of TR neurons. N = 11 slices from 4 mice.

(3) V-shaped, I-shaped, or O-shaped is not an intuitively understood nomenclature, consider changing. Further, the x/y axis for Figure 4a is not labeled, so it's not clear what the heat maps are supposed to represent.

The terms "V-shaped," "I-shaped," and "O-shaped" are an established nomenclature in the auditory neuroscience literature for describing frequency response areas (FRAs), and we use them for consistency with prior work. V-shaped: Neurons whose FRAs show decreasing frequency selectivity with increasing intensity. I-shaped: Neurons whose FRAs show constant frequency selectivity with increasing intensity. O-shaped: Neurons responsive to a small range of intensities and frequencies, with the peak response not occurring at the highest intensity level.

(Rothschild et al., 2010). We have included a more detailed description in the Methods.

The X-axis represents 11 pure tone frequencies, and the Y-axis represents 6 sound intensities. So, the heat map represents the FRA of neurons in A1, reflecting the responses for different frequencies and intensities of sound stimuli. In the revised manuscript, we have provided clarifications in the figure legend.

(4) Many references about projection neurons and cortical circuits are based on studies from visual or somatosensory cortex. Auditory cortex organization is not necessarily the same as other sensory areas. Auditory cortex references should be used specifically, and not sources reporting on S1, and V1.

We thank the reviewers for their valuable comments. We have made a concerted effort to ensure that claims about cortical circuit organization are supported by findings specifically from the auditory cortex wherever possible, strengthening the focus and specificity of our discussion.

Reviewer #3 (Public Review):

Summary:

The authors performed wide-field and 2-photon imaging in vivo in awake head-fixed mice, to compare receptive fields and tonotopic organization in thalamocortical recipient (TR) neurons vs corticothalamic (CT) neurons of mouse auditory cortex. TR neurons were found in all cortical layers while CT neurons were restricted to layer 6. The TR neurons at nominal depths of 200-400 microns have a remarkable degree of tonotopy (as good if not better than tonotopic maps reported by multiunit recordings). In contrast, CT neurons were very heterogenous in terms of their best frequency (BF), even when focusing on the low vs high-frequency regions of the primary auditory cortex. CT neurons also had wider tuning.

Strengths:

This is a thorough examination using modern methods, helping to resolve a question in the field with projection-specific mapping.

Weaknesses:

There are some limitations due to the methods, and it's unclear what the importance of these responses are outside of behavioral context or measured at single timepoints given the plasticity, context-dependence, and receptive field 'drift' that can occur in the cortex.

(1) Probably the biggest conceptual difficulty I have with the paper is comparing these results to past studies mapping auditory cortex topography, mainly due to differences in methods. Conventionally, the tonotopic organization is observed for characteristic frequency maps (not best frequency maps), as tuning precision degrades and the best frequency can shift as sound intensity increases. The authors used six attenuation levels (30-80 dB SPL) and reported that the background noise of the 2-photon scope is <30 dB SPL, which seems very quiet. The authors should at least describe the sound-proofing they used to get the noise level that low, and some sense of noise across the 2-40 kHz frequency range would be nice as a supplementary figure. It also remains unclear just what the 2-photon dF/F response represents in terms of spikes. Classic mapping using single-unit or multi-unit electrodes might be sensitive to single spikes (as might be emitted at characteristic frequency), but this might not be as obvious for Ca2+ imaging. This isn't a concern for the internal comparison here between TR and CT cells as conditions are similar, but is a concern for relating the tonotopy or lack thereof reported here to other studies.

We sincerely thank the reviewer for the thoughtful evaluation of our manuscript and for your positive assessment of our work.

(1)  Concern regarding Best Frequency (BF) vs. Characteristic Frequency (CF)

Our use of BF, defined as the frequency eliciting the highest response averaged across all sound levels, is a standard and practical approach in 2-photon Ca²⁺ imaging studies. (Issa et al., 2014; Rothschild et al., 2010; Schmitt et al., 2023; Tischbirek et al., 2019). This method is well-suited for functionally characterizing large numbers of neurons simultaneously, where determining a precise firing threshold for each individual cell can be challenging.

(2) Concern regarding background noise of the 2-photon setup

We have expanded the Methods section ("Auditory stimulation") to include a detailed description of the sound-attenuation strategies used during the experiments. The use of a custom-built, double-walled sound-proof enclosure lined with wedge-shaped acoustic foam was implemented to significantly reduce external noise interference. These strategies ensured that auditory stimuli were delivered under highly controlled, low-noise conditions, thereby enhancing the reliability and accuracy of the neural response measurements obtained throughout the study.

(3) Concern regarding the relationship between dF/F and spikes

While Ca²⁺ signals are an indirect and filtered representation of spiking activity, they are a powerful tool for assessing the functional properties of genetically-defined cell populations. As you note, the properties and limitations of Ca²⁺ imaging apply equally to both the TR and CT neuron groups we recorded. Therefore, the profound difference we observed—a clear tonotopic gradient in one population and a lack thereof in the other—is a robust biological finding and not a methodological artifact.

Related publications:

(1) Issa JB, Haeffele BD, Agarwal A, Bergles DE, Young ED, Yue DT. 2014. Multiscale optical Ca2+ imaging of tonal organization in mouse auditory cortex. Neuron 83: 944-59. DIO: https://doi.org/10.1016/j.neuron.2014.07.009, PMID: 25088366

(2) Rothschild G, Nelken I, Mizrahi A. 2010. Functional organization and population dynamics in the mouse primary auditory cortex. Nat Neurosci 13: 353-60. DIO: https://doi.org/10.1038/nn.2484, PMID: 20118927

(3) Schmitt TTX, Andrea KMA, Wadle SL, Hirtz JJ. 2023. Distinct topographic organization and network activity patterns of corticocollicular neurons within layer 5 auditory cortex. Front Neural Circuits 17: 1210057. DIO: https://doi.org/10.3389/fncir.2023.1210057, PMID: 37521334

(4) Tischbirek CH, Noda T, Tohmi M, Birkner A, Nelken I, Konnerth A. 2019. In Vivo Functional Mapping of a Cortical Column at Single-Neuron Resolution. Cell Rep 27: 1319-1326 e5. DIO: https://doi.org/10.1016/j.celrep.2019.04.007, PMID: 31042460

(2) It seems a bit peculiar that while 2721 CT neurons (N=10 mice) were imaged, less than half as many TR cells were imaged (n=1041 cells from N=5 mice). I would have expected there to be many more TR neurons even mouse for mouse (normalizing by number of neurons per mouse), but perhaps the authors were just interested in a comparison data set and not being as thorough or complete with the TR imaging?

As shown in the Figure 2- figure supplementary 2, a much higher fraction of TR neurons was "tuned" to pure tones (46% of 1041 neurons) compared to CT neurons (only 18% of 2721 neurons). To obtain a statistically robust and comparable number of tuned neurons for our core analysis (481 tuned TR neurons vs. 491 tuned CT neurons), it was necessary to sample a larger total population of CT neurons, which required imaging from more animals.

(3) The authors' definitions of neuronal response type in the methods need more quantitative detail. The authors state: "Irregular" neurons exhibited spontaneous activity with highly variable responses to sound stimulation. "Tuned" neurons were responsive neurons that demonstrated significant selectivity for certain stimuli. "Silent" neurons were defined as those that remained completely inactive during our recording period (> 30 min). For tuned neurons, the best frequency (BF) was defined as the sound frequency associated with the highest response averaged across all sound levels.". The authors need to define what their thresholds are for 'highly variable', 'significant', and 'completely inactive'. Is best frequency the most significant response, the global max (even if another stimulus evokes a very close amplitude response), etc.

We appreciate the reviewer's suggestions. We have added more detailed description in the Methods.

Tuned neurons: A responsive neuron was further classified as "Tuned" if its responses showed significant frequency selectivity. We determined this using a one-way ANOVA on the neuron's response amplitudes across all tested frequencies (at the sound level that elicited the maximal response). If the ANOVA yielded a p-value < 0.05, the neuron was considered "Tuned”. Irregular neurons: Responsive neurons that did not meet the statistical criterion for being "Tuned" (i.e., ANOVA p-value ≥ 0.05) were classified as "Irregular”. This provides a clear, mutually exclusive category for sound-responsive but broadly-tuned or non-selective cells. Silent neurons: Neurons that were not responsive were classified as "Silent". This quantitatively defines them as cells that showed no significant stimulus-evoked activity during the entire recording session. Best frequency (BF): It is the frequency that elicited the maximal mean response, averaged across all sound levels.

To provide greater clarity, we showed examples in the following figures.

Author response image 3.

Reviewer #1 (Recommendations For The Authors):

(1) A1 and AuC were used exchangeably in the text.

Thank you for pointing out this issue. Our terminological strategy was to remain faithful to the original terms used in the literature we cite, where "AuC" is often used more broadly. In the revised manuscript, we have performed a careful edit to ensure that we use the specific term "A1" (primary auditory cortex) when describing our own results and recording locations, which were functionally and anatomically confirmed.

(2) Grammar mistakes throughout.

We are grateful for the reviewer’s suggested improvement to our wording. The entire manuscript has undergone a thorough professional copyediting process to correct all grammatical errors and improve overall readability.

(3) The discussion should talk more about how/why L6 CT neurons don't possess the tonotopic organization and what are the implications. Currently, it only says 'indicative of an increase in synaptic integration during cortical processing'...

Thanks for this suggestion. We have substantially revised and expanded the Discussion section to explore the potential mechanisms and functional implications of the lack of tonotopy in L6 CT neurons.

Broad pooling of inputs: We propose that the lack of tonotopy is an active computation, not a passive degradation. CT neurons likely pool inputs from a wide range of upstream neurons with diverse frequency preferences. This broad synaptic integration, reflected in their wider tuning bandwidth, would actively erase the fine-grained frequency map in favor of creating a different kind of representation.

A shift from topography to abstract representation: This transformation away from a classic sensory map may be critical for the function of corticothalamic feedback. Instead of relaying "what" frequency was heard, the descending signal from CT neurons may convey more abstract, higher-order information, such as the behavioral relevance of a sound, predictions about upcoming sounds, or motor-related efference copy signals that are not inherently frequency-specific.’

Modulatory role of the descending pathway: The descending A1-to-MGB pathway is often considered to be modulatory, shaping thalamic responses rather than driving them directly. A modulatory signal designed to globally adjust thalamic gain or selectivity may not require, and may even be hindered by, a fine-grained topographical organization.

Reviewer #2 (Recommendations For The Authors):

(1) Given that the CT and TR neurons were imaged at different depths, the question as to whether or not these differences could otherwise be explained by layer-specific differences is still not 100% resolved. Control measurements would be needed either by recording (1) CT neurons in upper layers (2) TR in deeper layers (3) non-CT in deeper layers and/or (4) non-TR in upper layers.

We appreciate these constructive suggestions. To address this, we performed new experiments and analyses.

Comparison of TR neurons across superficial layers: we analyzed our existing TR neuron dataset to see if response properties varied by depth within the superficial layers. We found no significant differences in the fraction of tuned neurons, field IQR, or maximum bandwidth (BWmax) between TR neurons in L2/3 and L4. This suggests a degree of functional homogeneity within the thalamorecipient population across these layers.

Necessary control experiments.

(1) CT neurons in upper layers. CT neurons are thalamic projection neurons that only exist in the deeper cortex, so CT neurons do not exist in upper layers (Antunes and Malmierca, 2021).

(2) TR neurons in deeper layers. As we mentioned in the manuscript, due to high-titer AAV1-Cre virus labeling controversy (anterograde and retrograde labelling both exist), it is challenging to identify TR neurons in deeper layers.

(3) non-CT in deeper layers and/or (4) non-TR in upper layers.

To directly test if projection identity confers distinct functional properties within the same cortical layers, we performed the crucial control of comparing TR neurons to their neighboring non-TR neurons. We injected AAV1-Cre in MGB and a Cre-dependent mCherry into A1 to label TR neurons red. We then co-injected AAV-CaMKII-GCaMP6s to label the general excitatory population green.  In merged images, this allowed us to functionally image and directly compare TR neurons (yellow) and adjacent non-TR neurons (green). We separately recorded the responses of these neurons to pure tones using two-photon imaging. The results show that TR neurons are significantly more likely to be tuned to pure tones than their neighboring non-TR excitatory neurons. This finding provides direct evidence that a neuron's long-range connectivity, and not just its laminar location, is a key determinant of its response properties.

Related publications:

Antunes FM, Malmierca MS. 2021. Corticothalamic Pathways in Auditory Processing: Recent Advances and Insights From Other Sensory Systems. Front Neural Circuits 15: 721186. DIO: https://doi.org/10.3389/fncir.2021.721186, PMID: 34489648

(3) V-shaped, I-shaped, or O-shaped is not an intuitively understood nomenclature, consider changing. Further, the x/y axis for Figure 4a is not labeled, so it's not clear what the heat maps are supposed to represent.

The terms "V-shaped," "I-shaped," and "O-shaped" are an established nomenclature in the auditory neuroscience literature for describing frequency response areas (FRAs), and we use them for consistency with prior work. V-shaped: Neurons whose FRAs show decreasing frequency selectivity with increasing intensity. I-shaped: Neurons whose FRAs show constant frequency selectivity with increasing intensity. O-shaped: Neurons responsive to a small range of intensities and frequencies, with the peak response not occurring at the highest intensity level.

(Rothschild et al., 2010). We have included a more detailed description in the Methods.

The X-axis represents 11 pure tone frequencies, and the Y-axis represents 6 sound intensities. So, the heat map represents the FRA of neurons in A1, reflecting the responses for different frequencies and intensities of sound stimuli. In the revised manuscript, we have provided clarifications in the figure legend.

(4) Many references about projection neurons and cortical circuits are based on studies from visual or somatosensory cortex. Auditory cortex organization is not necessarily the same as other sensory areas. Auditory cortex references should be used specifically, and not sources reporting on S1, V1.

We thank the reviewers for their valuable comments. We have made a concerted effort to ensure that claims about cortical circuit organization are supported by findings specifically from the auditory cortex wherever possible, strengthening the focus and specificity of our discussion.

Reviewer #3 (Recommendations For The Authors):

I suggest showing some more examples of how different neurons and receptive field properties were quantified and statistically analyzed. Especially in Figure 4, but really throughout.

We thank the reviewer for this valuable suggestion. To provide greater clarity, we have added more examples in the following figure.

eLife Assessment

This study presents a valuable assessment of and solid evidence for increased similarity in visual appearance combined with increased chemical differences between two butterfly species in sympatry compared with differences between three populations of one of the two species in allopatry. The similarity in visual appearance hints to an evolutionary response to shared predators (but alternative explanations are possible). Thus, the difference in chemical signaling likely helps to avoid between-species mating in sympatry.

Joint Public Review:

Summary:

Ledamoisel et al. examined the evolution of visual and chemical signals in closely related Morpho butterfly species to understand their role in species coexistence. Using an integrative, state-of-the-art approach combining spectrophotometry, visual modeling, and behavioral mate choice experiments, they quantified differences in wing iridescence and assessed its influence on mate preference in allopatry and sympatry. They also performed chemical analyses to determine whether sympatric species exhibit divergent chemical cues that may facilitate species recognition and mate discrimination. The authors found iridescent coloration to be similar in sympatric Morpho species. Furthermore, male mate choice experiments revealed that in sympatry, males fail to discriminate conspecific females based on coloration, reinforcing the idea that visual signal convergence is primarily driven by predation pressure. In contrast, the divergence of chemical signals among sympatric species suggests their potential role in facilitating species recognition and mate discrimination. The authors conclude that interactions between ecological pressures and signal evolution may shape species coexistence.

Strengths:

The study is well-designed and integrates multiple methodological approaches to provide a thorough assessment of signal evolution in the studied species. We appreciate the authors' careful consideration of multiple selective pressures and their combined influence on signal divergence and convergence. Additionally, the inclusion of both visual and chemical signals adds an interesting and valuable dimension to the study, enhancing its importance. Beyond butterflies, this research broadens our understanding of multimodal communication and signal evolution in the context of species coexistence.

Reviewing Editor comment:

The authors have improved their submission after revisions and responded to the previous concerns of the reviewers.

Author response:

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

Reviewer #1 (Public review):

Summary:

In this study, Ledamoisel et al. examined the evolution of visual and chemical signals in closely related Morpho butterfly species to understand their role in species coexistence. Using an integrative, state-of-the-art approach combining spectrophotometry, visual modeling, and behavioral mate choice experiments, they quantified differences in wing iridescence and assessed its influence on mate preference in allopatry and sympatry. They also performed chemical analyses to determine whether sympatric species exhibit divergent chemical cues that may facilitate species recognition and mate discrimination. The authors found iridescent coloration to be similar in sympatric Morpho species. Furthermore, male mate choice experiments revealed that in sympatry, males fail to discriminate conspecific females based on coloration, reinforcing the idea that visual signal convergence is primarily driven by predation pressure. In contrast, the divergence of chemical signals among sympatric species suggests their potential role in facilitating species recognition and mate discrimination. The authors conclude that interactions between ecological pressures and signal evolution may shape species coexistence.

Strengths:

The study is well-designed and integrates multiple methodological approaches to provide a thorough assessment of signal evolution in the studied species. I appreciate the authors' careful consideration of multiple selective pressures and their combined influence on signal divergence and convergence. Additionally, the inclusion of both visual and chemical signals adds an interesting and valuable dimension to the study, enhancing its importance. Beyond butterflies, this research broadens our understanding of multimodal communication and signal evolution in the context of species coexistence.

Weaknesses:

(1) The broader significance of the findings needs to be better articulated. While the authors emphasize that comparing adaptive traits in sympatry and allopatry provides insights into selective processes shaping reproductive isolation and coexistence, it is unclear what key conceptual or theoretical questions are being addressed. Are these patterns expected under certain evolutionary scenarios? Have they been empirically demonstrated in other systems? The authors should explicitly state the overarching research question, incorporate some predictions, and better contextualize their findings within the existing literature. If the results challenge or support previous work, that should be highlighted to strengthen the study's importance in a broader context.

We thank the reviewer for their valuable feedback. We understand that the framing of the results and the discussion may fail to convey the broader significance of our findings. In the first version of the manuscript, we framed our manuscript around the processes shaping reproductive isolation and co-existence in sympatry, but now realize that this question was too broad in regards to our results. We thus strictly focused on outlining the importance of ecological interactions in the evolution of traits in sympatric species. In the revised version of the manuscript, we rewrote the first paragraph of the introduction to introduce context regarding the effect of ecological interactions on trait evolution (lines 43-60). We then explicitly introduce the theoretical question investigated in our paper (i.e. “we investigate how ecological interactions in sympatry can constrain natural and sexual selection shaping trait evolution”, lines 62-63) and our predictions regarding the evolution of traits in sympatry vs. allopatry (lines 74-80). We also added predictions regarding our experiments on Morpho at the end of the introduction (lines 146-157). As a result, the discussion is now better aligned with the introduction, by discussing the putative effect of predation and mate choice on the evolution of wing iridescence in Morpho.

(2) The motivation for studying visual signals and mate choice in allopatric populations (i.e., at the intraspecific level) is not well articulated, leaving their role in the broader narrative unclear. In particular, the rationale behind experiments 1, 2, and 3 is not well defined, as the authors have not made a strong case for the need for these intraspecific comparisons in the introduction. This issue is further compounded by the authors' primary focus on signal evolution in sympatry throughout both the results and the discussion. For instance, the divergence of iridescence in allopatry is a potentially interesting result. But the authors have not discussed its implications.

We now clearly state in the introduction our motivation for studying visual signals and mate choice in allopatric populations (lines 74-80, lines 146-157). We argued that intraspecific comparisons help identify whether visual cues can be used in mate recognition between phylogenetically close subspecies, between whom visual resemblance is supposed to be higher than between closely-related species (tetrad experiment, and experiment 1). As M. h. bristowi and M. h. theodorus have different wing pattern, we also used this comparison to identify the traits involved in male mate preference within a species, testing the importance of iridescent color (experiment 2) or iridescent patterning (experiment 3). The results of those experiments can then be used to assess whether these traits are used in species recognition between sympatric species. See also our answers to recommendations 11 and 15 from reviewer #1.

Overall, given that the primary conclusions are based on results and analyses in sympatry, the role of allopatric populations in shaping these conclusions needs to be better integrated and justified. Without a stronger link between the comparative framework and the study's key takeaways, the use of allopatric populations feels somewhat peripheral rather than central to the study's aim. Since the primary conclusions remain valid even without the allopatric comparisons, their inclusion requires a clearer rationale.

To make a stronger case for the use of the allopatric population in our manuscript, we strengthened the justification behind the study of intraspecific allopatric populations vs. interspecific sympatric populations, as the iridescence measurements and the mate choice experiments in allopatric populations can serve as a baseline in studying how species interactions can shape the evolution of traits and mate recognition when compared to sympatric populations. Following your major comment #1, we rewrote the introduction to include a justification to the need for studying allopatric vs. sympatric populations (lines 74-80), and also further highlighted the need to study iridescence in sympatric species to fully understand the trait evolution of sympatric species in the discussion (339-343).

(3) While the authors demonstrate that iridescence is indistinguishable to predators in sympatry, they overstate the role of predation in driving convergence. The present study does not experimentally demonstrate that iridescence in this species has a confusion effect or contributes to evasive mimicry. Alternatively, convergence could result from other selective forces, such as signal efficacy due to environmental conditions, rather than being solely driven by predation.

We acknowledge that our study does not directly demonstrate that iridescence contributes to evasive mimicry. We did tone down the interpretation of the results in the discussion and state that predation is not the only selective pressure that could have promoted a convergent evolution of iridescence in sympatric species, as iridescence is a trait that could be involved in thermoregulation (lines 346-353) and camouflage (lines 363-369) for example. We made sure to mention that convergence in iridescent signals in sympatry is only an indirect support to the evasive mimicry hypothesis, and that further research is still needed, including direct predation experiments, to show that this convergence is indeed triggered by predation (lines 391-396).  

Reviewer #2 (Public review):

This study presents an investigation of the visual and chemical properties and mating behaviour in Morpho butterflies, aimed at addressing the nature of divergence between closely related species in sympatry. The study species consists of three subspecies of Morpho helenor (bristowi, theodorus, and helenor), and the conspecific Morpho achilles achilles. The authors postulate that whereas the iridescent blue signals of all (sub)species should function as a predator reduction signal (similar to aposematism) and therefore exhibit convergence, the same signals should indicate divergence if used as a mating signal, particularly in sympatric populations. They also assess chemical profiles among the species to assess the potential utility of scent in mediating species/sex discrimination.

The authors first used reflectance spectrometry to calculate hue, brightness, and chroma, plus two measures of "iridescence" (perhaps better phrased as angular dependence) in each (sub)species. This indicated the ubiquitous presence of sexual dimorphism in brightness (males brighter), which also appears to be the case for iridescence (Figure 3A-B). Analysis of these data also indicated that whereas there is evidence for divergence among subspecies in allopatry, the same evidence is lacking for species in sympatry (P = 0.084). This was supported further by visual modelling, which showed that both conspecifics and birds should be (theoretically) capable of perceiving the colour difference among allopatric populations of M. helenor, whereas the same is not true for the sympatric species.

The authors then conducted mate choice trials, first using live individuals and second using female dummies. The live experiments indicated the presence of assortative mating among the two subspecies of M. helenor (bristowi and theodorus). The dummy presentations indicated (a) bristowi males prefer conspecific wings, whereas theodorus have no preference, (b) bristowi males prefer the con(sub)specific colour pattern, (c) theodorus prefer the con(sub)specific iridescence when the pattern is manipulated to be similar among female dummies. A fourth experiment, using sympatric M. achilles and M. helenor, indicated no preference for conspecific female dummies. Finally, chemical analysis indicated substantial differences between these two species in putative pheromone compounds, and especially so in the males.

The authors conclude that the similarity of iridescence among species in sympatry is suggestive of convergence upon a common anti-predation signal. Despite some behavioural evidence in favourof colour (iridescence)-based mate discrimination, chemical differences between Achilles and Helenor are posed as more likely to function for species isolation than visual differences.

Overall, I enjoyed reading this manuscript, which presents a valiant attempt at studying visual, chemical and behavioural divergence in this iconic group of butterflies.

Major comments

My only major comment concerns the authors' favoured explanation for aposematism (or evasive mimicry) for convergence among species, which is based upon the you-can't-catch-me hypothesis first presented by Young 1971. Although there is supporting work showing that iridescent-like stimuli are more difficult to precisely localize by a range of viewers, most of the evidence as applied to the Morpho system is circumstantial, and I'm not certain that there is widespread acceptance of this hypothesis. Given that the present study deals with closely-related  (sub)species, one alternative explanation - a "null" hypothesis of sorts - is for a lack of divergence (from a common starting point) as opposed to evolutionary convergence per se. in other words, two subspecies are likely to retain ancestral character states unless there is selection that causes them to diverge. I feel that the manuscript would benefit from a discussion of this alternative, if not others. Signalling to predators could very well be involved in constraining the extent of convergence, but this seems a little premature to state as an up-front conclusion of this work. There is also the result of a *dorsal* wing manipulation by Vieira-Silva et al. 2024 which seems difficult to reconcile in light of this explanation. Whereas this paper is cited by the authors, a more nuanced discussion of their experimental results would seem appropriate here.

We thank the reviewer for their constructive comments on our manuscript. We appreciate the reviewer’s concern regarding the way iridescence convergence between sympatric species is discussed in our manuscript, which align with similar concerns raised by Reviewer 1. Indeed, the you-can't-catch-me hypothesis has not been yet empirically tested in Morpho, this is currently a working hypothesis only supported by indirect lines of evidence.

Among the 30 known Morpho species, iridescence is most likely the ancestral character, notably because iridescence is a trait shared by a majority of Morpho (we now mention this in the introduction lines 108-110). In this paper, we thus did not aim to identify the evolutionary forces involved in the appearance of iridescence in this group, but rather wanted to understand to what extent ecological interactions can impact the diversification (or not) of this trait. As such, the dorsal manipulations performed in Vieira-Silva et al 2024 showing that iridescence in Morpho may have a similar effect than crypsis does not impact our working hypothesis. Instead, we use VieraSilva et al 2024 to discuss the potential anti-predator effect of iridescence, that could potentially promote convergent evolution of iridescent patterns.

In the main text, we now clearly mention our null hypothesis: under a scenario of neutral evolution of iridescence, we would expect that the divergence in wing coloration between two M. helenor subspecies would be lower than between two different Morpho species (M. helenor and M. achilles) and showed that our results sharply differ from this null expectation.

We then improved the discussion by adding alternative hypotheses potentially explaining the convergent iridescent signal detected in sympatric species: we discussed the expected effect under neutral evolution (lines 339-343), but also added alternative hypotheses regarding the diversification of iridescence due to camouflage (lines 363-369), predator evasion (lines 373-377) and thermoregulation (lines 346-353).

Reviewer #3 (Public review):

The authors investigated differences in iridescence wing colouration of allopatric (geographically separated) and sympatric (coexisting) Morpho butterfly (sub)species. Their aim was to assess if iridescence wing colouration of Morpho (sub)species converged or diverged depending on coexistence and if iridescence wing colouration was involved in mating behaviour and reproductive isolation. The authors hypothesize that iridescence wing colouration of different (sub)species should converge in sympatry and diverge in allopatry. In sympatry, iridescence wing colouration can act as an effective antipredator defence with shared benefits if multiple (sub)species share the same colouration. However, shared wing colouration can have potential costs in terms of reproductive interference since wing colouration is often involved in mate recognition. If the benefits of a shared antipredator defence outweigh the costs of reproductive interference, iridescence wing colouration will show convergence and alternative mate recognition strategies might evolve, such as chemical mate recognition. In allopatry, iridescence wing colouration is expected to diverge due to adaptation to different local conditions and no alternative mate recognition is expected.

Strengths:

(1) Using allopatric and sympatric (sub)species that are closely related is a powerful way to test evolutionary hypotheses

(2) By clearly defining iridescence and measuring colour spectra from a variety of angles, applying different methods, a very comprehensive dataset of iridescence wing colouration is achieved.

(3) By experimentally manipulating wing coloration patterns, the authors show visual mate recognition for M. h. bristowi and could, in theory, separate different visual aspects of colouration (patterns VS iridescence strength).

(4) Measurements of chemical profiles to investigate alternative mate recognition strategies in case of convergence of visual signals.

Weaknesses:

In my opinion, studies should be judged on the methods and data included, and not on additional measurements that could have been taken or additional treatments/species that should be included, since in most ecological and evolutionary studies, more measurements or treatments/species can always be included. However, studies do need to ensure appropriate replication and appropriate measurements to test their hypothesis AND support their conclusions. The current study failed to ensure appropriate replication, and in various cases, the results do not support the conclusions.

First, when using allopatric and sympatric (sub)species pairs to test evolutionary hypotheses, replication is important. Ideally, multiple allopatric and sympatric (sub)species pairs are compared to avoid outlier (sub)species or pairs that lead to biased conclusions. Unfortunately, the current study compares 1 allopatric and 1 sympatric (sub)species pair, hence having poor (no) replication on the level of allopatric and sympatric (sub)species pairs,

We would like to thank the reviewer for their constructive feedback. We agree that replication is important to test evolutionary hypotheses and that our study lacks replication for allopatric and sympatric Morpho populations. Ideally, one would require several allopatric and sympatric replicates to conclude on the effect of species interaction in trait evolution. Our study is a preliminary attempt at answering this question, covering a few Morpho populations but proposing a broad assessment of iridescence and mate preference for those populations. We clearly mentioned in the discussion that investigating multiple populations is needed to test whether the trend we observed in this paper can be generalized (line 388-392).

Second, chemical profiles were only measured for sympatric species and not for allopatric (sub)species, which limits the interpretation of this data. The allopatric (sub)species could have been measured as non-coexistence "control". If coexistence and convergence in wing colouration drives the evolution of alternative mate recognition signals, such alternative signals should not evolve/diverge for allopatric (sub)species where wing colouration is still a reliable mate recognition cue. More importantly, no details are provided on the quantification of butterfly chemical profiles, which is essential to understand such data. It is unclear how the chemical profiles were quantified and what data (concentrations, ratios, proportions) were used to perform NDMS and generate Figure 5 and the associated statistical tests.

We recognize that having the chemical profiles of the genitalia of the Morpho from the allopatric populations would have made a stronger case in favor of reinforcement acting on the divergence of the chemical compounds found on the genitalia of the sympatric Morpho species. Due to limited access to the biological material needed at the time of the chromatography, we could not test for lower divergence in the chemical profiles of allopatric Morpho butterflies. We made sure to mention this limitation in the discussion (lines 457-461). 

We already stated in the methods that we compiled the area under the peak of each components found in the chromatograms of our samples and that we performed all the statistical analyses on this dataset. To make it clearer, we mention in the new version of the manuscript that the area under the peak of each component allows to measure the concentration of the components (in the methods lines 720, 723, 733). We also added some precisions in the legend of Figure 5.

Third, throughout the discussion, the authors mention that their results support natural selection by predators on iridescent wing colouration, without measuring natural selection by predators or any other measure related to predation. It is unclear by what predators any of the butterfly species are predated on at this point

We made sure to mention in the introduction (line 132-136) and in the discussion (line 373-377) that previous predation experiments performed on Morpho and other butterflies showed evidence that birds are likely predators for these species. These observations lead us to test for the putative effect of predation on the evolution of their color pattern, without directly testing predatory rates. We made sure this information is transparent in the revised manuscript, and now precise that assessing wing convergence is only an indirect way of testing the escape mimicry hypothesis (line 393-396).

To continue on the interpretation of the data related to selection on specific traits by specific selection agents: This study did not measure any form of selection or any selection agent. Hence, it is not known if iridescent wing colouration is actually under selection by predators and/or mates, if maybe other selection agents are involved or if these traits converge due to genetic correlations with other traits under selection. For example, Iridescent colouration in ground beetles has functions as antipredator defence but also thermo- and water regulation. None of these issues are recognized or discussed.

The lack of discussion of alternative selective pressures involved in the evolution of iridescence was pointed out by all reviewers. We thus modified the text to account for this comment, and no longer limit our discussion to the putative effects of predation. We now specifically discuss alternative hypotheses, including crypsis (362-369) and thermoregulation (line 346-353).

Finally, some of the results are weakly supported by statistics or questionable methodology.

Most notably, the perception of the iridescence coloration of allopatric subspecies by bird visual systems. Although for females, means and errors (not indicated what exactly, SD, SE or CI) are clearly above the 1 JND line, for males, means are only slightly above this line and errors or CIs clearly overlap with the 1 JND line. Since there is no additional statistical support, higher means but overlap of SD, SE or CI with the baseline provides weak statistical support for differences.

We thank the reviewer for bringing interpretation issues concerning the chromatic distances of allopatric Morpho species measured with a bird vision model. We made sure to be nuanced in the description of this graph in the results section (line 208-212). Note that this addition does not change our main conclusion stating that Morpho and predator visual models better discriminate iridescence differences between allopatric subspecies than between sympatric species.

We now also clearly mention in the figure’s legend that the error bars represent the confidence intervals obtained after performing a bootstrap analysis, in addition to the mention of the nature of the error bars already mentioned in the methods (line 580).

Regarding the assortative mating experiment, the results are clearly driven by M. bristowi. For M. theodorus, females mate equally often with conspecifics (6 times) as with M. bristowi (5 times). For males, the ratio is slightly better (6 vs 3), but with such low numbers, I doubt this is statistically testable. Overall low mating for M. bristowi could indicate suboptimal experimental conditions, and hence results should be interpreted with care.

We recognize that the tetrad experiment results are mainly driven by M. bristowi’s behavior as already mentioned in the results (line 231-232) but we now also mention it in the discussion (lines 401-402). This experiment would have benefited from more replicates, but the limited access to live males and virgin females for both subspecies was a limiting factor. Fisher’s exact test used to assess assortative mating is specifically appropriate to small sample sizes. We recognize that the sampling size is not ideal, however it is still statistically testable.

Regarding the wing manipulation experiment, M. theodorus does not show a preference when dummies with non-modified wings are presented and prefers non-modified dummies over modified dummies. This is acknowledged by the authors but not further discussed. Certainly, some control treatment for wing modification could have been added.

The use of controls to consider the effect of wing modification and odor by the permanent marker were already mentioned in the methods (lines 636-639). Following your recommendation and comments from the other reviewers, we now mention the use of this control in the results (lines 278283). We also address a potential issue that would have resulted in the rejection of these modified dummies by live males: we cannot be sure whether butterflies perceive these modifications as equivalent to natural coloration (lines 281-282). An additional control could have been used, adding black ink on the black dorsal parts of the pattern to assess its potential visual effect. The constraints on sampling unfortunately did not allow to add another treatment.

Overall, the fact that certain measurements only provide evidence for 1 of the 2 (sub)species (assortative mating, wing manipulation) or one sex of one of the species (bird visual systems) means overall interpretation and overgeneralization of the results to both allopatric or sympatric species should be done with care, and such nuances should ideally be discussed.

The aim of the authors, "to investigate the antagonistic effects of selective pressures generated by mate recognition and shared predation" has not been achieved, and the conclusions regarding this aim are not supported by the results. Nevertheless, the iridescence colour measurements are solid, and some of the behavioural experiments and chemical profile measurements seem to yield interesting results. The study would benefit from less overinterpretation of the results in the framework of predation and more careful consideration of methodological difficulties, statistical insecurities, and nuances in the results.

Overall, we would like to thank all reviewers for their thorough assessment of our work. We understand that the imbalance between mate choice data, visual model data and chemical data only gives us a partial assessment of species recognition in Morpho butterflies, thus requiring more precision in the interpretation and the discussion of our results. We made sure to add balanced interpretations in our discussion, by mentioning the lack of replicates for allopatric and sympatric populations (lines 391-392), and the lack of chemical characterization of allopatric species (lines 458361, see previous comments) and by being more transparent on methodological limitations that we failed to convey in the first version of our manuscript. We brought nuance to our discussion and also discussed alternative hypotheses to predation to explain the convergence of iridescence found in sympatry.

Reviewing Editor Comments:

While all reviewers acknowledge the value of your data, they converge in their recommendations to tone down the evolutionary interpretations. Ideally, to test your main hypothesis, you would need several species pairs, or if only one, as in your case, replicated sympatric and allopatric sites for both species. Furthermore, your more specific hypotheses about convergence (vs. nondivergence), response to predators (vs. other environmental variables), and avoiding interspecific mating in sympatry (vs. not avoiding it in allopatry) would require appropriate alternative treatments/controls. We therefore recommend that you focus on those statements that you can support with your experiments and data, and introduce these statements in the introduction with reference to the appropriate literature.

Reviewer #1 (Recommendations for the authors):

(1) Line 25: This stated aim seems a bit off. The authors did not sensu stricto quantify 'how shared adaptive traits may shape genetic divergence' in this study. I suggest rewriting or deleting this whole sentence altogether. The study's aim is already clear in lines 29-34.

We deleted the mention of the characterization of genetic divergence, since this study did not focus on any genetic analysis.

(2) Line 34: The authors here state that they compared allopatric vs sympatric populations. This is strictly not true for M. Achilles. Further, the results after this sentence focus solely ondivergence/convergence in sympatry, nothing at the intraspecific level and implications of the findings

We now mention that we tested allopatric vs. sympatric species of M. helenor only (lines 28-29). We also mention that the behavioral experiments were based on intraspecific comparisons, and discuss the implications of this result in the discussion.

(3) Line 35: 'convergence driven by predation': this is a strong statement and cannot be directly inferred from the present set of experiments. Consider toning it down.

We added nuance to this statement by rephrasing it “suggesting that predation may favors local resemblance” (lines 32-33)

(4) Line 36: Replace 'behavioral results' with 'behavioral experiments' or something similar.

Corrected

(5) Line 45-49: These opening statements need some citations.

We provided references for the first few lines, by citing terHorst et al 2018 (line 44) underlining the importance of species interactions in trait evolution, and Blomberg et al 2003 (line 45) showing that closely-related species tend to resemble each other by quantifying the phylogenetic signal of various traits.

(6) Line 83, 165: 'visual effect', not sure what the authors are referring to. Please rewrite.

We defined “visual effect” as the way wing color patterns could be perceived by predators or mates. We removed mentions of “visual effect” and directly used its definition instead.

(7) Line 105 onwards: This section of the introduction could benefit from more concise writing. The authors might consider reducing the number of specific examples and instead offering broader general statements, supported by citations from multiple studies.

We reduced the number of examples given in this paragraph and used general statements supported by multiple citations as examples. (lines 102-119).

(8) Line 108-110: This sentence seems to be redundant with the previous one.

We merged this sentence with the previous one to improve clarity. (lines 103-105)

(9) Line 140: 'with chemical defenses': include citations here.

We added citations of Joron et al 1999 and Merrill et al 2014, which document the evolution of convergent wing patterns (mimicry) in butterfly species with chemical-defenses.

(10) Line 149: This is a bit of a stretch. Note that genetic divergence could be influenced by many other things, not only the processes that the authors examined.

We agree with the reviewer that the study of the convergent vs. divergent evolution of visual cues is not enough to fully understand the mechanisms allowing genetic divergence between species. Because this paper does not focus on characterizing genetic divergence, we removed it from the manuscript to avoid oversimplification.

(11) Line 151: Again. Here, the author's primary focus seems to be at an interspecific level. One is left to wonder about the need for comparisons at the intraspecific level in M.helenor and the implications. Please clarify

In the end of the introduction (lines 146-157), we specifically highlighted the importance of intraspecific comparisons. While studying the effect of sympatry on the evolution of the iridescent color pattern, we use this intraspecific comparison as a baseline to account for convergence or divergence of iridescence in a sympatric interspecific pair of Morpho, because under neutral evolution two subspecies are expected to be more similar than two different species (this assumption has been clarified line 147-148). We also used intraspecific mate choice to test for the use of visual cues in mate recognition (experiment 1) and to test what type of signal could be perceived by Morphos (the iridescent coloration or the iridescent pattern, experiment 2 and 3). These results help contextualize the interspecific mate choice, focused on determining whether visual cues could also be used in species recognition. Since we show that iridescent coloration is important in mate recognition at the intraspecific scale, it helps understand why species recognition is low at the interspecific scale because of wing color convergence between M. helenor and M. achilles.

(12) Line 154: 'signals on mate preferences'.

Corrected.

(13) Line 189: 'At the intraspecific level', maybe in the brackets include 'allopatric populations' just so the results are in a similar format as in the color contrast section below.

We added details to make clearer that the intraspecific level is studied between allopatric Morpho populations (line 189).

(14) Line 189-192: Please rearrange the figure (current B as A and vice versa) or present the results in order as in the figure (interspecific first and then intraspecific level).

We rearranged Figure 3 so that the intraspecific comparison (allopatric population) appears as A and the interspecific level (sympatric population) appears as B, to follow the order of presentation in the main text.

(15) Line 232: The motivation behind experiments 1, 2, and 3 is unclear. The authors have not made a strong point in the introduction about the need for these comparisons at an intraspecific level. Given that the authors are focused on divergence/convergence at an interspecific level, this set of experiments seems to be irrelevant to the present study. The implications of these findings are also not discussed.

We added motivation to the use of experiment 1, 2, and 3 in the introduction (lines 151-154) by stating that those experiments were used to assess whether blue color could indeed be used as a mating cue in Morpho helenor (experiment 1) and to try to understand what part of the visual signal is important in mate choice in Morpho helenor: the wing pattern (experiment 2) or the iridescent coloration (experiment 3). Although motivation for these experiments was not detailed in our manuscript, we already discussed the implications of the results of experiments 1, 2 and 3 in the discussion by stating that visual cues can take many forms and that considering both color AND pattern is important in understanding visual cues (lines 408-416). We carefully reworked this new version to make it more straightforward.

(16) Line 260: Insert 'wild-type' before model to ensure similar wording as in the previous section.

Corrected.

(17) Line 286: Insert 'sympatric' after mimetic.

Corrected.

(18) Line 307: Include a reference to the figures or table where these results are presented.

We now mention in the main text that the different proportions of beta-ocimene found between males M. helenor and M. achilles are shown in Table S2.

(19) Line 343: These inferences are speculative. Add a line here, something like 'although this warrants further research in this species'.

We detailed what additional experiments are needed lines 388-396.

(20) Line 357: The authors have not discussed their results on iridescence divergence in allopatric populations (line 190) and its implications.

We now made clear in the beginning of the discussion that the divergence of iridescence in allopatric populations is used as a baseline to test for convergent iridescence between species (lines 339-343).

(21) Line 361 onwards: This first paragraph is a bit confusing, as the results mainly focus on allopatry, while the title refers to sympatry.

To avoid confusion between the title and the content of the discussion, we divided the last part of the discussion into two different parts. As the first paragraph mainly focus on allopatry, we isolated it and titled it “Iridescent color patterns can be used as mate recognition cues in M. helenor” (line 498). The next paragraph of the discussion, focusing on the sympatric Morpho populations, has been titled “Evolution of visual and olfactory cues in mimetic sister-species living in sympatry” (line 418).

(21)  Line 383: visual cues 'as' poor species.

Corrected.

(23) Line 405: Why females here and not males? This is again confusing since the authors tested for male mate choice in the main experiments. Some background information on sex-specific mate choice in the methods might help.

In this specific sentence, we talk about performing mate choice experiments to test for the discrimination of olfactory cues by females (and not males) because we found a high divergence in the chemical compounds found on male genitalia. Although female chemical compounds could also be used as a cue by males in mate recognition, olfactive mate choice is often driven by female choice in butterflies. We recognize that this perspective does not line up with the mate choice presented in our results section which focused on male mate choice based on visual cues, because of ecological reasons (Morpho males tend to be attracted to bright blue colorations but not females) and technical reasons (in cages, females tend to hide away from the males or male dummies, and this behavior is not compatible with experiments involving flying around false males). In the discussion, we made sure to precise that the perspective we cite here is about testing the implications of divergence in male olfactory cues (line 454). We also added motivation to why we chose to investigate male (and not female) mate choice based on visual cues in the methods (lines 613-618) and in the results (219-223).

(24) Line 417: This inference is speculative. Consider toning it down.

We rewrote the sentence: “We find evidence of converging iridescent patterns in sympatry suggesting that predation could play a major role in the evolution of iridescence. Further work is nevertheless needed to directly test this hypothesis and establish the important of evasive mimicry in Morpho” (lines 465-468).

(25) Line 429: 'Convergent trait evolution leads to mutualistic interactions enhancing coexistence'. Careful here. It is not very evident how convergent trait evolution (iridescence) is mutualistic in this case, as there is no experimental evidence for evasive mimicry yet. Consider rewording or toning this sentence down.

We agree with the reviewer and removed this statement, only keeping the end of the sentence: “Altogether, this study addresses how convergence in one trait as a result of biotic interactions may alter selection on traits in other sensory modalities, resulting in a complex mosaic of biodiversity. (lines 479-481).

(26) Line 442: Since the samples come from a breeding farm, I have a few questions. How are the authors sure about the location where the specimens were collected? How long have they been kept in captivity? Have they been subjected to any artificial selection? More details are needed here.

Since M. helenor bristowi and M. helenor theodorus are only found in the wild in West and East Ecuador respectively, those M. helenor subspecies can only be collected in those two allopatric populations. Their phenotype is directly linked to their geographic repartition, this is how we made sure about their collect location. M. h. theodorus we used in this study were caught in East Ecuador in Tena, and M. h. bristowi were caught in West Ecuador in Pedro Vincente Madonado. We received pupae from the breeding farm, meaning that the Morpho used for the experiments were raised in captivity since their date of emergence. Upon emergence, they were transferred into cages for 4 to 5 days to wait for sexual maturity before performing the tetrad and mate choice experiments. This information was added to the method (lines 490-496).

(27) Line 476: Include some citations supporting this statement.

We now cite Bennett and Théry (2007), reviewing avian color vision, and Briscoe (2008), characterizing the sensitivity of the photoreceptors found in the eyes of butterflies. Both citations show that the 300-700nm range is seen by avian and butterfly visual systems.

(28) Line 480 onwards: Please clarify if the analysis used only one value (mean?) per species, sex, angle of measurement, and locality or included data from multiple individuals.

The analyses of both colorimetric variables and global iridescence were performed using iridescence data from multiple individuals (10 males and 10 females from M. h. bristowi, M. h. theodorus, M. h. helenor and M. a. achilles), for which we measured iridescence at 21 angles of illumination. Sampling size are mentioned lines 507, 515, 540-542.

(29) Line 510: Is there a specific reason that authors did not investigate achromatic contrasts? Provide some justification here. Or include the results of achromatic contrasts in the supplement.

We added the achromatic results in the supplement and in the results (lines 200-204). For both the avian visual model and the Morpho visual model, the confidence intervals always overlapped with the JND threshold, showing that neither birds nor butterflies could theoretically discriminate the wing reflectance brightness in allopatric and sympatric populations.

(30) Line 552 onwards: I may have missed it. It is not entirely clear why the authors focused on male mate choice rather than female preference for visual cues. The authors should explicitly justify this choice and cite previous studies demonstrating that male mate choice, rather than female preference, is important in this species. This should be stated in the results section as well.

We added a paragraph in the method (lines 613-618) to describe the ecological and technical reasons leading to testing only male mate choice using visual cues (also see our response to recommendation #23).

(31) Line 537 onwards: What was the criterion used to score that mating had occurred? Why first mating and not how long they were mating? Please add these details.

We stopped the experiment as soon as a male/female pair was formed by joining their genitalia (we added this information in the method lines 599-600). Since the tetrad experiment involves the interaction of two males and two females from different subspecies, we considered that mate choice happened before the formation of any couple, and is not necessarily dependent on how long they mate by observing their mating behavior. For instance, we witnessed avoidance behaviors from females that systematically hide their genitalia and refused to join their abdomen to some males, while being very ‘open’ to others (but did not quantify it).  

(32) Line 571: The authors used a black permanent marker to modify wing patterns but did not validate whether butterflies perceive these modifications as equivalent to natural coloration. It is possible that the alterations introduced unintended visual cues and may explain why most males rejected the dummies (line 267). The authors should acknowledge this limitation here.

We now acknowledge this limitation in the method (lines 638-639) and in the results section (lines 278-283).

(33) Line 591: Insert 'above' after protocol.

Corrected.

(34) Line 605: If the authors included random effects in their model, then it should be generalized linear mixed model (GLMM) and not GLM as they wrote.

We indeed included a random effect in our model accounting for male ID and trial number, we thus replaced “GLM” by “GLMM” in the manuscript.

(35) Line 615: This set of analyses does not seem to account for pseudo-replication, as the data were recorded from the same male more than once (Line 583). Please clarify and redo the analysis with the GLMM framework

We run new analyses using the GLMM framework: we used a binomial GLMM to test whether individuals preferentially interacted with dummy 1 vs. dummy 2 while accounting for pseudoreplication. The previously detected tendencies hold true with these new analyses, except for the visual mate discrimination of M. achilles: we now find statistical evidence that M. achilles tend to approach more their conspecifics during the mate choice experiment, although the signal is weak (line 297-307). Indeed, while we previously concluded that both species in sympatry (M. helenor and M. achilles) could not discriminate their conspecific mates, we now emphasize that M. achilles is somewhat sensitive to some visual signals. However, its estimated probability of approaching a conspecific is only 0.54, which is low compared to the estimated probability of approaching (0.61) or touching (0.84) a con-subspecific for M. bristowi. We thus concluded that even though some visual cues could be relevant for mate recognition, they are less reliable for male choice in sympatric populations were color patterns are more convergent, compared to allopatric populations. We thus updated Figure 4 and Figure S8 and S9, which are now picturing the probability of approaching or touching a conspecific or con-subspecific with the updated pvalues retrieved from the GLMM analyses. We also updated the results (line 297-307) and the discussion (lines 430-438) to bring nuance to our previous results.  

(36) Line 963: Figure 3D. Is there a particular reason for comparing allopatric populations only within Ecuador rather than between Ecuador and French Guiana for M. helenor? Please clarify.

We aimed at comparing the putative discrimination of blue coloration using visual models vs. what the butterflies actually discriminate using mate choice experiments. Since we only performed mate choice experiments involving M. h. bristowi x M. h. theodorus (allopatric populations within Ecuador) and M. h. helenor x M. a. achilles (sympatric population from Ecuador), we only looked at those comparisons using visual models. We added this precision lines (559-560).

(37) Line 980: Are these predicted probabilities or just mean proportions as written in line 614? Then the label should be changed to 'Proportion of approaches' or something similar.

Following our answer to recommendation #35, the points now represent the probability of touching a conspecific in the graph for each male, for every trial of every male tested. We corrected the legend of the figure. 

Reviewer #2 (Recommendations for the authors):

(1) Line 25: "...therefore facilitating co-existence in sympathy".

Corrected.

(2) Line 28: "contrasting" instead of contrasted.

Corrected.

(3) Line 33: begin a new sentence at the colon.

Corrected.

(4) Line 49: the phrase "habitat filtering" is unclear and should perhaps be defined or qualified.

We replaced “habitat filtering” by its definition and cited Keddy (1992), describing the community assembly rules and defining habitat filtering (line 46)

(5) Line 52: remove "even".

Corrected.

(6) Line 53: divergent suites may also result because traits are often constrained by genetic architecture (multivariate genetic covariances). This is discussed at length and specifically in relation to ornamental coloration by Kemp et al. 2023

We rewrote the introduction and focused on only reviewing the ecological interactions promoting trait divergence in sympatric species, and did not mention genetics in this paper.

(7) Line 87: (and throughout) refer to "colouration" or "colour pattern" rather than "colourations".

Corrected.

(8) Line 151: Remove "To do so,".

Corrected.

(9) Line 191: I would like to see the degrees of freedom for this test.

We added the F-statistic=2.09 and the degrees of freedom df=1 of this test, and for all the following tests.

(10) Line 201: (and throughout) replace "on" with "of".

Corrected.

(11) Line 205: modelling the visual properties of the wings allows one to infer what is theoretically visible/distinguishable. The modelling is useful but not necessarily definitive of vision/behaviour per se under different conditions in the wild. I therefore think it is appropriate to phrase the wording around the modelling approach more carefully. Perhaps refer to "theoretical" or "inferred" discriminability, or state (e.g.) that species should/should not be capable of perceiving differences based on the modelling data. You do this well in your wording of lines 207-209. This need not apply in the discussion because you're then dealing with the combination of modelling results and behaviour (mating trials).

We agree with the reviewer that visual modelling only allows to infer what is theoretically discriminated by the butterflies, and that the wording of our sentence is confusing. We therefore modified the sentence to account for those precisions: “Morpho butterflies and predators can theoretically visually perceive the difference in the blue coloration between different subspecies of M. helenor…… using both bird and Morpho visual models” (line 206-209).

(12) Line 222: Either the chi-square test or Fisher's exact test should be sufficient (why report both?)

Chi-square test relies on large-sample assumptions (expected counts>5) whereas Fischer’s exact test does not and is valid even with small or unbalanced sample sizes. Since the M. bristowi female/M. h. theodorus male paring only occurred 3 times, we do not meet the primary assumptions to apply a Chi-square test, although it is significant. We used a Fischer’s test to confirm the results. Using both and finding that both tests are significant shows that the results are robust, although they may appear redundant. To simplify, we remove the results of the Chisquare test and only keep the Fisher’s test in the methodology and the results.

(13) Line 224 (and throughout): Degrees of freedom should be provided for statistical tests.

We reported the statistic value and the degrees of freedom for all mentions of the statistical tests in the main text, except for the Fischer test which does not rely on an asymptotic distribution like the Chi-squared distribution as it is an exact test.

(14) Lines 266-267: This sentence has interest, but it is rather vague at present. Wouldn't your controls account for the effect of manipulation? This could be explained further.

During our mate choice experiments, all Morpho female dummies used for the experiments were painted with black markers, either on their dorsal blue band to modify their blue iridescent phenotype, or on their ventral side, thus controlling for the effect of manipulation. However, we cannot rule out that the modification of the dorsal blue iridescence could have had a “repulsive” effect for males for several reasons. For example, depending on the visual discrimination of darker colors by Morphos, the painted black band could have a slightly different color compared to the dark “brown” usually surrounding their blue iridescent patterns. We now explain this in the results (lines 278-283) and in the methodology (lines 638-639)  

(15) Line 316: I'm not certain that the similarity is best described as "striking", given a P-value of 0.084 for this contrast

We agree with the reviewer and removed this adjective for this line.

(16) Lines 387-390: This sentence is puzzling because, theoretically speaking, we should expect selection on visual preference to be heightened (not relaxed) in sympatry if colouration isincluded among the traits used in mate selection. I'm not certain I have understood the meaning here.

We would like to thank the reviewer for pointing out this typo. If shared predatory pressures favors convergent evolution of color pattern, then the visual signals become less reliable for species recognition. As a result, sexual selection on visual preference is heightened and becomes stronger, favoring the evolution of alternative cues used to discriminate conspecific mates. We changed the sentence and now write “the convergent evolution of iridescent wing patterns… may have negatively impact visual discrimination and favored the evolution of divergent olfactory cues” (lines 457-458).

(17) Line 529: Mating experiments. Given that these are quite large butterflies, I wondered whether a 3x3x2m cage would be sufficient in size to allow the expression of male courtship. A brief description of the courtship behaviour in these species or Morphos generally would be a useful addition to the paper.

A cage this size was enough for the males to express a flight behavior similar to what can be seen in nature, while also being able to see the females (live females or dummies). We tried to perform mate experiments in a larger cage (7m x 5m x 3m) but the trials were not conclusive because male did not find the dummies depending on where they were flying in the cage. A 3mx3mx2m cage is a good compromise maximizing interactions while still allowing enough space to fly. We now describe Morpho male behavior and female behavior in the methods (lines 613-618).

(18) Line 546: Why are both tests needed (chi-square AND Fisher's exact)?

Similarly to our answer on recommendations #12, were used both tests to show robustness in the statistical results. We only kept the Fisher’s test results to simplify the results.

eLife Assessment

This study presents important information about the role of mu opioid receptors in neurotransmission between the medial habenula and the interpeduncular nucleus. The authors provide convincing evidence that mu opioid receptor activation has differential effects on transmission from substance P neurons and cholinergic neurons, and that blockade of potassium channels can unmask a nicotinic cholinergic synaptic response. This work will be of high interest to those studying this brain region, and potentially to the larger neuroscience community studying motivated behavior.

Reviewer #1 (Public review):

Summary:

In this manuscript, the authors demonstrate for the first time that opioid signaling has opposing effects on the same target neuron depending on the source of the input. Further, the authors provide evidence to support the role of potassium channels in regulating a brake on glutamatergic and cholinergic signaling, with the latter finding being developmentally regulated and responsive to opioid treatment. This evidence solves a conundrum regarding cholinergic signaling in the interpeduncular nucleus that evaded elucidation for many years.

Strengths:

This manuscript provides 3 novel and important findings that significantly advance our understanding of the medial habenula-interpeduncular circuitry:

(1) Mu opioid receptor activation (mOR) reduces postsynaptic glutamatergic currents elicited from substance P neurons while simultaneously enhancing postsynaptic glutamatergic currents from cholinergic neurons, with the latter being developmentally regulated.

(2) Substance P neurons from the Mhb provide functional input to the rostral nucleus of the IPN, in addition to the previously characterized lateral nuclei.

(3) Potassium channels (Kv1.2) provide a break on neurotransmission in the IPN,

The findings here suggest that the authors have identified a novel mechanism for the normal function of neurotransmission in the IPN, so it would be expected to be observable in almost any animal. In the revised manuscript, the authors put forth significant effort to increase the n, thus increasing the confidence in the observations.

There are also significant sex differences in nAChR expression in the IPN that might not be functionally apparent using the low n presented here. In the revised manuscript, the authors increased the n, and provided data to the reviewers that no significant sex differences were apparent, although there was a trend. Future studies should examine sex differences in detail.

There are also some particularly novel observations that are presented but not followed up on, and this creates a somewhat disjointed story. For example, in Figure 2, the authors identify neurons in which no response is elicited by light stimulation of ChAT-neurons, but application of DAMGO (mOR agonist) un-silences these neurons. Are there baseline differences in the electrophysiological or morphological properties of these "silent" neurons compared to the responsive neurons? In the revised manuscript, the authors directly tested this with new experiments in SST+ neurons in the IPN, demonstrating convincingly that mOR activation unsilences these neurons.

With the revisions, the authors have addressed the reviewers' concerns and significantly improved the manuscript. I find no further weaknesses.

Reviewer #2 (Public review):

Summary:

In this paper, Chittajallu and colleagues present compelling evidence that mu opioid receptor (MOR) activation can potentiate synaptic neurotransmission in a medial habenula to interpeduncular nucleus (mHb-IPN) subcircuit. While, projections from mHb tachykinin 1 (Tac1) neurons onto lateral IPN neurons show a canonical opioid-induced synaptic depression in glutamate release, excitatory neurotransmission in mHb choline acetyltransferase (ChAT) projections to the rostral IPN is potentiated by opioids. This function emerges around age P27 in mice, when MOR expression in the IPN peaks.

Strengths:

Carefully executed electrophysiological experiments with appropriate controls. Interesting description of a neurodevelopmental change in the effects of opioids on mHb-IPN signaling.

Weaknesses:

A minor concern is that the genetic strategy used to target the mHb-IPN pathway (constitutive ChR2 expression in all ChAT+ and Tac1+ neurons) might not specifically target this projection. Future studies are needed to examine the precise mechanism whereby MOR signaling can potentiate glutamatergic neurotransmission in ChAT+ MHb-IPN projections."

Reviewer #3 (Public review):

Summary:

Here the authors describe the role of mORs in synaptic glutamate release from substance P and cholinergic neurons in the medial habenula to interpeduncular nucleus (IPN) circuit in adult mice. They show that mOR activation reduces evoked glutamate release from substance P neurons yet increases evoked glutamate release and Ach release from cholinergic neurons. Unlike glutamate release, Ach release is only detected when potassium channels are blocked with 4-AP or dendrotoxin. The authors also report a previously unidentified glutamatergic input to IPR from SP neurons and describe the developmental timing of mOR- facilitation in adolescent mice.

Strengths:

- The experiments provide new insight into the role of mORs in controlling evoked glutamate release in a circuit with high levels of mORs and established roles in relevant behaviors.

- The experiments are rigorous, and the results are clear cut. The conclusions are supported by the data.

- The findings will be of interest to those working in the field of synaptic transmission and those interested in the function of the medial habenula or interpeduncular nucleus, as well as those seeking to understand the role of opioids on normal and pathological behaviors.

Weaknesses:

- The mechanistic underpinnings of these interesting and novel results are not pursued.

Author response:

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

Reviewer #1 (Public Reviews):

Weaknesses: 

Overall I find the data presented compelling, but I feel that the number of observations is quite low (typically n=3-7 neurons, typically one per animal). While I understand that only a few slices can be obtained for the IPN from each animal, the strength of the novel findings would be more convincing with more frequent observations (larger n, more than one per animal). The findings here suggest that the authors have identified a novel mechanism for the normal function of neurotransmission in the IPN, so it would be expected to be observable in almost any animal. Thus,  it is not clear to me why the authors investigated so few neurons per slice and chose to combine different treatments into one group (e.g. Figure 2f), even if the treatments have the same expected effect.  

This is a well taken suggestion. However, we must  point out that we do perform statistical analyses on the original datasets and we believe that our conclusions are justified as acknowledged by the Reviewer. As the Reviewer is aware,  the IPN is a small nucleus and with the slicing protocol used, we typically attain 1-2 slices per mouse that are suitable for recordings. Since most of the experiments in the manuscript deals with some form of pharmacological interrogation, we were reticent to use slices that are not naïve and therefore in general did not perform more than 1 cell recording per slice. Having said this, to comply with the Reviewer’s suggestion we have now performed additional experiments to increase the n number for certain experiments. We have amended all figures and legends to incorporate the additional data. We must point out that during the replotting of the data in the summary Figure 8i (previously Figure 7i) we noticed an error with the data representation of the TAC IPL data and have now corrected this oversight  

Figure 2b,c. 

500nM DAMGO effect on TAC IPL AMPAR EPSC – n increased from 5 to 9

Figure 3g. 

500nM DAMGO effect on CHAT IPR AMPAR EPSC – n increased from 8 to 16 Effect of CTAP on DAMGO on CHAT IPR AMPAR EPSC – n increased from 4 to 7

Figure 3i. 

500nm DAMGO or Met-enk effect in “silent” CHAT IPR AMPAR EPSC – n increased    from 7 to 9

Figure 4e. 

500nM DAMGO effect on ES coupling – Note: in the original version the n number was 5 and not 7 as written in the figure legend. We have now increased the n from 5 – 9.

Figure 5e,f. 

500nM DAMGO effect on TAC IPR AMPAR EPSC – n increased from 5 to 9

Figure 7f.

Effect of DHE on EPSC amplitude after application of DNQX/APV/4-AP or DTX-α – n increased from 7-9.

Figure 7g.

Emergence of nAChR EPSC after DTX – n increased from 4 to 7

Figure 7i. 

Effect of ambenonium on nAChR amplitude and charge – n increased from 4 to 7

Supplementary Figure 3c and h

Effect of DAMGO after DNQX – n increased from 4 to 7

Effect of DNQX after DAMGO mediated potentiation – n increased from 3 to 5.

Throughout the study (Figs. 3i, 7f and 8h in the revised manuscript)  we do indeed pool datasets that were amassed from different conditions since we were not directly investigating the possibility of any deviation in the extent of response between said treatments. For example, and as pointed out by the Reviewer, in Fig. 2F (now Fig. 3i) the use of DAMGO and met-ENK were merely employed to ascertain whether light-evoked synaptic transmission (ChATCre:ai32 mice) in cells that had no measurable EPSC could be pharmacologically “unsilenced” by mOR activation. Thus, the means by which mOR receptor was activated was not relevant to this specific question. Note: 2 more recordings are now added to this dataset (Fig. 3i) that were taken from ChATChR2/SSTCre:ai9 mice in response to the comment by this Reviewer below (“Are there baseline differences in the electrophysiological or morphological properties of these "silent" neurons compared to the responsive neurons?”).  Similarly, in the revised Fig.7f we pooled data investigating the pharmacological block of the EPSC that emerged following application of either DNQX/APV/4-AP or DNQX/APV/DTX. Low concentrations 4-AP or DTX were interchangeably employed to reveal the DNQX-insensitive EPSC that we go on to show is indeed the nAChR response. Finally, in Fig. 8h, we pooled data demonstrating a  lack of effect of DAMGO in potentiating  both the glutamatergic and cholinergic arms of synaptic transmission in the OPRM1 KO mice. Again, here we were only interested in determining whether removal of mOR expression prevented potentiation of transmission mediated by mHB ChAT neurons irrespective of neurotransmitter modality.  Thus, overall we were careful to only pool data in those instances where it  would not change the interpretation and hence conclusions reached. 

There are also significant sex differences in nAChR expression in the IPN that might not be functionally apparent using the low n presented here. It would be helpful to know which of the recorded neurons came from each sex, rather than presenting only the pooled data.  

As the reviewer correctly states there are veins of literature concerning a divergence, based on sex, of not only nicotinic receptor expression but also behaviors associated with nicotine addiction. However, we have reanalyzed our datasets focusing on the extent of the mOR potentiation of glutamatergic and cholinergic transmission mediated by mHB ChAT neurons in IPR  between male and female mice. Please refer to the Author response image 1 below. Although there is a possible trend towards a higher potentiation of nAChR in female mice, this was not found to be of statistical significance (see Author response image 1 below). We therefore chose not to split our data in the manuscript based on gender.

Author response image 1.

Comparison of the mOR (500nM DAMGO) mediated potentiation on evoked (a) AMPAR and (b) nAChR  EPSCs in IPR between male and female mice.  

There are also some particularly novel observations that are presented but not followed up on, and this creates a somewhat disjointed story. For example, in Figure 2, the authors identify neurons in which no response is elicited by light stimulation of ChAT-neurons, but the application of DAMGO (mOR agonist) un-silences these neurons. Are there baseline differences in the electrophysiological or morphological properties of these "silent" neurons compared to the responsive neurons?  

Unfortunately, we did not routinely measure intrinsic properties of the recorded postsynaptic neurons nor systematically recovered biocytin fills to assess morphology. Therefore, it remains unclear whether the  neurons in which there were none or minimal AMPAR-mediated EPSCs are distinct to the ones displaying measurable responses. The IPR is resident to GABAergic SST neurons that comprise the most numerous neuron type in this IPN subdivision. Although heavily outnumbered by the SST neurons there are additionally VGluT3+ glutamatergic neurons in IPN. The Reviewer is likely referring to a recent study investigating synaptic transmission specifically onto  SST+ and VGluT3+ neurons in IPN demonstrating that mHB cholinergic mediated glutamatergic input is “weaker” onto the glutamatergic neurons. Furthermore, in some instances synaptic transmission onto this latter population can be “unsilenced” by GABAB receptor activation in a similar manner to that seen with mOR activation in this manuscript when IPR neurons are blindly targeted(Stinson & Ninan, 2025).  Using a similar strategy as in this recent study(Stinson & Ninan, 2025), we now include experiments in which the ChATChR2 mouse was crossed with  a SSTCre:Ai14. This allowed for recording of postsynaptic EPSCs in directly identified SST IPR neurons. We demonstrate that DAMGO can indeed increase glutamatergic EPSCs and in 2 of the cells where light activation demonstrated no appreciable AMPAR EPSC upon maximal LED light activation, DAMGO clearly “unsilenced” transmission.  Thus, our additional analyses directly demonstrate that our original observations concerning mOR modulation extend to the mHb cholinergic AMPAR mediated input onto IPR SST neurons. This additional data is in the revised manuscript (Figure 3D-F, I). Future experimentation will be required to determine if the propensity of encountering a  “silent” input that can be converted to robust synaptic transmission by mOR differs between these two cell types. Furthermore, it will be of interest to investigate if any differences exist in the magnitude of the cholinergic input or the mOR mediated potentiation of co-transmission between postsynaptic SST GABA and glutamatergic neuronal subtypes. 

Reviewer #2 (Public review)

Weaknesses: 

The genetic strategy used to target the mHb-IPN pathway (constitutive expression in all ChAT+ and Tac1+ neurons) is not specific to this projection.  

This is an important point made. We are acutely aware that the source of the synaptic input in IPN mediated by conditional expression of ChR2 employing  using transgenic cre driver lines does not confer specificity to mHB. This is particularly relevant considering one of the novel observations here relates to  a previously unidentified functional input from TAC1 neurons to the IPR. At this juncture we would like to point the Reviewer to the publicly available Connectivity Atlas provided by the Allen Brain Institute (https://connectivity.brain-map.org/). With reference to mHB TAC1 neuronal output, targeted viral injection into the habenula of Tac1Cre mice allows conditional expression of EGFP to SP neurons as evidenced by the predominant expression of reported fluorescence in dorsal mHB (see Author response image 2 a,b below). Tracing the axonal projections to the IPN clearly demonstrates dense fibers in IPL as expected but also arborization in  IPR (Author response image 2 a,c) . This pattern is reminiscent of that seen in the transgenic Tac1Cre:ai9 or ai32 mice used in the current study (Figs. 1c, 2a, 5c). Closer inspection of the fibers in the IPR reveals putative synaptic bouton like structures as we have shown in Fig. 5a,b (Author response image 2 d below).

Author response image 2.

Sterotaxic viral injection into mHB pf Tac1Cre mice taken from Allen Brain connectivity atlas (Link to Connectivity Atlas for mHb SP neuronal projection pattern)

These anatomical data suggest that part of the synaptic input to the IPR originates from mHB TAC1 neurons although we cannot fully discount additional synaptic input from other brain areas that may impinge on the IPR. Indeed, as the Reviewer points out, it is evident that other regions including the nucleus incertus send outputs to the IPN(Bueno et al., 2019; Liang et al., 2024; Lima et al., 2017). However, it is unclear if neuronal inputs from these alternate sources {Liang, 2024 #123;Lima, 2017 #33}{Bueno, 2019 #178} are glutamatergic in nature AND mediated by a TAC1/OPRM1-expressing neuronal population. Nevertheless, we have now modified text in the discussion to highlight the limitations of using a transgenic strategy (pg 12, para 1).

In addition, a braking mechanism involving Kv1.2 has not been identified.

It is unclear to what the Reviewer is referring to here. Although most of our experiments pertaining to the brake on cholinergic  transmission by potassium channels use low concentrations of 4-AP (50100M) which have been used to block Shaker Kv1 channels there although at these concentrations there are additional action at other K+-channels such as Kv3, for instance. However, we essentially demonstrate that a selective Kv1.1 and Kv1.2 antagonist dendrotoxin replicates the 4-AP effects. We have now also included RNAseq data demonstrating the relative expression levels of Kv1 channel mRNA in mHb ChAT neurons (KCNA1 through KCNA6; Figure 6b). The complete absence of KCNA1 yet a high expression level of KCNA2 transcripts highly suggests a central role of Kv1.2 in unmasking nAChR mediated synaptic transmission. 

Reviewer #3 (Public review)

Weaknesses:  

The significance of the ratio of AMPA versus nACh EPSCs shown in Figure 6 is unclear since nAChR EPSCs measured in the K+ channel blockers are compared to AMPA EPSCs in control (presumably 4-AP would also increase AMPA EPSCs). 

We understand the Reviewer’s concern regarding the calculation of nicotinic/AMPA ratios since they are measured under differing conditions i.e. absence and presence of 4-AP, respectively. As the reviewer correctly points point 4-AP likely increases the amplitude of the AMPA receptor mediated EPSC. However, our intention of calculating this ratio was not to ascertain a measure of relative strengths of fast glutamatergic vs cholinergic transmission onto a given postsynaptic IPN neuron per se. Rather, we used the ratio as a means to normalize the size of the nicotinic receptor EPSC to the strength of the light stimulation (using the AMPA EPSC as the normalizing factor) in each individual recording. This permits a more meaningful comparison across cells/slices/mice . We apologize for the confusion and have amended the text in the results section to reflect this (pg 9; para2).

The mechanistic underpinnings of the most now  results are not pursued. For example, the experiments do not provide new insight into the differential effects of evoked and spontaneous glutamate/Ach release by Gi/o coupled mORs, nor the differential threshold for glutamate versus Ach release. 

Our major goal of the current manuscript was to provide a much-needed roadmap outlining the effects of opioids in the habenulo-interpeduncular axis. Of course, a full understanding of the mechanisms underlying such complex opioid actions at the molecular level will be of great value. We feel that this is beyond the scope of this already quite result dense manuscript but will be essential if directed manipulation of the circuit is to be leveraged to alter maladaptive behaviors associated with addiction/emotion during adolescence and in adult. 

The authors note that blocking Kv1 channels typically enhances transmitter release by slowing action potential repolarization. The idea that Kv1 channels serve as a brake for Ach release in this system would be strengthened by showing that these channels are the target of neuromodulators or that they contribute to activity-dependent regulation that allows the brake to be released. 

The exact mechanistic underpinnings that can potentially titer Kv1.2 availability and hence nAChR transmission would be essential to shed light on potential in vivo conditions under which this arm of neurotransmission can be modulated. However, we feel that detailed mechanistic interrogation constitutes significant work but one that future studies should aim to achieve. Thus, it presently remains unclear under what physiological or pathological scenarios result in attenuation of Kv1.2 to subsequently promote nAChR mediated transmission but as mentioned in the existing discussion future work to decipher such mechanisms would be of great value.

Reviewer #1 (Recommendations for the authors): 

Overall I find this to be a very interesting and exciting paper, presenting novel findings that provide clarity for a problem that has persisted in the IPN field: that of the conundrum that light-evoked cholinergic signaling was challenging to observe despite the abundance of nAChRs in the IPN. 

Major concerns: 

(1) The n is quite low in most cases, and in many instances, data from one figure are replotted in another figure. Given that the findings presented here are expected in the normal condition, it should not be difficult to increase the n. A more robust number of observations would strengthen the novel findings presented here. 

Please refer to the response to the public review above.

(2) In general, I find the organization of the figures somewhat disjointed. Sometimes it feels as if parts of the information presented in the results are split between figures, where it would make more sense to be together in a figure. For example, all the histology for each of the lines is in Figure 1, but only ephys data for one line is included there. It would be more logical to include the histology and ephys data for each line in its own figure. It would also be helpful to show the overlap of mOR expression with Tac1-Cre and ChAT-Cre terminals in the IPN. Likewise, the summarized Tac1Cre:Ai32 IPR data is in Figure 4, but the individual data is in Figure 5. 

We introduce both ChAT and TAC1 cre lines in Figure 1 as an overview particularly for those readers who are not entirely familiar with the distinct afferent systems operating with the habenulointerpeduncular pathway.  However, in compliance with the Reviewer’s suggestion we have now restructured the Figures. In the revised manuscript, the functional data pertaining to the various transmission modalities mediated by the distinct afferent systems impinging on the subdivision of the IPN tested are now split into their own dedicated figure as follows:

Figure 2. 

mOR effect on TAC1neuronal glutamatergic output in IPL.

Figure 3. 

mOR effect on CHAT neuronal glutamatergic output in IPR.

Figure 5. 

mOR effect on TAC1neuronal glutamatergic output in IPR.

Figure 8.

mOR effect on CHAT neuronal cholinergic output in IPC.

Supp. Fig. 1 mOR effect on CHAT neuronal glutamatergic output in IPC.

We thank the Reviewer for their suggestions regarding the style of the manuscript. The restructuring has now resulted in a much better flow of the presented data.

(3) The discussion is largely satisfactory. However, a little more discussion of the integrative function of the IPN is warranted given the opposing effects of MOR activation in the Tac vs ChAT terminals, particularly in the context of both opioids and natural rewards. 

We thank the reviewer for this comment. However, we feel the discussion is rather lengthy as is and therefore we refrained from including additional text.  

Minor concerns: 

(1)  The methods are missing key details. For example, the stock numbers of each of the strains of mice appear to have been left out. This is of particular importance for this paper as there are key differences between the ChAT-Cre lines that are available that would affect observed electrophysiological properties. As the authors indicate, the ChAT-ChR2 mice overexpress VAChT, while the ChAT-IRES-Cre mice do not have this problem. However, as presented it is unclear which mice are being used. 

We apologize for the omission - the catalog numbers of the mice employed have now been included in the methods section.

We have now clearly included in each figure panel (single trace examples and pooled data) from which mice the data are taken from – in some instances the pooled data are from the two CHAT mouse strains employed. Despite the tendency of the ChATChR2 mice to demonstrate more pronounced nAChR mediated transmission (Fig. 7h),  we justify pooling the data since we see no statistical significance in the effect of mOR activation on either potentiating AMPA or nAChR EPSCs (Please refer to response to Reviewer 2, Minor Concern point 2)

(2) Likewise, antibody dilutions used for staining are presented as both dilution and concentration, which is not typical. 

We thank the reviewer for pointing out this inconsistency. We have amended the text in the methods to include only the working dilution for all antibodies employed in the study.

(3) There are minor typos throughout the manuscript. 

All typos have been corrected.

Reviewer #2 (Recommendations for the authors): 

The authors provide a thorough investigation into the subregion, and cell-type effect of mu opioid receptor (MOR) signaling on neurotransmission in the medial habenula to interpeduncular nucleus circuit (mHb-IPN). This circuit largely comprises two distinct populations of neurons: mHb substance P (Tac1+) and cholinergic (ChAT+) neurons. Corroborating prior work, the authors report that Tac1+ neurons preferentially innervate the lateral IPN (IPL) and rostral IPN (IPR), while ChAT+ neurons preferentially innervate the central IPN (IPC) and IPR. The densest expression of MOR is observed in the IPL and MOR agonists produce a canonical presynaptic depression of glutamatergic neurotransmission in this region. Interestingly, MOR signaling in the ChAT+ mHb projection to the IPR potentiates light-evoked glutamate and acetylcholine-mediated currents (EPSC), and this effect is mediated by a MOR-induced inhibition of Kv2.1 channels. 

Major concerns: 

(1) The method used for expressing channelrhodopsin (ChR2) into cholinergic and neurokinin neurons in the mHb (Ai32 mice crossed with Cre-driver lines) has limitations because all Tac1+/ChAT+ inputs to the IPN express ChR2 in this mouse. Importantly, the IPN receives inputs from multiple brain regions besides the IPN-containing neurons capable of releasing these neurotransmitters (PMID: 39270652). Thus, it would be important to isolate the contributions of the mHb-IPN pathway using virally expressed ChR2 in the mHb of Cre driver mice. 

Please refer to the response to the public review above. 

(2) Figure 4: The authors conclude that the sEPSC recorded from IPR originate from Tac1+ mHbIPR projections. However, this cannot be stated conclusively without additional experimentation. For instance, an optogenetic asynchronous release experiment. For these experiments it would also be important to express ChR2 virus in the mHb in Tac1- and ChAT-Cre mice since glutamate originating from other brain regions could contribute to a change in asynchronous EPSCs induced by DAMGO. 

This is a well taken point. The incongruent effect of DAMGO on evoked CHAT neuronal EPSC amplitude and sEPSC frequency prompted us  to consider the the possibility of differing effect of DAMGO on a  secondary input. We agree that we do not show directly if the sEPSCs originate from a TAC1 neuronal population. Therefore, we have tempered our wording with regards the origin of the sEPSCs and  have also restructured the Figure in question moving the sEPSC data into supplemental data (Supplemental Fig. 2) 

(3) Figure 5D: lt would be useful to provide a quantitative measure in a few mice of mOR fluorescence across development (e.g. integrated density of fluorescence in IPR). 

We have now included mOR expression density across development  (Fig. 6). Interestingly, the adult expression levels of mOR in the IPR are essentially reached at a very early developmental age (P10) yet we see stark differences in the role of mOR activation in modulating glutamatergic transmission mediated by mHB cholinergic neurons. Note: since we processed adult tissue (i.e. >p40) for these developmental analyses we utilized these slices to also include an analysis of the relative mOR expression density specifically in adults between the subdivisions of IPN in Fig. 1.

(4) Figure 6B: It would be useful to quantify the expression of Kcna2 in ChAT and Tac1 neurons (e.g. using FISH). 

We thank the Reviewer for this suggestion. We have now included mRNA expression levels available from publicly available 10X RNA sequencing dataset provided by the Allen Brain Institute (Figure 7b).  

(5) It would be informative to examine what the effects of MOR activation are on mHb projections to the (central) . 

In response to this suggestion, we now have included  additional data in the manuscript in putative IPC cells that clearly demonstrate a similar DAMGO elicited potentiation of AMPAR EPSC to that  seen in IPR. These data are now included in the revised manuscript  (Supplemental Fig. 1; Fig. 8i). 

(6) What is the proposed link between MOR activation and the inhibition of Kv1.2 (e.g. beta-Arrestin signaling, G beta-gamma interaction with Kv1.2, PKA inhibition?) 

We apologize for any confusion. We do not directly test whether the potentiation of EPSCs upon mOR activation occurs via inhibition of Kv1.2.Although we have not directly tested this possibility we find it an unlikely underlying cellular mechanism, especially for the potentiation of the cholinergic arm of neurotransmission since in the presence of DNQX/APV, the activation of mOR does not result in any emergence of any nAChR EPSC (see Supplementary Fig. 3a-c)

Minor concerns: 

(1) Methods: Jackson lab ID# for used mouse strains is missing. 

We apologize for this omission and have now included the mouse strain catalog numbers.

(2) The authors use data from both ChAT-Cre x Ai32 and ChAT-ChR2 mice. It would be helpful to show some comparisons between the lines to justify merging data sets for some of the analyses as there appear to be differences between the lines (e.g. Figure 6G). 

This is a well taken point. We have now provided a figure for the Reviewer (see below) that illustrates the lack of  significant difference between the mOR mediated potentiation of both mHB CHAT neuronal AMPAR and nAChR transmission between the two mouse lines employed despite a divergence in the extent of glutamatergic vs cholinergic transmission shown in Fig. 7g (previously Figure 6g). We have chosen not to include this data in the revised manuscript.

Author response image 3.

Comparison of the mOR (500nM DAMGO) mediated potentiation on evoked AMPAR (a) and nAChR (b)EPSCs in IPR between ChATCre:Ai32  and ChATChR2 mice.

(3)  Line 154: How was it determined that the EPSC is glutamatergic? 

We apologize for any confusion. In the revised manuscript we now clearly point to the relevant figures (see Supplementary Figs. 2a and 3) in the Results section (pg. 4, para 2; pg 7, para 1; pg 8, para2) where we determine that both the sEPSCs and ChAT mediated light evoked EPSCs recorded under baseline conditions are totally blocked by DNQX and hence are exclusively AMPAR events 

(4) It would be helpful to discuss the differences between GABA-B mediated potentiation of mHbIPN signaling and the current data in more detail. 

We are unclear as to what differences the Reviewer is referring to. At least from the perspective of ChAT neuronal mediated synaptic transmission, other groups (and in the current study; Fig. 7h) have clearly shown that GABA_B activation markedly potentiates synaptic transmission like mOR activation. Nevertheless, based on our novel findings it would be of interest to determine whether the influence of GABA_B is inhibitory onto the TAC mediated input in IPR and whether there is a developmental regulation of this effect as we demonstrate upon mOR activation. These additional comparisons between the effect of the two Gi-linked receptors may shed light onto the similarity, or lack thereof, regarding the underlying cellular mechanisms. We now have included a few sentences in the discussion to highlight this (pg 11, para 1).

Reviewer #3 (Recommendations for the authors): 

The abstract was confusing at first read due to the complex language, particularly the sentence starting with... Further, specific potassium channels... 

The authors might want to consider simplifying the description of the experiments and the results to clarify the content of the manuscript for readers who many only read the abstract. 

We have altered the wording of the abstract and hope it is now more reader friendly.

The opposite effect of mOR activation on spontaneous EPSCs versus electrical or ChR2-evoked EPSCs is very interesting and raises the issue of which measure is most physiologically relevant. For example, it is unclear whether sEPSCs arise primarily from cholinergic neurons (that are spontaneously active in the slice, Figure 3), and if so, does mOR activation suppress or enhance cholinergic neuron excitability and/or recruitment by ChR2? While a full analysis of this question is beyond the scope of this manuscript, the assumption that glutamate release assayed by electrical/ChR2 evoked transmission is the most physiologically relevant might merit some discussion since sEPSCs presumably also reflect action-potential dependent glutamate release. One wonders whether mORs hyperpolarize cholinergic neurons to reduce spontaneous spiking yet enhance fiber recruitment by ChR2 or an electrical stimulus (i.e. by removing Na channel inactivation). The authors have clearly stated that they do not know where the mORs are located, and that the effects arising from disinhibition are likely complex. But they also might discuss whether glutamate release following synchronous activation of a fiber pathway by ChR2 or electrode is more or less physiologically relevant than glutamate release assayed during spontaneous activity. It seems likely that an equivalent experiment to Figure 3D, E using spontaneous spiking of IPR neurons would show that spiking is reduced by mOR activation. 

We thank the Reviewer for this comment. As pointed it would be of interest to dissect the “network” effect of mOR activation but as the Reviewer acknowledges this is beyond the scope of the current manuscript. The Reviewer is correct in postulating that mOR activation results in hyperpolarization of mHB ChAT neurons.  A recent study(Singhal et al 2025) demonstrate that a subpopulation of ChAT neurons undergoes a reduction in firing frequency following DAMGO application. This is corroborated by our own observations although we chose not to include this data in our current manuscript (but see below).

Additionally, the Reviewer questions whether ChR2/electrical stimulation is physiological. This is a well taken point and of course the simultaneous activation of potentially all possible axonal release sites is not the mode under which the circuit operates. Nevertheless, our data clearly demonstrates the ability of mORs to modulate release under these circumstances that must reflect an impact on spontaneous action potential driven evoked release.  Although the suggested experiment  could shed light on the synaptic outcomes of mOR receptor activation on ES coupling of downstream IPN neurons. Interpretation of the outcome would be confounded by the fact that postsynaptic IPN neurons also express mORs . Thus,  we would not be able to isolate the effects of presynaptic changes in modulating ES coupling from any direct postsynaptic effect on the recorded cell when in current clamp. 

Together these additional sites of action of mOR (i.e. mHB ChAT somatodendritic and postsynaptic IPN neuron) only serve to further highlight the complex nature of the actions of opioids on the habenulo-interpeduncular axis warranting  future work to fully understand the physiological and pathological effects on the habenulo-interpeduncular axis as a whole.

The idea that Kv2.1 channels serve as a brake raises the question of whether they contribute to activity-dependent action potential broadening to facilitate Ach release during trains of stimuli. 

This is an interesting suggestion and one that we had considered ourselves. Indeed, as the Reviewer is likely aware and as mentioned in the manuscript, previous studies have shown nAChR signaling can be revealed under conditions of multiple stimulations given at relatively high frequencies.  We therefore attempted to perform high frequency stimulation (20 stimulations at 25Hz and 50Hz) in the presence of ionotropic glutamatergic receptor antagonists DNQX and APV. We have now included this data in the revised manuscript (Supplementary Fig 3b). As shown, this failed to engage nAChR mediated synaptic transmission in our hands. Interestingly there is evidence from reduced expression systems demonstrating that Kv1.2 channels undergo use-dependent potentiation(Baronas et al., 2015) in contrast to that seen with other K+-channels. Whether this is the case for the axonal Kv1.2 channels on mHB axonal terminals in situ is not known but this may explain the inability to reveal nAChR EPSCs upon delivery of such stimulation paradigms.  

References 

Baronas, V. A., McGuinness, B. R., Brigidi, G. S., Gomm Kolisko, R. N., Vilin, Y. Y., Kim, R. Y., … Kurata, H. T. (2015). Use-dependent activation of neuronal Kv1.2 channel complexes. J Neurosci, 35(8), 3515-3524. doi:10.1523/JNEUROSCI.4518-13.2015

Bueno, D., Lima, L. B., Souza, R., Goncalves, L., Leite, F., Souza, S., … Metzger, M. (2019). Connections of the laterodorsal tegmental nucleus with the habenular-interpeduncular-raphe system. J Comp Neurol, 527(18), 3046-3072. doi:10.1002/cne.24729

Liang, J., Zhou, Y., Feng, Q., Zhou, Y., Jiang, T., Ren, M., … Luo, M. (2024). A brainstem circuit amplifies aversion. Neuron. doi:10.1016/j.neuron.2024.08.010

Lima, L. B., Bueno, D., Leite, F., Souza, S., Goncalves, L., Furigo, I. C., … Metzger, M. (2017). Afferent and efferent connections of the interpeduncular nucleus with special reference to circuits involving the habenula and raphe nuclei. J Comp Neurol, 525(10), 2411-2442. doi:10.1002/cne.24217

Singhal, S. M., Szlaga, A., Chen, Y. C., Conrad, W. S., & Hnasko, T. S. (2025). Mu-opioid receptor activation potentiates excitatory transmission at the habenulo-peduncular synapse. Cell Rep, 44(7), 115874. doi:10.1016/j.celrep.2025.115874

Stinson, H.E., & Ninan, I. (2025). GABA(B) receptor-mediated potentiation of ventral medial habenula glutamatergic transmission in GABAergic and glutamatergic interpeduncular nucleus neurons. bioRxiv doi.10.1101/2025.01.03.631193

eLife Assessment

This important study elucidates the role of the exocyst component EXOC6A at distinct stages of ciliogenesis, which advances our understanding of ciliary membrane remodeling and cilium formation. The authors provide solid evidence that EXOC6A interacts with myosin-Va and is dynamically recruited via dynein-, microtubule-, and actin-dependent mechanisms, to support proper formation of the ciliary membrane. The study will be of interest to cell biologists and other researchers interested in vesicular trafficking, organellar membrane dynamics, and ciliogenesis.

Reviewer #1 (Public review):

Summary:

The study by Lin et al. studies the role of EXOC6A in ciliogenesis and its relationship with the interactor myosin-Va using a range of approaches based on the RPE1 cell line model. They establish its spatio-temporal organization at centrioles, the forming ciliary vesicle and ciliary sheath using ExM, various super-resolution techniques, and EM, including correlative light and electron microscopy. They also perform live imaging analyses and functional studies using RNAi and knockout. They establish a role of EXOC6A together with myosin-Va in Golgi-derived, microtubule- and actin-based vesicle trafficking to and from the ciliary vesicle and sheath membranes. Defects in these functions impair robust ciliary shaft and axoneme formation due to defective transition zone assembly.

Strengths:

The study provides very high-quality data that support the conclusions. In particular, the imaging data is compelling. It also integrates all findings in a model that shows how EXOC6A participates in multiple stages of ciliogenesis and how it cooperates with other factors.

Weaknesses:

The precise role of EXOC6A remains somewhat unclear. While it is described as a component of the exocyst, the authors do not address its molecular functions and whether it indeed works as part of the exocyst complex during ciliogenesis.

Reviewer #2 (Public review):

Summary:

The molecular mechanisms underlying ciliogenesis are not well understood. Previously, work from the same group (Wu et al., 2018) identified myosin-Va as an important protein in transporting preciliary vesicles to the mother vesicles, allowing for initiation of ciliogenesis. The exocyst complex has previously been implicated in ciliogenesis and protein trafficking to cilia. Here, Lin et al. investigate the role of exocyst complex protein EXOC6A in cilia formation. The authors find that EXOC6A localizes to preciliary vesicles, ciliary vesicles, and the ciliary sheath. EXOC6A colocalizes with Myo-Va in the ciliary vesicle and the ciliary sheath, and both proteins are removed from fully assembled cilia. EXOC6A is not required for Myo-Va localization, but Myo-VA and EHD1 are required for EXOC6A to localize in ciliary vesicles. The authors propose that EXOC6A vesicles continually remodel the cilium: FRAP analysis demonstrates that EXOC6A is a dynamic protein, and live imaging shows that EXOC6A fuses with and buds off from the ciliary membrane. Loss of EXOC6A reduces, but does not eliminate, the number of cilia formed in cells. Any cilia that are still present are structurally abnormal, with either bent morphologies or the absence of some transition zone proteins. Overall, the analyses and imaging are well done, and the conclusions are well supported by the data. The work will be of interest to cell biologists, especially those interested in centrosomes and cilia.

Strengths:

The TEM micrographs are of excellent quality. The quality of the imaging overall is very good, especially considering that these are dynamic processes occurring in a small region of the cell. The data analysis is well done and the quantifications are very helpful. The manuscript is well-written and the final figure is especially helpful in understanding the model.

Weaknesses:

Additional information about the functional and mechanistic roles of EXOC6A would improve the manuscript greatly.

Reviewer #3 (Public review):

Summary:

Lin et al report on the dynamic localization of EXOC6A and Myo-Va at pre-ciliary vesicles, ciliary vesicles, and ciliary sheath membrane during ciliogenesis using three-dimensional structured illumination microscopy and ultrastructure expansion microscopy. The authors further confirm the interaction of EXOC6A and Myo-Va by co-immunoprecipitation experiments and demonstrated the requirement of EHD1 for the EXOC6A-labeled ciliary vesicles formation. Additional experiments using gene-silencing by siRNA and pharmacological tools identified the involvement of dynein-, microtubule-, and actin in the transport mechanism of EXOC6A-labeled vesicles to the centriole, as they have previously reported for Myo-Va. Notably, loss of EXOC6A severely disrupts ciliogenesis, with the majority of cells becoming arrested at the ciliary vesicle (CV) stage, highlighting the involvement of EXOC6A at later stages of ciliogenesis. As the authors observe dynamic EXOC6A-positive vesicle release and fusion with the ciliary sheath, this suggests a role in membrane and potentially membrane protein delivery to the growing cilium past the ciliary vesicle stage. While CEP290 localization at the forming cilium appears normal, the recruitment of other transition zone components, exemplified by several MKS and NPHP module components, was also impaired in EXOC6A-deficient cells.

Strengths:

(1) By applying different microscopy approaches, the study provides deeper insight into the spatial and temporal localization of EXOC6A and Myo-Va during ciliogenesis.

(2) The combination of complementary siRNA and pharmacological tools targeting different components strengthens the conclusions.

(3) This study reveals a new function of EXOC6A in delivering membrane and membrane proteins during ciliogenesis, both to the ciliary vesicle as well as to the ciliary sheath.

(4) The overall data quality is high. The investigation of EXOC6A at different time points during ciliogenesis is well schematized and explained.

Weaknesses:

(1) Since many conclusions are based on EXOC6A immunostaining, it would strengthen the study to validate antibody specificity by demonstrating the absence of staining in EXOC6A-deficient cells.

(2) While the authors generated an EXOC6A-deficient cell line, off-target effects can be clone-specific. Validating key experiments in a second independent knockout clone or rescuing the phenotype of the existing clone by re-expressing EXOC6A would ensure that the observed phenotypes are due to EXOC6A loss rather than unintended off-target effects.

(3) Some experimental details are lacking from the materials and methods section. No information on how the co-immunoprecipitation experiments have been performed can be found. The concentrations of pharmacological agents should be provided to allow proper interpretation of the results, as higher or lower doses can produce nonspecific effects. For example, the concentrations of ciliobrevin and nocodazole used to treat RPE1 cells are not specified and should be included. More precise settings for the FRAP experiments would help others reproduce the presented data. Some details for the siRNA-based knockdowns, such as incubation times, can only be found in the figure legends.

Taken together, the authors achieved their goal of elucidating the role of EXOC6A in ciliogenesis, demonstrating its involvement in vesicle trafficking and membrane remodeling in both early and late stages of ciliogenesis. Their findings are supported by experimental evidence. This work is likely to have an impact on the field by expanding our understanding of the molecular machinery underlying cilia biogenesis, particularly the coordination between the exocyst complex and cytoskeletal transport systems. The methods and data presented offer valuable tools for dissecting vesicle dynamics and cilium formation, providing a foundation for future research into ciliary dysfunction and related diseases. By connecting vesicle trafficking to structural maturation of an organelle, the study adds important context to the broader description of cellular architecture and organelle biogenesis.

eLife Assessment

This valuable study investigates the role of HIF1a signaling in epicardial activation and neonatal heart regeneration in mice. Using a combination of genetic and pharmacological approaches, the authors demonstrate that stabilization of HIF1a enhances epicardial activation and extends the regenerative capacity of the heart beyond the typical neonatal window following myocardial infarction. The main conclusion is well supported by solid data, although some minor concerns regarding experimental interpretation require further clarification to ensure accuracy.

Reviewer #1 (Public review):

Summary:

The manuscript by Gamen et al. analyzed the functional role of HIF signaling in the epicardium providing evidence that stabilization of the hypoxia signaling pathway might contribute to neonatal heart regeneration. By generating different conditionally mouse mutants and performing pharmacological interventions, the authors demonstrate that stabilizing HIF signaling enhances cardiac regeneration after MI in P7 neonatal hearts.

Strengths:

The study presents convincing genetic and pharmacological approaches on the role of hypoxia signaling enhance the regenerative potential of the epicardium

Weaknesses:

The major weakness remains the lack of convincing evidence demonstrating the role of hypoxia signaling in EMT modulation in the epicardial cells. The authors claimed that EMT assays adopted in this study are based on similar previous studies. Surprisingly, two of the references provided correspond to their own research group (PMID: 17108969, PMID: 19235142), limiting the credit for such claims, and the other two (PMID: 27023710, PMID: 12297106) assessment of cell migration but not EMT is reported. Thus, EMT remains to be convincingly demonstrated.

Reviewer #2 (Public review):

Summary:

In this study, Gamen et al. investigated the roles of hypoxia and HIF1a signaling in regulating epicardial function during cardiac development and neonatal heart regeneration. The authors identified hypoxic regions in the epicardium during development and demonstrated that genetic and pharmacological stabilization of HIF1a during neonatal heart injury prolonged epicardial activation, preserved myocardium, enhanced infarct resolution, and maintained cardiac function beyond the normal postnatal regenerative window.

Strengths:

HIF1a signaling was manipulated in an epicardium-specific manner using appropriate genetic tools.

Weaknesses:

Some conclusions still need clarification.

Comments on revisions:

(1) The authors' comment on the partial overlap of HP1 and HIF1a IF signals (HIF1a is highly unstable ... broader regions of hypoxia) is reasonable and would help readers interpret the data if included in the text describing Fig. 1.

(2) The conclusion regarding WT1+ cells in Fig. 2a and b remains unclear. Both panels display larger and smaller magenta cells, and when all are taken into account, the overall number does not appear substantially different. Additional clarification is needed on how the quantification was performed.

(3) Regarding Figure 6-figure supplement 1c, it seems difficult to conclude the endothelial identity of WT1+ cells based on EMCN staining, as the markers do not overlap. The authors note that WT1 is upregulated in endothelial cells, but this has been reported in the context of injury, which differs from the context of the present study involving Molidustat.

Reviewer #3 (Public review):

Summary:

The author's research here was to understand the role of hypoxia and hypoxia-induced transcription factors Hif-1a in the epicardium. The authors noted that hypoxia was prevalent in the embryonic heart and this persisted into neonatal stages until post natal day 7 (P7). Hypoxic regions in the heart were noted in the outer layer of the heart and expression of Hif-1a coincided with the epicardial gene WT1. It has been documented that at P7, the mouse heart cannot regenerate after myocardial infarction and the authors speculated that the change in epicardial hypoxic conditions could play a role in regeneration. The authors then used genetic and pharmacological tools to increase the activity of Hif genes in the heart and noted that there was a significant improvement in cardiac function when Hif-1a was active in the epicardium. The authors speculated that the presence of Hif-1a improved cell survival.

Strengths:

A focus on hypoxia and its effects on the epicardium in development and after myocardial infraction. This study outlines a potential to extend the regenerative time window in neonatal mammalian hearts.

Weaknesses:

While the observations of improved cardiac function is clear, the exact mechanism of how increased Hif-1a activity causes these effects is not completely revealed. The authors mention improved myocardium survival, but do not include studies to demonstrate this.

There is an indication that fibrosis is decreased in hearts where Hif activity is prolonged, but there are no studies to link hypoxia and fibrosis.

Comments on revisions:

In the manuscript revision, the authors address my comments. They outline differences between genetic disruption of Phd2 and chemical inactivation could be due to dosing and drug half-life of Molidustat. The other comments are addressed by explaining that they have analyzed enough heart sections and hearts to come to their conclusions. The authors also state they cannot generate more numbers for this study, therefore I accept their conclusions as stated.

Author response:

The following is the authors’ response to the original reviews

eLife Assessment

This valuable study investigates the role of HIF1a signalling in epicardial activation and neonatal heart regeneration in mice. Through a combination of genetic and pharmacological approaches, the authors show that stabilization of HIF1a enhances epicardial activation and extends the regenerative capacity of the heart beyond the typical neonatal window following myocardial infarction (MI). However, several aspects of the study remain incomplete and would benefit from further clarification and additional experimental support to solidify the conclusions.

We reveal herein prolonged epicardial activation following myocardial infarction (MI) beyond post-natal days 1-7 (P1-P7) by genetic or pharmacological stabilisation of HIF-signalling. This extends the so-called “regenerative window” during an adult-like response to injury, leading to enhanced survived myocardium and functional improvement of the heart, even against a backdrop of persistent, albeit reduced, fibrosis. The epicardium is known to enhance cardiomyocyte proliferation and myocardial growth during heart development via trophic growth factor (for example, IGF-1, FGF, VEGF, TGFβ and BMP) signalling (reviewed in PMID:29592950) and epicardium-derived cell-conditioned medium reduces infarct size and improves heart function (PMID: 21505261). Further experiments, outside of the scope of the current study, are required to determine whether activated neonatal epicardium elicits similar paracrine support to sustain the myocardium and heart function after injury beyond P7 into adulthood.

Public Reviews:

Reviewer #1 (Public review):

Summary:

The manuscript by Gamen et al. analyzed the functional role of HIF signaling in the epicardium, providing evidence that stabilization of the hypoxia signaling pathway might contribute to neonatal heart regeneration. By generating different conditionally mouse mutants and performing pharmacological interventions, the authors demonstrate that stabilizing HIF signaling enhances cardiac regeneration after MI in P7 neonatal hearts.

Strengths:

The study presents convincing genetic and pharmacological approaches to the role of hypoxia signaling in enhancing the regenerative potential of the epicardium.

Weaknesses:

The major weakness is the lack of convincing evidence demonstrating the role of hypoxia signaling in EMT modulation in epicardial cells. Additionally, novel experimental approaches should be performed to allow for the translation of these findings to the clinical arena.

We respectfully disagree that we have not convincingly demonstrated a role for HIF-signalling in promoting epicardial EMT. We adopt epicardial explant assays utilising a well characterised ex vivo protocol previously described for studying EMT in embryonic, neonatal and adult epicardium (PMID: 27023710, PMID: 12297106; PMID: 17108969, PMID: 19235142). These assays demonstrate in WT1^CreERT2;Phd2^fl/fl explants enhanced cobblestone to spindle-like change in cell morphology, increased cell migration, appearance of stress fibres and an up-regulation of the mesenchymal marker alpha-smooth muscle actin (αSMA); all parameters associated with EMT. In addition, our in vivo analyses of Wt1^CreERT2;Phd2^fl/fl hearts, in response to neonatal injury, reveal elevated numbers of WT1+ epicardial cells within the sub-epicardial region and underlying myocardium as is associated with active EMT and subsequent migration from the epicardium.

Reviewer #2 (Public review):

Summary:

In this study, Gamen et al. investigated the roles of hypoxia and HIF1a signaling in regulating epicardial function during cardiac development and neonatal heart regeneration. They found that WT1⁺ epicardial cells become hypoxic and begin expressing HIF1a from mid-gestation onward. During development, epicardial HIF1a signaling regulates WT1 expression and promotes coronary vasculature formation. In the postnatal heart, genetic and pharmacological upregulation of HIF1a sustained epicardial activation and improved regenerative outcomes.

Strengths:

HIF1a signaling was manipulated in an epicardium-specific manner using appropriate genetic tools.

Weaknesses:

There appears to be a discrepancy between some of the conclusions and the provided histological data. Additionally, the study does not offer mechanistic insight into the functional recovery observed.

We respectfully disagree with the comment that our histological data does not support our conclusions and expand on this in the response to specific reviewer comments. We agree that further mechanistic experiments outside of the scope of the current study are required to identify precisely how activated neonatal epicardium results in increased healthy myocardium after injury beyond post-natal day 7 (P7).

Reviewer #3 (Public review):

Summary:

The authors' research here was to understand the role of hypoxia and hypoxia-induced transcription factor Hif-1a in the epicardium. The authors noted that hypoxia was prevalent in the embryonic heart, and this persisted into neonatal stages until postnatal day 7 (P7). Hypoxic regions in the heart were noted in the outer layer of the heart, and expression of Hif-1a coincided with the epicardial gene WT1. It has been documented that at P7, the mouse heart cannot regenerate after myocardial infarction, and the authors speculated that the change in epicardial hypoxic conditions could play a role in regeneration. The authors then used genetic and pharmacological tools to increase the activity of Hif genes in the heart and noted that there was a significant improvement in cardiac function when Hif-1a was active in the epicardium. The authors speculated that the presence of Hif-1a improved cell survival.

Strengths:

A focus on hypoxia and its effects on the epicardium in development and after myocardial infarction. This study outlines the potential to extend the regenerative time window in neonatal mammalian hearts.

We thank the reviewer for this positive endorsement and recognition of the importance of mechanistic insight into how to extend the window of neonatal heart regeneration.

Weaknesses:

While the observations of improved cardiac function are clear, the exact mechanism of how increased Hif-1a activity causes these effects is not completely revealed. The authors mention improved myocardium survival, but do not include studies to demonstrate this.

We report an increase in healthy myocardium arising from prolonged activation of the epicardium during the neonatal window and following injury at post-natal day 7 (P7). We speculate this recapitulates the role of the epicardium during heart development which is known to be a source of trophic growth factors that can enhance myocardial growth. Further experiments are required, out-of-scope of this study, to define a mechanistic link between HIF-signalling, epicardial activation and myocardial survival in the setting of prolonged neonatal heart regeneration.

There is an indication that fibrosis is decreased in hearts where Hif activity is prolonged, but there are no studies to link hypoxia and fibrosis.

We believe the decreased fibrosis is a natural consequence of the increase in survived myocardium arising from the activated epicardium. There is strong precedent here following injury at post-natal day 1 (P1) in which fibrosis is evident early-on but is resolved over time with growth of the myocardium in the regenerating heart (PMID: 23248315).

Recommendations for the authors:

Reviewing Editor Comments:

(1) Address issues related to image quality, colocalization, sample labeling, appropriate controls, and quantification - particularly in Figures 1, 2, 6, and Supplementary Figure 9. Increase sample size as noted by reviewers.

The issues of co-localisation and sample labelling have been addressed under response to reviewers. We are unable to increase sample numbers but have clarified the number of regions per section and numbers of sections per heart analysed where appropriate.

(2) Clarify the effects of epicardial HIF1a activation on neovascularization.

We have removed reference in the abstract to an effect on neovascularisation.

(3) Extend assessments of epicardial hypoxia and HIF1a expression to earlier embryonic stages, when epicardial EMT is more active.

Our earliest timepoint of E12.5 marks the onset of epicardial EMT and E13.5 is the stage with the most significant mobilisation of epicardium-derived cells (EPDCs) into the sub-epicardial region and underlying myocardium (PMID: 32359445). In the same study, E11.5 lineage tracing of epicardial cells is restricted to outer layer of the heart; thus, our timepoints are representative in capturing both the onset and progression of in vivo EMT.

(4) Strengthen EMT assays and mechanistic modeling. Provide evidence from physiologically relevant models, as current 2D culture assays do not adequately support conclusions about EMT. Include additional EMT markers and quantification where appropriate.

We respectfully disagree that epicardial explants are not a valid assay for assessing EMT. As noted under responses to reviewers, such primary explants have been widely described elsewhere (PMID: 27023710, PMID: 12297106; PMID: 17108969, PMID: 19235142) and enable documentation of multiple parameters that are associated with active EMT, including an assessment of the extent of cell migration, cobblestone (epithelial) to spindle-like (mesenchymal) cell morphologies, stress fibre formation and expression of alpha-smooth muscle actin as a mesenchymal marker. We support our findings in explants by revealing reduced WT1+ epicardium-derived cells (EPDCs) in the sub-epicardial region and underlying myocardium of WT1^CreERT2/+;Hif1a^fl/fl embryonic hearts (data in Figure 2) indicative of impaired epicardial EMT and migration of EPDCs and in vivo following neonatal MI with pharmacological inhibition of PHD2, where we observe the reciprocal phenotype of increased numbers of epicardium-derived cells emerging from the outer epicardial layer (data in Figure 6).

(5) Strengthen mechanistic insights into the role of epicardial cells in the functional recovery observed in MI hearts.

We agree that further experiments are required, out-of-scope of this study, to define a mechanistic link between HIF-signalling, epicardial activation and myocardial survival in the setting of prolonged neonatal heart regeneration.

Reviewer #1 (Recommendations for the authors):

The manuscript by Gamen et al. analyzed the functional role of HIF signaling in the epicardium, providing evidence that stabilization of the hypoxia signaling pathway might contribute to neonatal heart regeneration. By generating different conditionally mouse mutants and performing pharmacological interventions, the authors demonstrate that stabilizing HIF signaling enhances cardiac regeneration after MI in P7 neonatal hearts. The study is potentially interesting, but it presents several major caveats.

(1) One of the critical points reported in the early stages of this study is the early co-localization of Wt1, the hypoxic report (HP1), and HIF signaling pathways master regulators (i.e., HIF1a and HIF1b) during embryonic development. Figure 1 is meant to report such findings. However, unfortunately, I hardly see any co-localization at all in the Wt1+ epicardial cells for HP1, with some colocalization is seen for HIF1 and 2 alpha, although none of these data are quantified. Thus, it is hard to believe such co-localization.

We respectfully disagree with this comment. We highlight cells in Figure 1 that are co-stained for WT1+ and HP1. In addition, we identify HIF1-α and HIF2- α positive cells which either reside within the epicardium, as the outer cell layer, or within the underlying sub-epicardial region, respectfully.

(2) The authors claimed that they have analyzed the expression of the hypoxic report, as well as Wt1 and the HIF signaling pathways master regulators (i.e., HIF1a and HIF1b) in the AV groove, as compared to the apex, in embryonic heart ranging from E12.5 to E18.5 (Figure 1). Unfortunately, all images provided that are tagged as AV groove are rather misleading. They do not represent the AV groove but part of the right ventricular free wall. If the authors want to refer to the AV groove, AV cushions should be visible underneath.

We have removed specific reference to the AV groove and refer to the highlighted regions as the “Base” of the heart.

(3) The authors analyzed the hypoxic condition of the developing heart from E12.5 to E18.5. However, it remains unclear why the authors only explored the hypoxic conditions from E12.5 onwards, since epicardial EMT mainly occurs earlier than this time point, i.e., E10.5 onwards. Therefore, it would be needed to explore it already at this earlier time point.

We respectfully disagree with the reviewer and refer to the comment above regarding the fact that E12.5 marks the onset of epicardial EMT and E13.5 is the stage with the most significant mobilisation of epicardium-derived cells (EPDCs) into the sub-epicardial region and underlying myocardium (PMID: 32359445).

(4) The authors reported a conditional mouse model of HIF1alpha deletion by using the Wt1CreERT2 driver. Curiously, Wt1 is dependent on hypoxia signaling (i.e., HIF1a). Therefore, it is unclear whether there is a negative feedback loop between the deletion of Hif1alpha and the activation of the Cre driver might have functional consequences. Convincing evidence should be provided that such crosstalk does not interfere with Hif1alpha inactivation, and therefore, appropriate controls should be run in parallel.

We discount a negative feedback loop in this instance based on the fact we have utilised heterozygous mice for the WT1^CreERT2/+ line and observe a consistent and reproducible phenotype for the developing hearts on a Wt1^CreERT2/+;Hif1a^fl/fl background and following injury in Wt1^CreERT2/+;Phd2^fl/fl mice. Collectively this indicates that the WT1-CreERT2 driver is active in the context of diminishing HIF-1α and Phd2, respectively. In addition, have carried out parallel experiments using epicardial explants derived from R26R-CreERT2;Phd2^fl/fl (Figure 3) to circumvent any potential confounding issues; the results of which are consistent with increased epicardial EMT in support of our overall hypothesis.

(5) On Figure 2a-f the authors reported that epicardial cells are diminished in Wt1CreERT2Hif1alpha mice as compared to controls. I am very sorry, but I do not see any difference. Furthermore, it is unclear to me how the authors quantified such differences, i.e., what marker signal did they use and how it was performed (Figure 2c and d)?

We respectfully disagree with the reviewer and draw attention to the single channel panels of WT1+ staining in Figure 2, which show clear differences between numbers of epicardial cells in the mutant mice compared to controls (comparing magenta cells in panels a) versus b). Quantification was carried out for numbers of WT1+ cells residing within the PDPN-positive epicardium (and underlying PDPN-negative myocardium) across multiple images from multiple sections and multiple hearts.

(6) On Figure 2g, the authors reported differences in total vessel length. Are they referring to impaired microvasculature development? Or is this analysis also including major coronary vessels? What about the major coronary vessels and trees, is there any affection?

This analysis refers to the microvasculature and not the major coronary arteries or coronary trees.

(7) The authors reported that there might be some differences in EMT markers, but unfortunately, all of them are analyzed on 2D cultures, where no substrate for EMT is present, i.e., an underlying ECM bed. Thus, the authors cannot claim that EMT is altered. Additional experiments using either collagen substrate and/or Matrigel are required to fully demonstrate that EMT is impaired. Furthermore, quantitative analyses of such differences should be provided.

The 2D cultures are epicardial explants from mutant versus wild type hearts and represent a widely adopted previously published ex-vivo assay for investigating epicardial EMT across embryonic to adult stages (PMID: 27023710, PMID: 12297106; PMID: 17108969, PMID: 19235142); including an assessment of the extent of migration and cobblestone (epithelial) to spindle-like (mesenchymal) cell morphologies, stress fibre formation and expression of alpha-smooth muscle actin as a mesenchymal marker. We do not understand the comment regarding an “underlying ECM bed” as the cells exhibit EMT routinely on tissue culture plastic and will deposit their own ECM during the culture time course and in response to EMT/cell migration. In terms of quantification this was carried out for scratch assay experiments, as a proxy for EMT and emergent mesenchymal cell migration, as presented in Figure 3i, j with significant enhanced scratch closure and cell migration following Molidustat treatment.

(8) The description of data provided on Supplementary Figure 5 is spurious and should be removed. A note in the discussion might be sufficient.

We respectfully disagree. The ChIP-seq data, in what is now Figure 2- figure supplement 3, highlights a HIF-1 α binding site within the Wt1 locus suggesting putative upstream regulation of WT1 by HIF-1α. Thus this provides a potential explanation as to how HIF-1α may activate the epicardium through up-regulation of Wt1/WT1.

(9) On Figure 3, the authors further illustrate the change of EMT markers using ex vivo cardiac explants. They reported increased expression of Snai2 that, although statistically significant, is most likely of no biological relevance (increase of only 20% at transcript level). What about Snai1, Prrx1, and other EMT promoters? Are they also induced? As previously stated, these 2D cultures do not provide supporting evidence that EMT is occurring, thus 3D gel assays should be performed in which Z-axis analyses will provide evidence on the different migratory behaviour of those cells.

We respectfully suggest that a 20% change in snai2 expression is biologically meaningful with respect to EMT. This in-turn is supported by associated cell migration, reduced ZO-1 expression, increased stress fibres and increased alpha-SMA as a mesenchymal marker; all properties associated with active EMT. Other suggested markers have not been validated as formally required for EMT, for example Snai1 (PMID: 23097346). The migratory capacity of targeted versus epicardial cells was assessed by combined explant and scratch assay experiments.

(10) The description of single-cell analyses is very incomplete. Which mice were used for these analyses, wildtype control, or hypoxic mice? Please provide a clearer description of the samples used. Additionally, the entire rationale of these analyses is dubious. Doing single-cell analyses to analyze a couple or three markers in a very small cell population is rather ridiculous. qPCR might be far more appropriate and convincing, or a bulk RNAseq analysis of isolated epicardial cells.

The single-cell analyses represent an unbiased assessment of different pathways in epicardial cells (identified bioinformatically) between intact P1 and P7 stages in wild type (control) hearts, with a focus on hypoxia-related gene expression and HIF-dependent pathways. It was not designed to analyse a small number of genes, rather global differences in the hypoxic states between P1 and P7 hearts. Selected genes (Vegfa, Pdk3, Egln 1 (Phd2)) were analysed to highlight the key differences in hypoxic signalling across the regenerative window. The fact the hearts were uninjured/intact is clarified in the text and legends for Figure 4 and now Figure 4-figure supplement 1.

(11) The analyses provided in Figure 5 are very interesting and their findings are very relevant. However, I would think that the complementary experimental approach should also be done, i.e, MI followed by activation with tamoxifen, since that situation would be more realistic in the clinical setting.

Tamoxifen causes respiratory failure in neonates with MI, so the two cannot be combined at the same time or soon after surgery. Moreover, tamoxifen takes significant time to take effect on targeted gene down-regulation which may negate sufficient activation of the epicardium following injury.

The experiments in Figure 5 were designed to demonstrate that prolonged heart regeneration could be elicited in a cell-specific (epicardial-specific) manner via a genetic approach. The pharmacological experiments in Figure 6 are complementary in this regard by demonstrating equivalent effects with drug (Molidustat) delivery to reduce PHD2 and stabilise HIF post-MI.

(12) In Figure 6, expression of Wt1 is highly prominent in P7 controls, mainly restricted to the epicardial lining while in the experimental setting, such Wt1 expression is broadly distributed on the subepicardial space, nicely demonstrating epicardial activation. However, it is very surprising to see such Wt1 expression in controls, something that is not expected, as compared to the data reported in Figure 4g. Could the authors please reconcile these findings?

Figure 6 represents the injury setting and Figure 4g the intact setting (as clarified above, in the text and revised figure legends). Hence in the latter WT1 expression is significantly reduced in the P7 heart, as anticipated. With injury at P7 we anticipate activation of WT1 in control hearts, albeit restricted to the epicardial layer (as occurs in adult hearts, PMID: 21505261). In contrast, following Molidustat-treatment of P7 hearts post-MI we observe extensive epicardial expansion into the sub-epicardial region and EPDC migration into the underlying myocardium (Figure 6b).

Reviewer #2 (Recommendations for the authors):

The role of hypoxia and HIF1a signaling in epicardial activation is an important topic, and the genetic approaches employed in this study are appropriate. However, several aspects of the study remain unclear and would benefit from further clarification or explanation by the authors:

(1) The authors detected hypoxic regions using an anti-pimonidazole fluorescence-conjugated monoclonal antibody (HP1). The data would become more compelling if negative and positive controls were provided.

We believe the HP1 staining is compelling in the images shown and is consistent with hypoxic regions of the developing heart. We reveal HP1 staining at cellular resolution with neighbouring cells positive and negative for the HP1 signal in the apex of the heart and within the epicardium and sub-epicardial regions at E12.5 (Figure 1a) and diminished/altered hypoxic/HP1 regional signal through subsequent developmental stages at E14.5-18.5 (Figure 1a-d).

(2) Many HIF1a-positive cells in the AV groove region do not appear to overlap with HP1 staining (Figure 1a). Providing a low-magnification image of HIF1α expression would be helpful to better assess the extent of overlap with HP1 staining

HIF-1 is highly unstable and hence detection of HIF-1+ cells will likely only sample of cells compared to HP1 which is a surrogate for broader regions of hypoxia.

(3) Although the authors conclude that epicardial HIF1a deletion results in a significant reduction of WT1⁺ cells in both the epicardium and myocardium (Figure 2a-d), the provided images are not sufficiently clear to fully support this interpretation. Providing additional evidence to support this conclusion would be helpful.

We respectfully disagree with the reviewer and draw attention to the single channel panels of WT1+ staining which show clear differences between numbers of epicardial cells in the mutant mice compared to controls (Figure 2a versus 2b; magenta WT1+ staining).

(4) Similar to the point raised above, the authors' conclusion regarding the increased expression of WT1 following Molidustat treatment does not appear to be fully supported by the provided images (Figure 6b-f). Immunofluorescence staining for WT1 does not clearly demonstrate epicardial expression in the remote zone of either the control or Molidustat-treated hearts. In addition, while an increase of WT1⁺ cells is observed in the infarct zone of the Molidustat-treated heart, it is somewhat unexpected that such expansion is not evident in the corresponding region of the control heart, given that epicardial cells typically expand near the infarct area. Clarification on these points would be helpful.

Figure 6b reveals WT1 expression in controls (upper panel set) that is reactivated proximal to the infarct region, given WT1 is not expressed in adult epicardium but restricted to the epicardial layer (as occurs in injured adult mouse hearts PMID: 21505261). This contrasts with what is observed in the Molidustat-treated P7 hearts post-MI, where we observe epicardial expansion and migration of WT1+ cells into the underlying myocardium (Figure 6b, lower panel set, infarct zone).

(5) The authors conclude that WT1⁺ cells in the myocardial tissue exhibit endothelial identity based on the colocalization of WT1 and EMCN signals (Supplementary Figure 9c). However, this interpretation is difficult to assess, as WT1 is a nuclear marker and EMCN is a membrane protein, which makes precise colocalization challenging to confirm with confidence. Additional supporting evidence may be necessary to substantiate this conclusion.

WT1 is known to be up regulated in endothelial cells in response to injury as shown previously in several studies (for example, PMID: 25681586). Here we show clear co-localisation of nuclear WT1 and cytoplasmic Endomucin (EMCN) in what is now Figure 6- figure supplement 1c and would encourage the reviewer and readers to magnify the image by zooming-in on the relevant co-stained panel.

(6) The authors conclude that activation of epicardial HIF1a signaling has no effect on neovascularization in postnatal MI hearts (Figure 5c). However, the abstract states: "Finally, a combination of genetic and pharmacological stabilisation of HIF ... increased vascularisation, augmented infarct resolution and preserved function beyond the 7-day regenerative window" (Lines 38-41). Clarification regarding this apparent discrepancy would be appreciated.

The abstract has been altered to remove the statement of increased vascularisation.

(7) The study appears somewhat incomplete, as it lacks mechanistic insight into the functional recovery observed following epicardial Phd2 deletion and Molidustat treatment in postnatal MI hearts. Although the authors suggest a potential paracrine role of the epicardium in protecting cardiomyocytes from apoptosis, this hypothesis has not been experimentally addressed. Incorporating such analysis would help to reinforce the study's conclusions.

Further experiments are required, which are out-of-scope of this study, to define a mechanistic link between the genetic or pharmacological stabilisation of HIF-signalling, epicardial activation and myocardial survival in the setting of prolonged neonatal heart regeneration.

Other points:

(1) Providing single-channel images for Figures 1a-d and 6g would be helpful for clarity and interpretation.

We believe the combined channel views of co-staining for two markers on a background of DAPI staining to pin-point cell nuclei, are informative and support our conclusions.

(2) Have the authors considered using AngioTool to quantify the number of vessels in Figure 5b-c?

AngioToolTM was used to quantify the vessels, as we have used previously (PMID: 33462113) and this is now added to the methods and legend of Figure 2.

Reviewer #3 (Recommendations for the authors):

There are several areas where the manuscript can be improved, such that its conclusions can be solidified.

(1) The authors highlight a point where blocking Phd2 can enhance survival of cardiac tissue, but did not report on survival markers. They surmised that apoptosis could be decreased in Phd2 mutant or Molidustat treatment but did not show this. The authors should determine if apoptosis is decreased in the myocardium and epicardium.

We show evidence of increased levels of healthy myocardium in the genetic and pharmacological models of stabilised HIF-signalling. We exclude increased cardiac hypertrophy or increased cardiomyocyte proliferation as causative, so suggest as a reasonable alternative enhanced survival, albeit this need not necessarily be via an apoptotic pathway given the incidence of necrotic cell death during MI. We are unable to generate new surgeries and mutant/treated heart samples to analyse for apoptotic markers at this stage.

(2) There appears to be no difference in cardiomyocyte proliferation in Molidustat-treated animals, but the experiment was only performed on 2 to 3 animals. This is too small a sample size to conclude from these results. The authors should increase the sample size to make this assertion.

We respectfully disagree that we are unable to conclude no effect on cardiomyocyte proliferation. We analysed multiple heart regions per section, for EdU+/cTnT+ colocalised signals across several sections per heart, set against a consistency of effect on other parameters in hearts treated with Molidustat. We are unable to generate more P7 heart surgeries +/- Molidustat and +/- EdU at this stage.

(3) It is curious as to how, after myocardial infarction, the fibrotic scar tissue is decreased in the Phd2 deletion but not as profound in Molidustat-treated mice at d21. Can the authors speculate why the difference exists and how this decrease arises? For example, are there decreased pro-inflammatory signals in Phd2 deleted mice? Is there decreased collagen deposition and ECM gene expression? Do macrophage recruitment into the infarct zone differ between mutant/treated vs WT?

The representative images in Figure 6k reveal a trend towards reduced fibrosis with Molidistat treatment (Figure 6l), but across all hearts analysed this was not as significant as observed in the epicardial-specific deletion injured hearts (Figure 5g, h). This may be due to the relatively short half-life of Molidustat (approximately 4-10 hours, PMID: 32248614), the dosing regimen for the drug and/or the fact that it was not specifically delivered/targeted to the epicardium.

(4) The magnified images in Figure 1 do not match the boxes in the whole heart images. It is unclear what the white boxes signify.

The white boxes have been removed from Figure 1. The magnified image panels are from serial heart sections and this is now clarified in the Figure 1 legend.

eLife Assessment

This fundamental work substantially advances our understanding of how the glycocalyx of cells provide a non-specific barrier for the interaction of viruses with cell-surface receptors. Using both in vitro experiments and in vivo manipulations they provide compelling evidence for the properties of the glycocalyx to serve as an energy barrier as a main attribute of its mode of action. The work will be of broad interest to virologists and the cell biology community that studies host-pathogen interactions.

Joint Public Review:

This manuscript tests the notion that bulky membrane glycoproteins suppress viral infection through non-specific interactions. Using a suite of biochemical, biophysical, and computational methods in multiple contexts (ex vivo, in vitro, and in silico), the authors collect compelling evidence supporting the notion that (1) a wide range of surface glycoproteins erect an energy barrier for the virus to form stable adhesive interface needed for fusion and uptake and (2) the total amount of glycan, independent of their molecular identity, additively enhanced the suppression.

As a functional assay the authors focus on viral infection starting from the assumption that a physical boundary modulated by overexpressing a protein-of-interest could prevent viral entry and subsequent infection. Here they find that glycan content (measured using the PNA lectin) of the overexpressed protein and total molecular weight, that includes amino acid weight and the glycan weight, is negatively correlated with viral infection. They continue to demonstrate that it is in effect the total glycan content, using a variety of lectin labelling, that is responsible for reduced infection in cells. Because the authors do not find a loss in virus binding this allows them to hypothesize that the glycan content presents a barrier for the stable membrane-membrane contact between virus and cell. They subsequently set out to determine the effective radius of the proteins at the membrane and demonstrate that on a supported lipid bilayer the glycosylated proteins do not transition from the mushroom to the brush regime at the densities used. Finally, using Super Resolution microscopy they find that above an effective radius of 5 nm proteins are excluded from the virus-cell interface.

The experimental design does not present major concerns and the results provide insight on a biophysical mechanism according to which, repulsion forces between branched glycan chains of highly glycosylated proteins exert a kinetic energy barrier that limits the formation of a membrane/viral interface required for infection.

In their revised manuscript and rebuttal, the authors address several general and specific concerns that were raised about their first submission. The revised manuscript now makes the strength of the evidence supporting their claims, compelling.

Author response:

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

Public Review

GENERAL QUESTIONS:

(1) For many enveloped viruses, the attachment factors - paradoxically - are also surface glycoproteins, often complexed with a distinct fusion protein. The authors note here that the glycoportiens do not inhibit the initial binding, but only limit the stability of the adhesive interface needed for subsequent membrane fusion and viral uptake. How these antagonistic tendencies might play out should be discussed.

When the surface density of receptor molecules for a virus with glycans increases, the density of free glycans not bound to the virus increases along with the amount of virus adsorbed. However, if the total amount of glycans is considered to be a function of the receptor density, the reaction may become more complicated. This complication may also be affected by the prolonged infection. If the receptor density on the cell surface is high, the infection inhibitory effect of glycans may not be obtained in a system in which a high concentration of virus is supplied from the outside world for a long time. This is because once viruses have entered the cell, they accumulate inside the cell, and viral infection is affected by the total accumulated amount, which is the integration of the number of viruses that have entered over time. This distinction indicates that the virus entry reaction and the total amount of infection in the cell must be considered separately. This is an important point, but it was not clearly mentioned in the original manuscript.

Our experiments were conducted under conditions that clearly allowed us to detect the virusinhibiting function of glycans without being affected by the above points. In order to clarify these points, we will revise this article as follows, referring to an experiment that is somewhat related to this discussion (the Adenovirus infection experiment into HEK293T cells shown in Figure S1F)..

(Page-3, Introduction)

While there are known examples of glycans that function as viral receptors (Thompson et al., 2019), these results demonstrate that a variety of glycoproteins negatively regulate viral infection in a wide range of systems. All of these results suggest that bulky membrane glycoproteins nonspecifically inhibit viral infection.

(Page 20, Discussion)

When the virus receptor is a glycoprotein or glycan itself, the inhibition of virus infection by glycans becomes more complex because the total amount of glycans is also a function of the receptor density. It is also important to note that the total amount of infection into a cell is the time integral of virus entry. Even if the probability of virus entry is significantly reduced by glycans, the cumulative number of virus entries may increase if high concentrations of virus continue to be supplied from outside the cell for a long period of time. In the case of Adenovirus, which continues to amplify in HEK293T cells after infection, we showed that MUC1 on the cell surface has an inhibitory effect on long-term cumulative infection (Supplementary Figure 1F). However, such an accumulation effect may be caseby-case depending on the virus cell system, and may be more pronounced when the cell surface density of virus receptor molecules is high. As a result, if the virus receptor molecule is a glycan or glycoprotein and infection continues for a long period of time, the infection inhibition effect may not be observed despite an apparent increase in the total amount of glycans in the cell. In any case, our results clarified the factor of virus entry inhibition dependent on the total amount of glycans because appropriate conditions were set.

(2) Unlike polymers tethered to solid surface undergoing mushroom-to-brush transition in densitydependent manner, the glycoproteins at the cell surface are of course mobile (presumably in a density-dependent manner). They can thus redistribute in spatial patterns, which serve to minimize the free energy. I suggest the authors explicitly address how these considerations influence the in vitro reconstitution assays seeking to assess the glycosylation-dependent protein packing.

We performed additional experiments using lipid bilayers that had lost fluidity, and found that there is no significant difference in protein binding between fluid and nonfluid bilayers. The redistribution of molecules due to molecular fluidity may play some roles but not in our experimental systems. It suggests that glycoproteins can generate intermolecular repulsion even in fluid conditions such as cell membranes, just as they do on the solid phase. This experiment was also very useful because it allowed us to compare our results in the fluid bilayer with solid-state measurements of saturation molecular density and the brush transition. This comparison gave us confidence that in the reconstituted membrane system, even at saturation density, the membrane proteins are not as stretched as they are in the condensed brush state. We have therefore added a new paragraph and a new figure (Supplementary Fig. 5B) to discuss this issue, as follows:

The molecular structural state of these proteins needs to be further discussed to estimate the contribution of f_el, which represents resistance to molecular elongation. Our results suggest that these densely packed nonglycosylated molecules are no longer in a free mushroom state. However, their saturation density was several times lower than previously reported brush transition densities, such as 65000 µm^-2 for 17 kDa polyacrylamide (R_F ~ 15 nm) on a solid surface (Wu et al., 2002). To compare our data on fluid bilayers with previously reported data on solid surfaces, we performed additional experiments with lipid bilayers that lost fluidity. No significant changes in protein binding between fluid and nonfluid bilayers were observed for both b-MUC1 and g-MUC1 molecules (Supplementary Figure 5B). This result suggests that membrane fluidity does not affect the average intermolecular distance or other relevant parameters that control molecular binding in the reconstituted system. Based on these, we speculate that the saturated protein density observed in our experiments is lower than or at most comparable to the actual brush transition density. Thus, although these crowded proteins may be restricted from free random motion, they are not significantly extended as in the condensed brush state, in which the contribution of resistance to molecular extension f_el is expected to be small relative to the overall free energy of the system.

(3) The discussion of the role of excluded volume in steric repulsion between glycoprotein needs clarification. As presented, it's unclear what the role of "excluded volume" effects is in driving steric repulsion? Do the authors imply depletion forces? Or the volume unavailable due to stochastic configurations of gaussian chains? How does the formalism apply to branched membrane glycoproteins is not immediately obvious.

Regarding the excluded volume due to steric repulsion between glycoproteins, we considered the volume that cannot be used by glycans as Gaussian chains branching from the main chain. We would like to expand on this point by adding several papers that make similar arguments. I'm glad you brought this up because we hadn't considered depletion forces - the excluded volume between glycoproteins should generate a depletion force, but in this case we believe this force will not have a significant effect on viruses that are larger than the glycoproteins. We also attempted to clarify the discussion in this section by focusing on intermolecular repulsion, and restructured paragraphs, which are also related to General Question 2 and Specific Question 2. The relevant part has been revised as follows. (page 15~page16)：

To compare the packing of proteins with different molecular weights and R_F, These were smaller than the coverage of molecules at hexagonal close packing that is ~90.7%. In contrast, the coverage of b-CD43 and b-MUC1 at saturated binding was estimated to be greater than 100% under this normalization standard, indicating that the mean projected sizes of these molecules in surface direction were smaller than those expected from their R_F Thus, it is clear that glycosylation reduces the saturation density of membrane proteins, regardless of molecular size.

Highly glycosylated proteins resisted densification, indicating that some intermolecular repulsion is occurring. In the framework of polymer brush theory, the intermolecular repulsion of densely packed highly glycosylated proteins is due to an increase in either f_el, f_int (d<R_F), or both (Hansen et al., 2003; Wu et al., 2002). The term of intermolecular interaction, f_int, is regulated by intermolecular steric repulsion, which occurs when neighboring molecules cannot approach the excluded volume created by the stochastic configuration of the polymer chain (Attili et al., 2012; Faivre et al., 2018; Kreussling and Ullman, 1954; Kuo et al., 2018; Paturej et al., 2016). The magnitude of this steric repulsion depends largely on R_F in dilute solutions, but the molecular structure may also affect it when molecules are densified on a surface. In other words, the glycans protruding between molecules can cause steric inhibition between neighboring proteins (Figure 5D). Such intermolecular repulsion due to branched side chains occurs only when the molecules are in close proximity and sterically interact on a twodimensional surface, but not in dilute solution, and does not occur in unbranched polymers such as underglycosylated proteins (Figure 5D). Based on the above, we propose the following model for membrane proteins: Only when the membrane proteins are glycosylated does strong steric repulsion occur between neighboring molecules during the densification process, suppressing densification.

The molecular structural state of these proteins needs to be further discussed to estimate the contribution of f_el, which represents resistance to molecular elongation. Our results suggest that these densely packed nonglycosylated molecules are no longer in a free mushroom state. However, their saturation density was several times lower than previously reported brush transition densities, such as 65000 µm^-2 for 17 kDa polyacrylamide (R_F ~ 15 nm) on a solid surface (Wu et al., 2002). To compare our data on fluid bilayers with previously reported data on solid surfaces, we performed additional experiments with lipid bilayers that lost fluidity. No significant changes in protein binding between fluid and nonfluid bilayers were observed for both b-MUC1 and g-MUC1 molecules (Supplementary Figure 5B). This result suggests that membrane fluidity does not affect the average intermolecular distance or other relevant parameters that control molecular binding in the reconstituted system. Based on these, we speculate that the saturated protein density observed in our experiments is lower than or at most comparable to the actual brush transition density. Thus, although these crowded proteins may be restricted from free random motion, they are not significantly extended as in the condensed brush state, in which the contribution of resistance to molecular extension f_el, is expected to be small relative to the overall free energy of the system.

Note that this does not mean that glycoproteins cannot form condensed brush structures: in fact, highly glycosylated molecules (e.g., MUC1) can form brush structures in cells when such proteins are expressed at very high densities. (Shurer et al., 2019). In these cells, ………. Such membrane deformation results in the increase of total surface area to reduce the density of glycoproteins, indicating that there is strong intermolecular repulsion between glycoproteins. In any case, the free energy of the system is determined by the balance between protein binding and insertion into the membrane, protein deformation, and repulsive forces between proteins, which determine the density of proteins depending on the configuration of the system. Thus, although strong intermolecular repulsions were prominently observed in our simplified system, this may not be the case in other systems. ……

(4) The authors showed that glycoprotein expression inversely correlated with viral infection and link viral entry inhibition to steric hindrance caused by the glycoprotein. Alternative explanations would be that the glycoprotein expression (a) reroutes endocytosed viral particles or (b) lowers cellular endocytic rates and via either mechanism reduce viral infection. The authors should provide evidence that these alternatives are not occurring in their system. They could for example experimentally test whether non-specific endocytosis is still operational at similar levels, measured with fluid-phase markers such as 10kDa dextrans.

The results of the experiment suggested by the reviewer are shown in the new Supplementary Figure 3B. (This results in generation of a new Supplementary Figure 3, and previous Supplementary Figures 4-5 are now renumbered as Supplementary Figures 5-6). Endocytosis of 10KDa dextran was attenuated by the expression of several large-sized molecules, but was not affected by the expression of many other glycoproteins that have the ability to inhibit infection. These results were clearly different from the results in which virus infection was inhibited more by the amount of glycan than by molecular weight. Therefore, it was found that many glycoproteins inhibit virus infection through processes other than endocytosis. Based on the above, we have added the following to the manuscript: （p9 New paragraph:）

We also investigated the effect of membrane glycoproteins on membrane trafficking, another process involved in viral infection. Expression of MUC1 with higher number of tandem repeats reduced the dextran transport in the fluid phase, while expression of multiple membrane glycoproteins that have infection inhibitory effects, including truncated MUC1 molecules, showed no effect on fluid phase endocytosis, indicating a molecular weight-dependent effect (Supplementary Figure 3B). The molecular weight-dependent inhibition of endocytosis may be due to factors such as steric inhibition of the approach of dextran molecules or a reduction in the transportable volume within the endosome. In any case, it is clear that many low molecular weight glycoproteins inhibit infection by disturbing processes other than endocytosis. Based on the above, we focus on the effect of glycoproteins on the formation of the interface between the virus and the cell membrane.

(5) The authors approach their system with the goal of generalizing the cell membrane (the cumulative effect of all cell membrane molecules on viral entry), but what about the inverse? How does the nature of the molecule seeking entry affect the interface? For example, a lipid nanoparticle vs a virus with a short virus-cell distance vs a virus with a large virus-cell distance?

Thank you for your interesting comment. If the molecular size of the ligand is large, it should affect virus adsorption and molecular exclusion from the interface. In lipid nanoparticle applications, controlling this parameter may contribute to efficiency. In addition, a related discussion is the influence of virus shell molecules that are not bound to the receptor. I will revise the text based on the above.

Discussion (as a new paragraph after the paragraph added in Q1):

In this study, we attempted to generalize the surface structure on the cell side, but the surface structure on the virus side may also have an effect. The efficiency of virus adsorption and the efficiency of cell membrane protein exclusion from the interface will change depending on the molecular length of the receptor-ligand, although receptor priming also has an effect. In addition, free ligands of the viral envelope or other coexisting glycoproteins may also have an effect as they are also required for exclusion from the virus-cell interface. In fact, there are reports that expression of CD43 and PSGL-1 on the virus surface reduces virus infection efficiency (Murakami et al., 2020). Such interface structure may be one of the factors that determine the infection efficiency that differs depending on the virus strain. More generally, modification of the surface structure may be effective for designing materials such as lipid nanoparticles that construct the interface with cell.

SPECIFIC QUESTIONS:

(1) The proposed mechanism indicates that glycosylation status does not produce an effect in the "trapping" of virus, but in later stages of the formation of the virus/membrane interface due to the high energetic costs of displacing highly glycosylated molecules at the vicinity of the virus/membrane interface. It is suggested to present a correlation between the levels of glycans in the Calu-3 cell monolayers and the number of viral particles bound to cell surface at different pulse times. Results may be quantified following the same method as shown in Figure 2 for the correlation between glycosylation levels and viral infection (in this case the resulting output could be number of viral particles bound as a function of glycan content).

The results of this experiment are now shown as Supplementary Figure 2F and 2G. We compared the amount of virus bound after incubation for 10 minutes or for 3 hours as in the infection experiment, but no negative correlation was found between the total amount of glycans on the surface of the Calu3 monolayer and the amount of virus bound. Interestingly, there was a sight positive correlation was detected, which may be due to concentrated virus receptor expressions in glycan-enriched cells. This result shows that glycoproteins do not strongly inhibit virus binding. We will amend the text as follows (see also Q6).

(Page 10)

Glycans could be one of the biochemical substances ……We found that a large number of SARS-CoV2-PP can still bind to cells even when cells expressed sufficient amounts of the glycoprotein that could account for the majority of glycans within these cells and inhibit viral infection (Figure 3A). Similarly, on the two-dimensional culture surface of Calu-3 cells, no negative correlation was observed between the number of viruses bound and the total amount of glycans on the cell surface (Supplementary Figure 2F-G). The slight positive correlation between bound virus and glycans may be due to higher expression levels of viral receptors in glycan-rich cells. ….

(2) The use of the purified glycosylated and non-glycosylated ectodomains of MUC1 and CD-43 to establish a relationship between glycosylation and protein density into lipid bilayers on silica beads is an elegant approach. An assessment of the impact of glycosylation in the structural conformation of both proteins, for instance determining the Flory radius of the glycosylated and non-glycosylated ectodomains by the FRET-FLIM approach used in Figure 4 would serve to further support the hypothesis of the article.

Unfortunately, the proposed experiment did not provide a strong enough FRET signal for analysis. This was due in part to the difficulty in constructing a bead-coated bilayer incorporating PlasMem Bright Red, which established a good FRET pair in cell experiments. We also tried other fluorescent molecules, but were unable to obtain a strong and stable FRET signal. Another reason may be that the curvature of the beads is larger than that of the cells, making it difficult to obtain a sufficient cumulative FRET effect from multiple membrane dyes. We plan to improve the experimental system in the future.

On the other hand, we also found that in this system, the signal changes were very subtle, making it difficult to detect molecular conformational changes using FRET. After reconsidering general questions (2) and (3), we speculated that the molecular density in the experiment, even at saturation binding, was below or at most equivalent to the brush transition point. In other words, proteins on the bead-coated bilayer may not be significantly extended in the vertical direction. Therefore, the conformational changes of these proteins may not be large enough to be detected by the FRET assay. We updated Figure 3C and Figure 5D (model description) to better reflect the above discussion and introduced the following discussion in the manuscript.

(page11)

We introduced the framework of conventional polymer brush theory to study the structure of viruscell interfaces containing proteins……. Numerous experimental measurements of the formation of polymer brushes have also been reported (Overney et al., 1996; Wu et al., 2002; Zhao and Brittain, 2000). In these measurements, the transition to a brush typically occurs at a density higher than that required to pack a surface with hemispherical polymers of diameter R_F. This is the point at which the energy loss due to repulsive forces between adjacent molecules (f_int) exceeds the energy required to stretch the polymer perpendicularly into a brush (f_el).

(page16)

The molecular structural state of these proteins needs to be further discussed to estimate the contribution of f_el, which represents resistance to molecular elongation. Our results suggest that these densely packed nonglycosylated molecules are no longer in a free mushroom state. However, their saturation density was several times lower than previously reported brush transition densities, such as 65000 µm^-2 for 17 kDa polyacrylamide (R_F ~ 15 nm) on a solid surface (Wu et al., 2002). To compare our data on fluid bilayers with previously reported data on solid surfaces, we performed additional experiments with lipid bilayers that lost fluidity. No significant changes in protein binding between fluid and nonfluid bilayers were observed for both b-MUC1 and g-MUC1 molecules (Supplementary Figure 5B). This result suggests that membrane fluidity does not affect the average intermolecular distance or other relevant parameters that control molecular binding in the reconstituted system. Based on these, we speculate that the saturated protein density observed in our experiments is lower than or at most comparable to the actual brush transition density. Thus, although these crowded proteins may be restricted from free random motion, they are not significantly extended as in the condensed brush state, in which the contribution of resistance to molecular extension f_el is expected to be small relative to the overall free energy of the system.

Note that this does not mean that glycoproteins cannot form condensed brush structures: in fact, highly glycosylated molecules (e.g., MUC1) can form brush structures in cells when such proteins are expressed at very high densities. (Shurer et al., 2019). In these cells, ………. Such membrane deformation results in the increase of total surface area to reduce the density of glycoproteins, indicating that there is strong intermolecular repulsion between glycoproteins. In any case, the free energy of the system is determined by the balance between protein binding and insertion into the membrane, protein deformation, and repulsive forces between proteins, which determine the density of proteins depending on the configuration of the system. Thus, although strong intermolecular repulsions were prominently observed in our simplified system, this may not be the case in other systems. ……

(3) The MUC1 glycoprotein is reported to have a dramatic effect in reducing viral infection shown in Fig 1F. On the contrary, in a different experiment shown in Fig2D and Fig2H MUC1 has almost no effect in reducing viral infection. It is not clear how these two findings can be compatible.

The immunostaining results show that the density of MUC1 molecules is very low in the experimental system in Figure 2 (Figure 2C), which is supported by the SC-RNASeq data (as shown in Supplementary Figure 2A, MUC1 is not listed as a top molecule). In other words, the MUC1 expression level in this experiment is too low to affect virus infection inhibition. On the other hand, the Pearson correlation function represents the strength of the linear relationship between two variables, so it is not the most appropriate indicator for seeing the correlation with the MUC1 expression level, which has little change (Figure 2D, 2F). In fact, even TOS analysis, which can see the correlation by focusing on the cells with the highest expression level, cannot detect the correlation (Figure 2H).Therefore, the MUC1 data in Figure 2DFH will be annotated and corrected in the figure legend.

Figure2 Legend:

MUC1 has a small mean expression level and variance, and is more affected by measurement noise than other molecules when calculating the Pearson correlation function (Figure 2C-2F). In addition, the number of cells in which expression can be detected is small, so no significant correlation was detected by TOS analysis (Figure 2H).

(4) Why is there a shift in the use of the glycan marker? How does this affect the conclusions? For the infection correlation relating protein expression with glycan content the PNA-lectin was used together with flow cytometry. For imaging the infection and correlating with glycan content the SSA-lectin is used.

For each cell line, we selected the lectin that could be measured over the widest dynamic range. This lectin is thought to recognize the predominant glycan species in the cell line (Fig. S1C, Fig. 2D). In our model, we believe that viral infection inhibition is not specific to the type of sugar, but is highly dependent on the total amount of glycans. If this hypothesis is correct, the reason we used different lectins in each experiment is simply to select the lectin that recognizes the most predominant glycan species that is most convenient for predicting the total amount of glycans in cells. This hypothesis is consistent with our observations, where the total amount of glycans estimated by different lectins could explain the infection inhibition in a similar way in the experiments in Figures 1 and 2, and the TOS analysis in Figure 2 showed that minor glycans also have an infection inhibitory effect. On the other hand, it is of course possible to predict the total amount of glycans more accurately by obtaining as much information on glycans as possible (related to Q5). Based on the above discussion, the manuscript will be revised as follows.

Page5

Using HEK293T cell lines exogenously expressing genes of these proteins tagged with fluorescent markers, their glycosylation was measured by binding of a lectin from Arachis hypogaea (PNA), and the number of these proteins in the cells was measured simultaneously. PNA was used for the measurement because it has a wider dynamic range than other lectins (Supplementary Figure 1C). This suggests that GalNAc recognized by PNA is predominantly present on glycans of HEK293T cells, especially on the termini of glycans that are amenable to lectin binding, compared to other saccharides.. …

page9  

Our findings suggest that membrane glycoproteins nonspecifically inhibit viral infection, and we hypothesize that their inhibitory function is also nonspecific depending on the type of glycan. Our hypothesis is consistent with the observations in the TOS analysis. Although minor saccharide species in the system (such as GlcNAc and GalNAc recognized by DSA, WGA, or PNA) showed anticolocalization with infection, their scores were much lower than those of major saccharide species. This suggests that all major and minor saccharide species have an infection inhibitory effect, but cells enriched with minor type glycans are only partially present in the system, and the contribution of these cells to virus inhibition is also partial. It is also consistent with the observation that the amount of GalNAc recognized by PNA determines the virus infection inhibition in HEK 293T cells (Figure 1). Therefore, we believe that our assay using a single type of predominantly expressed lectin is still useful for estimating the total glycan content. Nevertheless, the virus infection rate may show a better correlation with a more accurately estimated total glycan in each cell. For example, the use of multiple lectins with appropriate calibration to integrate multiple signals to simultaneously detect a wider range of saccharide species would allow for more accurate estimation. It should be noted that the amount of bound lectin does not necessarily measure the overall glycan composition but likely reflects the sugar population at the free end of the glycan chain to which the lectin binds most.

(5) The authors in several instances comment on the relevance and importance of the total glycan content. Nevertheless, these conclusions are often drawn when using only one glycan-binding lectin. In fact, the anti-correlation with viral infection is distinct for the various lectins (Fig 2D and Fig 2H). Would it make more sense to use a combination of lectins to get a full glycan spectrum?

As stated in the answer to Q4, we believe that we were able to detect the infection-suppressing effect of the total glycan amount by using the measurement value of the major component glycan as an approximation. However, as you pointed out, if we could accurately measure the minor glycan components and add up their values, we believe that we could measure the total glycan amount more accurately. In order to measure multiple glycans simultaneously and with high accuracy, some kind of biochemical calibration may be necessary to compare the measurements of lectin-glycan pairs with different binding constants. We believe that these are very useful techniques, and would like to consider them as a future challenge. The corrections listed in Q4 are shown below.

(Page 9)

Nevertheless, the virus infection rate may show a better correlation with a more accurately estimated total glycan in each cell. For example, the use of multiple lectins with appropriate calibration to integrate multiple signals to simultaneously detect a wider range of glycans would allow for more accurate estimation. …….

(6) Fig 3A shows virus binding to HEK cells upon MUC1 expression. Please provide the surface expression of the MUC1 so that the data can be compared to Fig 1F. Nevertheless, it is not clear why the authors used MUC expression as a parameter to assess virus binding. Alternatively, more conclusive data supporting the hypothesis would be the absence of a correlation between total glycan content and virus binding capacity.

The relationship between the expression level of MUC1 in each cell and the amount of virus binding is shown in Supplementary Figure 3A. There is no correlation between the two. In HEK293T cells, many glycans are modified with MUC1, so MUC1 was used as the indicator for analysis (Supplementary Figure 1C). As you pointed out, it is better to use the amount of glycan as an indicator, so we analyzed the relationship between the amount of bound virus and the amount of glycan on the surface on the Calu-3 monolayer (Supplementary Figure 2F, 2G, introduced in the answer to Specific (Q1)). In any case, no correlation was found between virus binding and surface glycans. I will correct the manuscript as follows.

(page 9)

Glycans could be one of the biochemical substances that link the intracellular molecular composition and macroscopic steric forces at the cell surface. To clarify this connection, we further investigated the mechanism by which membrane glycoproteins inhibit viral infection. First, we measured viral binding to cells to determine which step of infection is inhibited. We found that a large number of SARS-CoV2-PP can still bind to cells even when cells expressed sufficient amounts of the glycoprotein that could account for the majority of glycans within these cells and inhibit viral infection (Figure 3A). Similarly, on the two-dimensional culture surface of Calu-3 cells, no correlation was observed between the number of viruses bound and the total amount of glycans on the cell surface (Supplementary Figure 2F-G). These results indicate that glycoproteins do not inhibit virus binding to cells, but rather inhibit the steps required for subsequent virus internalization.

(7) While the use of the Flory model could provide a simplification for a (disordered) flexible structure such as MUC1, where the number of amino acids equals N in the Flory model, this generalisation will not hold for all the proteins. Because folding will dramatically change the effective polypeptide chain-length and reduce available positioning of the amino acids, something the authors clearly measured (Fig 4G), this generalisation is not correct. In fact, the generalisation does not seem to be required because the authors provide an estimation for the effective Flory radius using their FRET approach

Current theories generalizing the Flory model to proteins are incomplete, and it is certainly not possible to accurately estimate the size of individual molecules undergoing different folding. However, we found such a generalized model to be useful in understanding the overall properties of membrane proteins. In our experiments, we were indeed able to obtain the R_Fs of some individual molecules by FRET measurements. However, this modeling made it possible to estimate the distribution range of the RFs, including for larger proteins that cannot be measured by FRET. For example, from our results, we can estimate that the upper limit of the RFs of the longest membrane proteins is about 10.5 nm, assuming that the proteins follow the Flory model in all respects except for the shortening of the effective length due to folding. These analyses are useful for physical modeling of nonspecific phenomena, as in our case.

In order to discuss the balance between such theoretical validity and the convenience of practical handling, we revise the manuscript as follows.

(page 13) 

This shift in ν indicates that glycosylation increases the size of the protein at equilibrium, but the change in R_F is slight, e.g., a 1.3-fold increase for one of the longest ectodomains with N = 4000 when these values of ν are applied. This calculation also gives a rough estimate of the upper limit of the R_F of the extracellular domains of all membrane proteins in the human genome (approximately 10.5 nm). Physically, this change in ν by glycosylation may be caused by the increased intramolecular exclusion induced sterically between glycan chains. This estimated ν are much smaller than that of 0.6 for polymers in good solvents, possibly due to protein folding or anchoring effects on the membrane. In fact, the ν of an intrinsically disordered protein in solution has been reported to be close to 0.6 (Riback et al., 2019; Tesei et al., 2024). Overall, these analyses using the Flory model provide information on the size distribution of membrane proteins and the influence of glycans, although the model cannot predict the exact size of each protein due to its specific folding.

MINOR COMMENTS/EDITS:

(1) In Figures 2A and 2C, as well as Supplemental Figure 2C, the fluorescent images indicate that GFP expression differs among the various groups. Ideally, these should be at the same GFP expression level, as the glycan and antibody staining occurred post-viral infection. For instance, ACE2 is a well-known positive control and should enhance SARS-CoV-2 infection. Yet, based on the findings presented in Supplemental Figure 2C, ACE2 appears to correlate with the lowest infection rate. The relationship between the infection rate and key glycoproteins needs clearer quantification.

We measured the virus inhibition effect specific to each molecule using a cell line expressing low levels of viral receptors and glycoproteins (Fig. 1). On the other hand, the system in Fig. 2 contains diverse viral receptors and glycoproteins and has not been genetically manipulated. (We apologize that there was a typo in our description of experiment, which will be corrected, as shown below). The variation in infection rate between samples was caused by multiple factors but was not related to the molecule for which the correlation was measured. The receptor-based normalization used in the experiment in Fig. 1 cannot be applied in this system in Fig.2 due to the complexity of the gene expression profile. Therefore, instead of such parameter-based normalization, we applied Pearson correlation and TOS analysis. In the calculation of Pearson correlation, intensities are normalized. TOS analysis allows the analysis of colocalization between the groups with the highest fluorescence intensity. Therefore, in both cases of variation in overall infection rate and variation in the distribution of infected populations, samples with large variations can be reasonably compared by Pearson correlation and TOS analysis, respectively. We extend the discussion on statistics and revise the manuscript as follows.

(page 8-9)

To test this hypothesis, we infected a monolayer of epithelial cells endogenously expressing highly heterogeneous populations of glycoproteins with SARS-CoV-2-PP, and measured viral infection from cell to cell visually by microscope imaging. …

Pearson correlation is effective for comparing samples with varying scales of data because it normalizes the data values by the mean and variance. However, as observed in our experiments, this may not be the case when the distribution of data within a sample varies between samples. In addition, as has already been reported, the distribution of infected cells often deviates significantly from the normal distribution of data that is the premise of Pearson correlation (Russell et al., 2018) (Figure 2B). To further analyze data in such nonlinear situations, we applied the threshold overlap score (TOS) analysis (Figure 2G-H, Supplementary Figure 2E). This is one statistical method for analyzing nonlinear correlations, and is specialized for colocalization analysis in dual color images (Sheng et al., 2016). TOS analysis involves segmentation of the data based on signal intensity, as in other nonlinear statistics (Reshef et al., 2011). The computed TOS matrix indicates whether the number of objects classified in each region is higher or lower than expected for uniformly distributed data, which reflects co-localization or anti-localization in dual-color imaging data. For example, calculated TOS matrices show strong anti-localization for infection and glycosylation when both signals are high (Figure 2GH). This confirms that high infection is very unlikely to occur in cells that express high levels of glycans. The TOS analysis also yielded better anti-localization scores for some of the individual membrane proteins, especially those that are heterogeneously distributed across cells (Figure 2H). This suggests that TOS analysis can highlight the inhibitory function of molecules that are sparsely expressed among cells, reaffirming that high expression of a single type of glycoprotein can create an infection-protective surface in a single cell and that such infection inhibition is not protein-specific. In contrast, for more uniformly distributed proteins such as the viral receptor ACE2, TOS analysis and Pearson correlation showed similar trends, although the two are mathematically different (Figure 2D, 2H). Because glycoprotein expression levels and virus-derived GFP levels were treated symmetrically in these statistical calculations, the same logic can be applied when considering the heterogeneity of infection levels among cells. Therefore, it is expected that TOS analysis can reasonably compare samples with different virus infection level distributions by focusing on cells with high infection levels in all samples.

(2) For clarity, the authors should consider separating introductory and interpretive remarks from the presentation of results. These seem to get mixed up. The introduction section could be expanded to include more details about glycoproteins, their relevance to viral infection, and explanations of N- and O-glycosylation.

Following the suggestion, (1) we added an explanation of the relationship between glycoproteins and viral infection, and N-glycosylation and O-glycosylation to the Introduction section, and (2) moved the introductory parts in the Results section to the Introduction section, as follows.

（１; page3）

While there are known examples of glycans that function as viral receptors (Thompson et al., 2019), these results demonstrate that a variety of glycoproteins negatively regulate viral infection in a wide range of systems. These glycoprotein groups have no common amino acid sequences or domains. The glycans modified by these proteins include both the N-type, which binds to asparagine, and the O-type, which binds to serine and threonine. Furthermore, there have been no reports of infection-suppressing effects according to the specific monosaccharide type in the glycan. All of these results suggest that bulky membrane glycoproteins nonspecifically inhibit viral infection.

(2 : Page 4-5)

To confirm that glycans are a general chemical factor of steric repulsion, an extensive list of glycoproteins on the cell membrane surface would be useful. The wider the range of proteins to be measured, the better. Therefore, we collect information on glycoproteins on the genome and compile them into a list that is easy to use for various purposes. Then, by analyzing sample molecules selected from this list, it may be possible to infer the effect of the entire glycoprotein population on the steric inhibition of virus infection, despite the complexity and diversity of the Glycome (Dworkin et al., 2022; Huang et al., 2021; Moremen et al., 2012; Rademacher et al., 1988). Elucidation of the mechanism of how glycans regulate steric repulsion will also be useful to quantitatively discuss the relationship between steric repulsion and intracellular molecular composition. For this purpose, we apply the theories of polymer physics and interface chemistry.

Results

List of membrane glycoproteins in human genome and their inhibitory effect on virus infection

To test the hypothesis that glycans contribute to steric repulsion at the cell surface, we first generate a list of glycoproteins in the human genome and then measure the glycan content and inhibitory effect on viral infection of test proteins selected from the list (Figure 1A). To compile the list of glycoproteins, we ….

(3) In the sentence, "glycoproteins expressed lower than CD44 or other membrane proteins including ERBB2 did not exhibit any such correlation, although ERBB2 expressed ~4 folds higher amount than CD44 and shared ~7% among all membrane proteins," it is unclear which protein has a higher expression level: CD44 or ERBB2? Furthermore, the use of the word "although" needs clarification.

Corrected as follows:

(page 8)

……showed a weak inverse correlation with viral infection; even such a weak correlation was not observed with other proteins, including ERBB2, which is approximately four-fold more highly expressed than CD44

(4) In Supplementary Figure 5, please provide an explanation of the data in the figure legend, particularly what the green and red signals represent.

Corrected as follows:

STORM images of all analyzed cells, expressing designated proteins. The detected spots of SNAPsurface Alexa 647 bound to each membrane protein are shown in red, and the spots of CF568conjugated anti-mouse IgG secondary antibody that recognizes Spike on SARS-CoV2-PP are shown in green. For cells, a pair of two-color composite images and a CF658-only image are shown. Numbers on axes are coordinates in nanometer.

(5) It would be good to see a comprehensive demonstration of the exact method for estimation of membrane protein density (in the SI), since this is an integral part of many of the analyses in this paper. The method is detailed in the Methods section in text and is generally acceptable, but this methodology can vary quite widely and would be more convincing with calibration data provided.

We added flow cytometry and fluorometer data for calibration (Supplementary Figure 1L,M) and introduced a sentence explaining the procedure for obtaining the values used for calibration as follows:

(page 54)

…….Liposome standards containing fluorescent molecules (0.01– 0.75 mol% perylene (Sigma), 0.1– 1.25 mol% Bodipy FL (Thermo), and 0.005– 0.1% DiD) as well as DOPC (Avanti polar lipids) were measured in flow cytometry (Supplmentary Figure 1L). Meanwhile, by fluorimeter, fluorescence signals of these liposomes and known concentrations of recombinant mTagBFP2, AcGFP and TagRFP-657 proteins and SNAP-Surface 488 and Alexa 647 dyes (New England Biolabs) were measured in the same excitation and emission ranges as in flow cytometry assays (Supplementary Figure 1M). Ratios between the integral of fluorescent intensities in this range between two dyes of interest are used for converting the signals measured in flow cytometry. Additional information needed for calibration is the size difference between liposomes and cells. The average diameter of liposomes is measured to be 130 nm, and the diameter of HEK 293T cells is estimated to be 13 µm (Furlan et al., 2014; Kaizuka et al., 2021b; Ushiyama et al., 2015). From these data, the signal from cells acquired by flow cytometry can be calibrated to molecular surface density. For example, the Alexa 647 signal acquired by flow cytometry can be converted to the signal of the same concentration of DID dye using fluorometer data, but the density of the dye is unknown at this point. This converted DID signal can then be calibrated to the density on liposomes rather than cells using liposome flow cytometry data. Finally, adjusted for the size difference between liposomes and cells, the surface molecular density on cells is determined. By going through one cycle of these procedures, we could obtain calibration unit, such as 1 flow cytometry signal for a cell in the designated illumination and detection setting = 0.0272 mTagBFP2 µm^-2 on cell surface.

(Figure legend, Supporting Figure 1: )

… L. Flow cytometry measurements for liposomes containing serially diluted dye-conjugated lipids and fluorescent membrane incorporating molecules (Bodipy-FL, peryelene, and DID) with indicated mol%. Linear fitting shown was used for calibration.  M. Fluorescence emission spectrum for equimolar molecules (50µM for green and far-red channels, and 100µM for blue channel), excited at 405 nm, 488 nm, and 638 nm, respectively. Membrane dyes were measured as incorporated in liposomes. Purified recombinant mTagBFP2 was used.

(6) Fig 2A: The figure legend should describe the microscopy method for a quick and easy reference.

Corrected as follows:

(Figure legend, Figure 2)

A. Maximum projection of Z-stack images at 1 µm intervals taken with a confocal microscope. SARSCoV2-pp-infected, air-liquid interface (ALI)-cultured Calu-3 cell monolayers were chemically fixed and imaged by binding of Alexa Fluor 647-labeled Neu5AC-specific lectin from Sambucus sieboldiana (SSA) and GFP expression from the infecting virus.

(7) Fig 2B: what is the color bar supposed to represent? Is it the pixel density per a particular value? Units and additional description are required. In addition, these are "arbitrary units" of fluorescence, but you should tell us if they've been normalized and, if so, how. They must have been normalized, since the values are between 0 and 1, but then why does the scale bar for SSA only go to 0.5?

The color bar shows the number of pixels for each dot, resulting in the scale for density scatter plot. The scale on the X-axis was incorrect. All these issues have been fixed in this revision, in the figure and in the legend as follows.

(Figure legend, Figure 2)

B. Density scatter plot of normalized fluorescence intensities in all pixels in Figure 2A in both GFP and SSA channels. Color indicates the pixel density.  

(8) Fig 3D has a typo: this should most likely be "grafted polymer."

(9) Fig 3E has a suspected typo: in the text, the author uses the word "exclusion" instead of "extrusion." The former makes more sense in this context.

(10) Fig 5A has a typo: "Suppoorted" instead of Supported Lipid Bilayer.

(11) Fig 7E-F has a suspected typo: Again, this should most likely be the word "exclusion" instead of "extrusion."

Thank you so much for pointing out these mistakes, I have corrected them all as suggested.

(12) Which other molecules are referred to, on page 6 (middle), that do not have an inhibitory effect? Please specify.

We specified the molecules that have inhibitory effects, and revised as follows: 

These proteins include those previously reported (MUC1, CD43) as well as those not yet reported (CD44, SDC1, CD164, F174B, CD24, PODXL) (Delaveris et al., 2020; Murakami et al., 2020). In contrast, other molecules (VCAM-1, EPHB1, TMEM123, etc.) showed little inhibitory effect on infection within the density range we used.

(13) Fig 2 B: the color LUT is not labelled nor explained.

Corrected as described in (7)

(14) Please provide the scale bars for figures Fig 2A, C, E and Suppl Fig 2C, D.

Corrected. 

(15) Please provide the name for the example of a 200 aa protein that is meant to inhibit viral infection but is not bigger than ACE2. Also providing the densities in Fig 3A would help to correlate the data to Fig 1F.

Corrected as follows: 

(page 10)

We found that a large number of SARS-CoV2-PP can still bind to cells even when cells expressed sufficient amounts of the glycoprotein (mean density ~50 µm^-2) that could account for the majority of glycans within these cells and inhibit viral infection (Figure 3A). …..

In our measurements, a protein with extracellular domain of ~200 amino acids (e.g. CD164 (138aa)) at a density of ~100 μm-2 showed significant inhibition in viral infection. This molecule is shorter than the receptor ACE2 (722 aa),

(16) In the experiments conducted in HeK cells expressing the different glycoproteins studies it is mentioned that results of infection were normalised by the amount ACE2 expression. Is the expression of receptor homogenous in the experiments conducted in Figure 2? Clarify in the methods if the expression of receptor has been quantified and somehow used to correct the intensity values of GFP used to determine infection.

As also explained for Q1, the system in Fig. 2 contains diverse viral receptors and glycoproteins, and the receptor-based normalization used in the experiment in Fig. 1 cannot be applied. Instead, we applied Pearson correlation and TOS analysis. In the calculation of Pearson correlation, intensities are normalized. TOS analysis allows the analysis of colocalization between the groups with the highest fluorescence intensity. Therefore, in both cases of variation in overall infection rate and variation in the distribution of infected populations, samples with large variations can be reasonably compared by Pearson correlation and TOS analysis, respectively. We extend the discussion on statistics and revise the manuscript as follows.

(page 8-9)

Pearson correlation is effective for comparing samples with varying scales of data because it normalizes the data values by the mean and variance. However, as observed in our experiments, this may not be the case when the distribution of data within a sample varies between samples. In addition, as has already been reported, the distribution of infected cells often deviates significantly from the normal distribution of data that is the premise of Pearson correlation (Russell et al., 2018) (Figure 2B). To further analyze data in such nonlinear situations, we applied the threshold overlap score (TOS) analysis (Figure 2G-H, Supplementary Figure 2E). This is one statistical method for analyzing nonlinear correlations, and is specialized for colocalization analysis in dual color images (Sheng et al., 2016). TOS analysis involves segmentation of the data based on signal intensity, as in other nonlinear statistics (Reshef et al., 2011). The computed TOS matrix indicates whether the number of objects classified in each region is higher or lower than expected for uniformly distributed data, which reflects co-localization or anti-localization in dual-color imaging data. For example, calculated TOS matrices show strong anti-localization for infection and glycosylation when both signals are high (Figure 2GH). This confirms that high infection is very unlikely to occur in cells that express high levels of glycans. The TOS analysis also yielded better anti-localization scores for some of the individual membrane proteins, especially those that are heterogeneously distributed across cells (Figure 2H). This suggests that TOS analysis can highlight the inhibitory function of molecules that are sparsely expressed among cells, reaffirming that high expression of a single type of glycoprotein can create an infection-protective surface in a single cell and that such infection inhibition is not protein-specific. In contrast, for more uniformly distributed proteins such as the viral receptor ACE2, TOS analysis and Pearson correlation showed similar trends, although the two are mathematically different (Figure 2D, 2H). Because glycoprotein expression levels and virus-derived GFP levels were treated symmetrically in these statistical calculations, the same logic can be applied when considering the heterogeneity of infection levels among cells. Therefore, it is expected that TOS analysis can reasonably compare samples with different virus infection level distributions by focusing on cells with high infection levels in all samples.

(17) Can you provide additional details about the method of thresholding to eliminate "background" localisations in STORM?

Method section was corrected as follows: 

(page 59)

…Viral protein spots not close to cell membranes were eliminated by thresholding with nearby spot density for cell protein. Specifically, the entire image was pixelated with a 0.5µm square box and all viral protein signals within the box that had no membrane protein signals were removed. Also, viral protein spots only sparsely located were eliminated by thresholding with nearby spot density for viral protein. This thresholding process removed any detected viral protein spot that did not have more than 100 other viral protein spots within 1µm.

(18) The article says "It was shown that the number of bound lectins correlated with the amount of glycans, not with number of proteins (Figure 1E)". Figure 1E correlates experimental PNA/mol with predicted glycosylation sites, not with the number of expressed proteins. Correct sentence with the right Figure reference.

As you pointed out, the meaning of this sentence was not clear. We have amended it as follows to clarify our intention:

(page 8)

Since a wide range of glycoproteins inhibit viral infection, it is possible that all types of glycoproteins have an additive effect for this function. ……. In this cell line, this inverse correlation was most pronounced when quantifying N-acetylneuraminic acid (Neu5AC, recognized by lectins SSA and MAL) compared to the various types of glycans, while some other glycans also showed weak correlations (Supplementary Figure 2C). These results showed that the amount of virus infection in cell anticorrelated with the amount of total glycans on the cell surface. As amount of glycans is determined by the total population of glycocalyx, infection inhibitory effect can be additive by glycoprotein populations as we hypothesized.

If the inhibitory effect is nonspecific and additive, the contribution of each protein is likely to be less significant. To confirm this, we also measured the correlation between the density of each glycoprotein and viral infection. CD44, which was shown to…….. Our results demonstrate that total glycan content is a superior indicator than individual glycoprotein expression for assessing infection inhibition effect generated by cell membrane glycocalyx. These results are consistent with our hypothesis regarding the additive nature of the nonspecific inhibitory effects of each glycoprotein.

eLife Assessment

Endothelial cell-specific loss of TGF-beta signaling in mice leads to CNS vascular defects, specifically impairing retinal development and promoting immune cell infiltration. The data are solid, showing that loss of TGF-beta signaling triggers vascular inflammation and attracts immune cells specific to CNS vasculature. These findings are important, highlighting TGF-beta's role in maintaining vascular-immune homeostasis and its therapeutic potential in neurovascular inflammatory diseases.

Reviewer #1 (Public review):

Summary:

The manuscript analyses the effects of deleting the TgfbR1 and TgfbR2 receptors from endothelial cells at postnatal stages on vascular development and blood-retina barrier maturation in the retina. The authors find that deletion of these receptors affects vascular development in the retina but importantly it affects the infiltration of immune cells across the vessels in the retina. The findings demonstrate that Tgf-beta signaling through TgfbR1/R2 heterodimers regulates primarily the immune phenotypes of endothelial cells in addition to regulating vascular development, but has minor effects on the BRB maturation. The data provided by the authors provides a solid support for their conclusions.

Strengths:

(1) The manuscript uses a variety of elegant genetic studies in mice to analyze the role of TgfbR1 and TgfbR2 receptors in endothelial cells at postnatal stages of vascular development and blood-retina barrier maturation in the retina.

(2) The authors provide a nice comparison of the vascular phenotypes in endothelial-specific knockout of TgfbR1 and TgfbR2 in the retina (and to a lesser degree in the brain) with those from Npd KO mice (loss of Ndp/Fzd4 signaling) or loss of VEGF-A signaling to dissect the specific roles of Tgf-beta signaling for vascular development in the retina.

(3) The snRNAseq data of vessel segments from the brains of WT versus TgfbR1 -iECKO mice provides a nice analysis of pathways and transcripts that are regulated by Tgf-beta signaling in endothelial cells.

Weaknesses (Original Submission):

(1) The authors claim that choroidal neovascular tuft phenotypes are similar in TgfbrR1 KO and TgfbrR2 KO mice. However, the phenotypes look more severe in the TgfbrR1 KO rather than TgfbrR2 KO mice. Can the authors show a quantitative comparison of the number of choroidal neovascular tufts per whole eye cross-section in both genotypes?

(2) In the analysis of Sulfo-NHS-Biotin leakage in the retina to assess blood-retina barrier maturation, the authors claim that there is increased vascular leakage in the TgfbR1 KO mice. However, there does not seem like Sulfo-NHS-biotin is leaking outside the vessels. Therefore, it cannot be increased vascular permeability. Can the authors provide a detailed quantification of the leakage phenotype?

(3) The immune cell phenotyping by snRNAseq seems premature as the number of cells is very small. The authors should sort for CD45+ cells and perform single cell RNA sequencing.

(4) The analysis of BBB leakage phenotype in TgfbR1 KO mice needs to be more detailed and include some tracers in addition to serum IgG leakage.

(5) A previous study (Zarkada et al., 2021, Developmental Cell) showed that EC-deletion of Alk5 affects the D tip cells. The phenotypes of those mice look very similar to those shown for TgfbrR1 KO mice. Are D tip cells lost in these mutants by snRNAseq?

Comments on revisions:

The authors have addressed the major weaknesses that I raised with the original submission adequately in the revised manuscript.

Reviewer #2 (Public review):

Summary:

The authors meticulously characterized EC-specific Tgfbr1, Tgfbr2, or double knockout in the retina, demonstrating through convincing immunostaining data that loss of TGF-β signaling disrupts retinal angiogenesis and choroidal neovascularization. Compared to other genetic models (Fzd4 KO, Ndp KO, VEGF KO), the Tgfbr1/2 KO retina exhibits the most severe immune cell infiltration. The authors proposed that TGF-β signaling loss triggers vascular inflammation, attracting immune cells - a phenotype specific to CNS vasculature, as non-CNS organs remain unaffected.

Strengths:

The immunostaining results presented are clear and robust. The authors performed well-controlled analyses against relevant mouse models. snRNA-seq corroborates immune cell leakage in the retina and vascular inflammation in the brain.

Comments on revisions:

The authors have revised the manuscript and addressed all my questions.

Author response:

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

Reviewer #1 (Public review): 

Weaknesses: 

(1) The authors claim that choroidal neovascular tuft phenotypes are similar in TgfbrR1 KO and TgfbrR2 KO mice. However, the phenotypes look more severe in the TgfbrR1 KO rather than TgfbrR2 KO mice. Can the authors show a quantitative comparison of the number of choroidal neovascular tufts per whole eye cross-section in both genotypes? 

Thank you for asking about this.  Each VE-cad-CreER;TGFBR1 CKO/- and VE-cad-CreER;TGFBR2 CKO/- retina exhibits multiple zones of choroidal neovascularization.  The examples in Figures 1 and Figure 1 – Figure supplements 1 and 2 are mostly from retinas with loss of TGFBR1, but we could have chosen similar examples from retinas with loss of TGFBR2.  The quantification in the original version of Figure 1- Figure supplement 1 panel C had a labeling error.  It actually showed the quantification choroidal neovascularization (CNV) in the sum of both VE-cad-CreER;TGFBR1 CKO/- and VE-cad-CreER;TGFBR2 CKO/- retinas, not only in VE-cad-CreER;TGFBR1 CKO/- retinas as originally labeled.  The point that it made is that CNV is seen with loss of TGF-beta signaling but not in control retinas or retinas with loss of Norrin signaling.  We have now updated that plot by separating the data points for VE-cad-CreER;TGFBR1 CKO/- and VE-cad-CreER;TGFBR2 CKO/- retinas, so that they can be compared to each other.   The result shows ~2.5-fold more CNV in VE-cad-CreER;TGFBR2 CKO/- retinas compared to VE-cad-CreER;TGFBR1 CKO/-.  We think it likely that a more extensive sampling would show little or no difference between these two genotypes – but the data is what it is. This is now described in the Results section. 

We have also added a panel D to Figure 1- Figure supplement 1, which shows a retina flatmount analysis of CNV.  This is done by mounting the retina with the photoreceptor side up so that the outer retina can be optimally imaged. 

(2) In the analysis of Sulfo-NHS-Biotin leakage in the retina to assess blood-retina barrier maturation. The authors claim that there is increased vascular leakage in the TgfbR1 KO mice. However, it does not seem like Sulfo-NHS-biotin is leaking outside the vessels. Therefore, it cannot be increased vascular permeability. Can the authors provide a detailed quantification of the leakage phenotype? 

Thank you for raising this point.  Your comment prompted us to look at this question in greater depth with more experiments.  We have expanded Figure 2 to show and quantify a comparison between control (i.e. phenotypically WT), NdpKO, and TGFBR1 endothelial KO and we have expanded the associated part of the Results section (Figure 2C and D).  In a nutshell, control retinas show little Sulfo-NHS-biotin accumulation in or around the vasculature or in the parenchyma; NdpKO retinas show Sulfo-NHS-biotin accumulation in the vasculature and in the parenchyma (i.e., the area between the vessels); and VEcadCreER;Tgfbr1CKO/- retinas show Sulfo-NHS-biotin accumulation in the vascular tufts with minimal accumulation in the non-tuft vasculature and minimal leakage into the parenchyma.   The conclusion is that the bulk of the retinal vasculature in TGFBR1 endothelial KO mice is minimally or not at all leaky – very different from the situation with loss of Norrin/Frizzled4 signaling.

(3) The immune cell phenotyping by snRNAseq is premature, as the number of cells is very small. The authors should sort for CD45+ cells and perform single-cell RNA sequencing. 

Thank you for raising this point.  For the revised manuscript, we have performed additional snRNAseq analyses using the same tissue processing protocol as for our original snRNAseq data.  We have opted to homogenize the tissue and prepare nuclei (our original method) rather than dissociate the tissue and FACS sorting for CD45+ cells because the nuclear isolation approach is unbiased – we assume that nuclei from all cell types are present after tissue homogenization.  By contrast, we cannot be certain that CD45 FACS will capture the full range of immune cells since some cells may not express CD45, may express CD45 at low level, or may be tightly adherent to other cells, such as vascular endothelial cell.  Additionally, by following the original protocol, we can combine the original snRNAseq dataset and the new snRNAseq dataset.  In the revised manuscript we present the snRNAseq data from the combination of the original and the more recent snRNAseq datasets (revised Figure 4; N=628 immune cell nuclei).  The new analysis comes to the same conclusions as the original analysis: the immune cell infiltrate in the mutant retinas is composed of a wide variety of immune cells.

(4) The analysis of BBB leakage phenotype in TgfbR1 KO mice needs to be more detailed and include tracers as well as serum IgG leakage. 

As described in our response to query 2, we have conducted additional experiments to look at vascular leakage in control, VE-cad-CreER;TGFBR1 CKO/-, and NdpKO retinas.  We have also looked at Sulfo-NHS-biotin leakage in the VE-cadCreER;TGFBR1 CKO/- brain, and it is indistinguishable from WT controls.  Since Sulfo-NHS-biotin is a low MW tracer (<1,000 kDa), this implies that loss of TGF-beta signaling does not increase non-specific diffusion of either low or high MW molecules.  Therefore, the elevated levels of IgG in the brain parenchyma in young VE-cad-CreER;TGFBR1 CKO/- mice (Figure 8A) likely represents specific transport of IgG across the BBB.  Such transport is known to occur via Fc receptors expressed on vascular endothelial cells, although it is normally greater in the brain-to-blood direction than in the blood-to-brain direction.  For example, see Lafrance-Vanasse et al (2025) Leveraging neonatal Fc receptor (FcRn) to enhance antibody transport across the blood brain barrier.  Nat Commun. 16:4143.  This is now described in greater detail in the Results section.

(5) A previous study (Zarkada et al., 2021, Developmental Cell) showed that EC-deletion of Alk5 affects the D tip cells. The phenotypes of those mice look very similar to those shown for TgfbrR1 KO mice. Are D-tip cells lost in these mutants by snRNAseq? 

Please note: Alk5 is another name for TGFBR1.  This is noted in the second sentence of paragraph 4 of the Introduction.  The reviewer is correct: there are a lot of similarities because these are exactly the same KO mice.  Also, Zarkada and we used the same VEcadCreER to recombine the CKO allele.  The proposed snRNAseq analysis would serve as an independent check on the diving (D) tip vs stalk cell analyses published in Zarkada et al (2021) Specialized endothelial tip cells guide neuroretina vascularization and blood-retina-barrier formation. Dev Cell 56:2237-2251.  We have not gone in this direction because the question of tip vs. stalk cells and of subtypes of tip cells in WT vs. mutant retinas is beyond our focus on choroidal neovascularization and the role of immune cells and vascular inflammation.  The proposed snRNAseq analysis would also require a major effort since tip cells are rare and must be harvested from large numbers of early postnatal retinas followed by FACS enrichment for vascular endothelial cells.  Finally, we have no reason to doubt the results of Zarkada et al.

Reviewer #2 (Public review): 

Summary:

The authors meticulously characterized EC-specific Tgfbr1, Tgfbr2, or double knockout in the retina, demonstrating through convincing immunostaining data that loss of TGF-β signaling disrupts retinal angiogenesis and choroidal neovascularization. Compared to other genetic models (Fzd4 KO, Ndp KO, VEGF KO), the Tgfbr1/2 KO retina exhibits the most severe immune cell infiltration. The authors proposed that TGF-β signaling loss triggers vascular inflammation, attracting immune cells - a phenotype specific to CNS vasculature, as non-CNS organs remain unaffected. 

Strengths: 

The immunostaining results presented are clear and robust. The authors performed well-controlled analyses against relevant mouse models. snRNA-seq corroborates immune cell leakage in the retina and vascular inflammation in the brain. 

Weaknesses: 

The causal link between TGF-β loss, vascular inflammation, and immune infiltration remains unresolved. The authors' model posits that EC-specific TGF-β loss directly causes inflammation, which recruits immune cells. However, an alternative explanation is plausible: Tgfbr1/2 KO-induced developmental defects (e.g., leaky vessels) permit immune extravasation, subsequently triggering inflammation. The observations that vein-specific upregulation of ICAM1 staining and the lack of immune infiltration phenotypes in the non-CNS tissues support the alternative model. Late-stage induction of Tgfbr1/2 KO (avoiding developmental confounders) could clarify TGF-β's role in retinal angiogenesis versus anti-inflammation. 

Thank you for raising this point.  Your comment prompted us to look at this question in greater depth with more experiments.  We have expanded Figure 2 to show and quantify a comparison between control (i.e. phenotypically WT), NdpKO, and TGFBR1 endothelial KO and we have expanded the associated part of the Results section (Figure 2C and D).  In a nutshell, control retinas show little Sulfo-NHS-biotin accumulation in or around the vasculature or in the parenchyma; NdpKO retinas show Sulfo-NHS-biotin accumulation in the vasculature and in the parenchyma (i.e., the area between the vessels); and VEcadCreER;Tgfbr1CKO/- retinas show Sulfo-NHS-biotin accumulation in the vascular tufts with minimal accumulation in the non-tuft vasculature and minimal leakage into the parenchyma.   The conclusion is that the bulk of the retinal vasculature in TGFBR1 endothelial KO mice is minimally or not at all leaky – very different from the situation with loss of Norrin/Frizzled4 signaling.

In the revised manuscript, we have expanded the Discussion section to address the two alternative hypotheses raised by the reviewer.  Here are the relevant data in a nutshell: (1) vascular leakage into the parenchyma, as measured with sulfo-NHSbiotin, in TGFBR1 endothelial CKO retinas is far less than in NdpKO retinas, where nearly all ECs convert to a fenestration+ (PLVAP+) phenotype and there is leakage of sulfo-NHS-biotin, (2) ICAM1 in ECs in TGFBR1 endothelial CKO retinas increases several-fold more than in NdpKO or Frizzled4KO retinas, (3) TGFBR1 endothelial CKO retinas have more infiltrating immune cells than NdpKO or Frizzled4KO retinas, and (4) in TGFBR1 endothelial CKO retinas large numbers of immune cells are observed within and adjacent to blood vessels.  We think that the simplest explanation for these data is that loss of TGFbeta signaling in ECs causes an endothelial inflammatory state with enhanced immune cell extravasation.  That said, the case for this model is not water-tight, and there could be less direct mechanisms at play.  In particular, this model does not explain why the inflammatory phenotype is limited to CNS (and especially retinal) vasculature.

Regarding the last sentence of the reviewer’s comment (“Late stage induction…”), we have tried activating CreER recombination at different ages and we observe a large reduction in the inflammatory phenotype when recombination is initiated after vascular development is complete.   This observation suggests that the vascular developmental/anatomic defect – and perhaps the resulting retinal hypoxia response – is required for the inflammatory phenotype.  In the revised manuscript we have expanded the Results and Discussion sections to describe this observation.

Reviewer #1 (Recommendations for the authors): 

Suggestions for experiments: 

(1) The authors need to show a quantitative comparison of the number of choroidal neovascular tufts per whole eye crosssection in both genotypes (TgfbR1 and TgfbR2 KO mice). 

Thank you for raising this point.  The quantification in the original version of Figure 1- Figure supplement 1 panel C was mis-labeled.  It quantifies choroidal neovascularization (CNV) in both VE-cad-CreER;TGFBR1 CKO/- and VE-cadCreER;TGFBR2 CKO/- retinas, not VE-cad-CreER;TGFBR1 CKO/- retinas only as originally labeled.  The point it makes is that CNV is seen with loss of TGF-beta signaling but not in control retinas or retinas with loss of Norrin signaling.  We have now corrected that plot by separating the data points for VE-cad-CreER;TGFBR1 CKO/- and VE-cad-CreER;TGFBR2 CKO/- retinas, so that they can be compared to each other.   The result shows ~2.5-fold more CNV in VE-cad-CreER;TGFBR2 CKO/- retinas compared to VE-cad-CreER;TGFBR1 CKO/-.  This is now described in the Results section. 

(2) In the analysis of Sulfo-NHS-Biotin leakage in the retina to assess blood-retina barrier maturation. The authors should provide a detailed quantification of the leakage phenotype outside the vessels into the CNS parenchyma, both in the retina and brain, in TgfbR1 KO mice. 

Thank you for raising this point.  There is no detectable Sulfo-NHS-biotin leakage into the brain parenchyma in VE-cadCreER;TGFBR1 CKO/- mice.  We have expanded Figure 2 to show and quantify the data for retinal vascular leakage (Figure 2C and D).  The data show that in VE-cad-CreER;TGFBR1 CKO/- mice there is accumulation of Sulfo-NHS-biotin in the vascular tufts but minimal accumulation elsewhere in the retinal vasculature and minimal leakage of Sulfo-NHS-biotin into the retinal parenchyma.

(3) The immune cell phenotyping by snRNAseq is premature, as the number of cells is very small. The authors should sort for CD45+ cells and perform single-cell RNA sequencing to ascertain these preliminary data. 

Thank you for raising this point.  We have performed additional snRNAseq analyses using the same tissue processing protocol as for our original snRNAseq data to increase the numbers of cells.  We have opted to homogenize the tissue and prepare nuclei (our original method) rather than dissociating the cells and FACS sorting for CD45+ cells because the nuclear isolation approach is unbiased – we assume that nuclei from all cell types are present.  By contrast, we cannot be certain that CD45 FACS will capture the full range of immune cells, since some cells may not express CD45, may express CD45 at low level, or may be tightly adherent to other cells, such as vascular endothelial cell.  Additionally, by following the original protocol, we can combine the original snRNAseq dataset of and the new snRNAseq dataset.  In the revised manuscript we present the snRNAseq data from the combination of the original and the more recent snRNAseq datasets (revised Figure 4; N=628 immune cell nuclei).  The new analysis comes to the same conclusion as in the original submission, namely that the immune cell infiltrate in the mutant retinas is composed of a wide variety of immune cells.  The Results section has been expanded to describe this new data and analysis.    

(4) The analysis of BBB leakage phenotype in TgfbR1 KO mice needs to be more detailed and include tracers as well as serum IgG leakage. 

Sulfo-NHS biotin leakage in the VE-cad-CreER;TGFBR1 CKO/- brain is minimal, and it is indistinguishable from WT controls.  Since Sulfo-NHS biotin is a low MW tracer (<1,000 kDa), this implies that loss of TGF-beta signaling does not increase non-specific diffusion of either low or high MW molecules.  Therefore, the elevated levels of IgG in the brain parenchyma in young VE-cad-CreER;TGFBR1 CKO/- mice (Figure 8A) likely represents specific transport of IgG across the BBB.  Such transport is known to occur via Fc receptors expressed on vascular endothelial cells, although it is normally greater in the brain-to-blood direction than in the blood-to-brain direction.  For example, see Lafrance-Vanasse et al (2025) Leveraging neonatal Fc receptor (FcRn) to enhance antibody transport across the blood brain barrier.  Nat Commun. 16:4143.  This is now described in greater detail in the Results section.

(5) The authors should perform a more detailed RNAseq analysis of tip and stack (stalk) cells in TgfbrR1 KO mice to determine whether D tip cells are lost in these mutants by snRNAseq. 

The proposed snRNAseq analysis would serve as an independent check on the diving (D) tip vs stalk cell analyses published by Zarkada et al, who analyzed the same VE-cad-CreER;TGFBR1 CKO/- mutant mice, although they refer to the TGFBR1 gene by its alternate name ALK5 [Zarkada et al (2021) Specialized endothelial tip cells guide neuroretina vascularization and blood-retina-barrier formation. Dev Cell 56:2237-2251].  We have not gone in this direction because the question of tip vs. stalk cells and of subtypes of tip cells in WT vs. mutant retinas is beyond our focus on choroidal neovascularization and the role of immune cells and vascular inflammation.  The proposed snRNAseq analysis would also require a major effort since tip cells are rare and must be harvested from large numbers of early postnatal retinas followed by FACS enrichment for vascular endothelial cells.

Suggestions for improving the manuscript:  

(6) The statement that ECs acquire properties of immune cells (Page 2, Line 90) is incorrect. Endothelial cells may acquire characteristics of antigen presenting cells. 

Thank you for that correction.  Based on the review from Amersfoort et al (2022) (Amersfoort J, Eelen G, Carmeliet P. (2022) Immunomodulation by endothelial cells - partnering up with the immune system? Nat Rev Immunol 22:576-588) and the articles cited in it, we have changed the sentence to “Although vascular endothelial cells (ECs) are not generally considered to be part of the immune system, in some locations and under some conditions they acquire properties characteristic of immune cells, including secretion of cytokines, surface display of co-stimulatory or co-inhibitory receptors, and antigen presentation in association with MHC class II proteins (Pober and Sessa, 2014; Amersfoort et al., 2022).”  

(7) The statement in Page 3, Line 100-101 [In CNS ECs, quiescence is maintained in part by the actions of astrocyte-derived Sonic Hedgehog, with the result that few immune cells other than resident microglia are found within the CNS (Alvarez et al., 2011).] is incomplete. Wnt signaling also suppresses the expression of leukocyte adhesion molecules from endothelial cells and therefore helps with immune cell quiescence. 

Thank you for raising that point.  We have expanded that sentence to include Wnt signaling in CNS endothelial cells, as described in the following reference: Lengfeld JE, Lutz SE, Smith JR, Diaconu C, Scott C, Kofman SB, Choi C, Walsh CM, Raine CS, Agalliu I, Agalliu D. (2017) Endothelial Wnt/beta-catenin signaling reduces immune cell infiltration in multiple sclerosis. Proc Natl Acad Sci USA 114:E1168-E1177.

(8) It may be beneficial for the reader to separate the results of the vascular phenotypes related to choroidal neovascularization compared to retinal vascular development. 

Thank you for this suggestion.  The two topics are partly overlapping: choroidal neovascularization is described in Figure 1, and retinal development is described in Figures 1 and 2.  The challenge is that some of same images illustrate both phenotypes as in Figure 1, so the topics cannot be easily separated.

(9) In addition to comparing the phenotypes in Tgfb signaling mutant mice with Wnt signaling and VEGF-A signaling mutants, the authors should compare and contrast their data with those found in Alk5 KO mice, as there are a lot of similarities. 

The reviewer has alerted us to a nomenclature challenge which we will try to resolve in the introduction: Alk5 is just another name for TGFBR1.  The reviewer is correct: there are a lot of similarities between the present study and that of Zarkada et al (2021) because both use the same TGFBR1(=Alk5) CKO mice.

Reviewer #2 (Recommendations for the authors): 

Figure 2 

For 2B, the authors should clarify whether the two regions shown in the Tgfbr1 KO retina (P14) represent central vs. peripheral areas, as phenotype severity varies. 

For 2C, does the uneven biotin accumulation reflect developmental gradients (e.g., central-peripheral maturation timing)? 

Thank you for raising these points.  Regarding Figure 2B, these images are all from the mid-peripheral retina, where the phenotype is moderately severe.  This is now noted in the figure legend.

Regarding Figure 2C, the reviewer is correct that the pattern of Sulfo-NHS-biotin is uneven in VEcadCreER;Tgfbr1CKO/- retinas – it accumulates only in the tufts.  We have expanded Figure 2C to show a comparison between control (i.e.

phenotypically WT), NdpKO, and TGFBR1 endothelial KO retinas, and we have expanded the associated part of the Results section.  In a nutshell, control retinas show little Sulfo-NHS-biotin accumulation in the vasculature or in the parenchyma; NdpKO retinas show Sulfo-NHS-biotin accumulation in the vasculature and in the parenchyma (i.e., the area between the vessels); and VEcadCreER;Tgfbr1CKO/- retinas show Sulfo-NHS-biotin accumulation in the vascular tufts with minimal accumulation in the non-tuft vasculature and minimal leakage into the parenchyma.   The conclusion is that the bulk of the retinal vasculature in TGFBR1 endothelial KO mice is not leaky – very different from the situation with loss of Norrin/Frizzled4 signaling.

Figure 6 

The claim that PECAM1+ rings on veins reflect EC-immune cell binding is uncertain, as PECAM1 is also known to be expressed by immune cells. The complete correlation of PECAM1 and CD45 staining signals suggests that a subset of immune cells upregulates PECAM1. The VEcadCreER;Tgfbr1 flox/-; SUN1:GFP reporter would be helpful to delineate ECimmune cell proximity. Super-resolution imaging with Z-stacks could also resolve spatial relationships (luminal vs. abluminal immune cell adhesion). 

Thank you for this comment.  The reviewer is correct that, at the resolution of these images, we cannot determine whether the PECAM1 immunostaining signal is derived from ECs, from leukocytes, or from both.  This is now stated in the Results section.  The PECAM1-rich endothelial ring structure associated with leukocyte extravasation has been characterized in various publications, for example in (1) Carman CV, Springer TA. (2004) A transmigratory cup in leukocyte diapedesis both through individual vascular endothelial cells and between them. J Cell Biol 167:377-388 and (2) Mamdouh Z, Mikhailov A, Muller WA. (2009) Transcellular migration of leukocytes is mediated by the endothelial lateral border recycling compartment. J Exp Med 206:2795-2808.  The ring structures visualized in Figure 6D by PECAM1 immunostaining conform to the ring structures described in these and other papers.  In showing these structures, our point is simply that they likely represent sites of leukocyte extravasation.  This is now clarified in the text.  We have also added some additional references on leukocyte extravasation and the ring structures.

Figure 7 

A time-course analysis of ICAM1 would strengthen the mechanistic model. Does ICAM1 upregulation precede immune infiltration (supporting inflammation as the primary defect)? Given that immune cells appear by P14 (per snRNA-seq), is ICAM1 elevated earlier? 

This is an interesting idea, but based on what is known about leukocyte adhesion and extravasation we predict that there will not be a clean temporal separation between ICAM1 induction and leukocyte adhesion/infiltration.  That is, if the proinflammatory state causes an increase in the number of leukocytes, then as ICAM1 levels increase, leukocyte adhesion would also increase.  Similarly, if the presence of leukocytes increases the pro-inflammatory state, then as the number of leukocytes increases, the levels of ICAM1 would be predicted to increase.  Thus, we think that a time course analysis is unlikely to provide a definitive conclusion.

Figure 8-SF1 

In brain slices, a transient pan-IgG accumulation suggests a self-resolving defect in the BBB. However, this BBB impairment appears to be spatiotemporally distinct from ICAM1 upregulation. ICAM1 staining is restricted to the lesion site, aligning with immune cell-driven inflammation. 

Thank you for raising these points.  The reviewer is correct that these observations don’t fit together in a clear way.  There does not appear to be a general increase in brain vascular permeability in VE-cad-CreER;TGFBR1 CKO/- mice, as shown by sulfo-NHS-biotin.  However, there is a large and transient increase in IgG in the brain parenchyma, suggestive of a general vascular alteration, and – as the reviewer correctly notes – it is not accompanied by a generalized increase in ICAM1 vascular immunostaining.  At this point, we don’t have any real insight into the mechanistic basis of the transient IgG increase.

Thank you for handling this manuscript.

eLife Assessment

This cleverly designed and potentially important work supports our understanding regarding how and whether social behaviours promoting egalitarianism can be learned, even when implementing these norms entails a cost for oneself. However, the evidence supporting the major claims is currently incomplete, with the major limitation being whether Ps truly learn egalitarianism from a teacher or instead exhibit reduced guilt across time that is reduced when observing others behaving more selfishly. With a strengthening of the supporting evidence, this work will be of interest to a wide range of fields, including cognitive psychology/neuroscience, neuroeconomics, and social psychology, as well as policy making.

Reviewer #1 (Public review):

Summary:

Zhang et al. addressed the question of whether advantageous and disadvantageous inequality aversion can be vicariously learned and generalized. Using an adapted version of the ultimatum game (UG), in three phases, participants first gave their own preference (baseline phase), then interacted with a "teacher" to learn their preference (learning phase), and finally were tested again on their own (transfer phase). The key measure is whether participants exhibited similar choice preference (i.e., rejection rate and fairness rating) influenced by the learning phase, by contrasting their transfer phase and baseline phase. Through a series of statistical modeling and computational modeling, the authors reported that both advantageous and disadvantageous inequality aversion can indeed be learned (Study 1), and even be generalised (Study 2).

Strengths:

This study is very interesting, that directly adapted the lab's previous work on the observational learning effect on disadvantageous inequality aversion, to test both advantageous and disadvantageous inequality aversion in the current study. Social transmission of action, emotion, and attitude have started to be looked at recently, hence this research is timely. The use of computational modeling is mostly appropriate and motivated. Study 2 that examined the vicarious inequality aversion on conditions where feedback was never provided is interesting and important to strengthen the reported effects. Both studies have proper justifications to determine the sample size.

Weaknesses:

Despite the strengths, a few conceptual aspects and analytical decisions have to be explained, justified, or clarified.

INTRODUCTION/CONCEPTUALIZATION

(1) Two terms seem to be interchangeable, which should not, in this work: vicarious/observational learning vs preference learning. For vicarious learning, individuals observe others' actions (and optionally also the corresponding consequence resulted directly by their own actions), whereas, for preference learning, individuals predict, or act on behalf of, the others' actions, and then receive feedback if that prediction is correct or not. For the current work, it seems that the experiment is more about preference learning and prediction, and less so about vicarious learning. But the intro and set are heavily around vicarious learning, and late the use of vicarious learning and preference learning is rather mixed in the text. I think either tone down the focus on vicarious learning, or discuss how they are different. Some of the references here may be helpful: Charpentier et al., Neuron, 2020; Olsson et al., Nature Reviews Neuroscience, 2020; Zhang & Glascher, Science Advances, 2020

EXPERIMENTAL DESIGN

(2) For each offer type, the experiment "added a uniformly distributed noise in the range of (-10 ,10)". I wonder how this looks like? With only integers such as 25:75, or even with decimal points? More importantly, is it possible to have either 70:30 or 90:10 option, after adding the noise, to have generated an 80:20 split shown to the participants? If so, for the analyses later, when participants saw the 80:20 split, which condition did this trial belong to? 70:30 or 90:10? And is such noise added only to the learning phase, or also to the baseline/transfer phases? This requires some clarification.

(3) For the offer conditions (90:10, 70:30, 50:50, 30:70, 10:90) - are they randomized? If so, how is it done? Is it randomized within each participants, and/or also across participants (such that each participant experienced different trial sequences)? This is important, as the order especially for the leanring phase can largely impact on the preference learning of the participants.

STATISTICAL ANALYSIS & COMPUTATIONAL MODELING

(4) In Study 1 DI offer types (90:10, 70:30), the rejection rate for DI-AI averse looks consistently higher than that for DI averse (ie, blue line is above the yellow line). Is this significant? If so, how come? Since this is a between-subject design, I would not anticipate such a result (especially for the baseline). Also, for the LME results (eg, Table S3), only interactions were reported but not the main results.

(5) I do not particularly find this analysis appealing: "we examined whether participants' changes in rejection rates between Transfer and Baseline, could be explained by the degree to which they vicariously learned, defined as the change in punishment rates between the first and last 5 trials of the Learning phase." Naturally, participants' behavior in the first 5 trials in the learning phase will be similar to those in the baseline; and their behavior in the last 5 trials in the learning phase would echo those at the transfer phase. I think it would be stronger to link the preference learning results to the chance between baseline and transfer phase, eg, by looking at the difference between alpha (beta) at the end of the learning phase and the initial alpha (beta).

(6) I wonder if data from the baseline and transfer phases can also be modeled, using a simple Fehr-Schimdt model? This way, the change in alpha/beta can also be examined between the baseline and transfer phase.

(7) I quite liked Study 2 that tests the generalization effect, and I expected to see an adapted computational modeling to directly reflect this idea. Indeed, the authors wrote "[...] given that this model [...] assumes the sort of generalization of preferences between offer types [...]". But where exactly did the preference learning model assumed the generalization? In the methods, the modeling seems to be only about Study 1; did the authors advise their model to accommodate Study 2? The authors also ran simulation for the learning phase in Study 2 (Figure 6), and how did the preference updated (if at all) for offers (90:10 and 10:90) where feedback was not given? Extending/Unpacking the computational modeling results for Study2 will be very helpful for the paper.

Comments on revisions:

I kept my original public review, so that future readers can see the progress and development of the manuscript.

The authors have largely addressed my original questions/concerns, and I have two outstanding comments.

(a) Related to my original comment #6, where I suggested to apply the F-S model also to the baseline and transfer phase. The authors were inclined not to do it, but in fact later in comment #7 and in the manuscript they opted to use a more complex F-S-based model to their learning phase. I agree that the rejection rate is indeed a clear indication, but for completeness, it'd be more consistent and compelling if the paper follows a model-free (model-agnostic) and model-based approach in all phases of the experiment.

(b) Related to my original comment #4, I appreciate that the authors have provided more details of their LMM models. But I don't think it is accurate regardless. First, all offer levels (50:50, 30:70, 10:90), should not be coded as pure categorical levels. In fact, they have an ordinal meaning, a single ordinal predictor with three levels should be used. This also avoids the excessive number of interactions the authors have pointed out.

Second, running a model with only interactions without main effects is flawed. All textbooks on stats emphasize that without the presence of the main effects, the interpretation of interaction only is biased.

So these LMMs needs to be revised before the manuscript eventually gets to a version of record.

Reviewer #2 (Public review):

Summary:

This study investigates whether individuals can learn to adopt egalitarian norms that incur a personal monetary cost, such as rejecting offers that benefit them more than the giver (advantageous inequitable offers). While these behaviors are uncommon, two experiments aim to demonstrate that individuals can learn to reject such offers by observing a "teacher" who follows these norms. The authors use computational modelling to argue that learners adopt these norms through a sophisticated process, inferring the latent structure of the teacher's preferences, akin to theory of mind.

Strengths:

This paper is well-written and tackles an important topic relevant to social norms, morality, and justice. The findings are promising (though further control conditions are necessary to support the conclusions). The study is well-situated in the literature, with a clever experimental design and a computational approach that may offer insights into latent cognitive processes. In the revision, the authors clarified some questions related to the initial submission.

Weaknesses:

Despite these strengths, I remain unconvinced that the current evidence supports the paper's central claims. Below, I outline several issues that, in my view, limit the strength of the conclusions.

(1) Experimental Design and Missing Control Condition:

The authors set out to test whether observing a "teacher" who is averse to advantageous inequity (Adv-I) will affect observers' own rejection of Adv-I offers. However, I think the design of the task lacks an important control condition needed to address this question. At present, participants are assigned to one of two teachers: DIS or DIS+ADV. Behavioral differences between these groups can only reveal relative differences in influence; they cannot establish whether (and how) either teacher independently affects participants' own behavior. For example, a significant difference between conditions can emerge even if participants are only affected by the DIS teacher and are not affected at all by the DIS+ADV teacher. What is crucially missing here is a no-teacher control condition, which can then be compared with each teacher condition separately. This control condition would also control for pure temporal effects unrelated to teacher influence (e.g., increasing Adv-I rejections due to guilt build-up).

While this criticism applies to both experiments, it is especially apparent in Experiment 2. As shown in Figure 4, the interaction for 10:90 offers reflects a decrease in rejection rates following the DIS teacher, with no significant change following the DIS+ADV teacher. Ignoring temporal effects, this pattern suggests that participants may be learning NOT to reject from the DIS teacher, rather than learning to reject from the DIS+ADV teacher. On this basis, I do not see convincing evidence that participants' own choices were shaped by observing Adv-I rejections.

In the Discussion, the authors write that "We found that participants' own Adv-I-averse preferences shifted towards the preferences of the Teacher they just observed, and the strength of these contagion effects related to the degree of behavior change participants exhibited on behalf of the Teachers, suggesting that they internalized, at least somewhat, these inequity preferences." However, there is no evidence that directly links the degree of behaviour change (on the teacher's behalf) to contagion effects (own behavioural change). I think there was a relevant analysis in the original version, but it was removed from the current version.

(2) Modelling Efforts: The modelling approach is underdeveloped. The identification of the "best model" lacks transparency, as no model-recovery results are provided. Additionally, behavioural fits for the losing models are not shown, leaving readers in the dark about where these models fail. Readers would benefit from seeing qualitative/behavioural patterns that favour the winning model. Moreover, the reinforcement learning (RL) models used are overly simplistic, treating actions as independent when they are likely inversely related. For example, the feedback that the teacher would have rejected an offer provides evidence that rejection is "correct" but also that acceptance is "an error," and the latter is not incorporated into the modelling. In other words, offers are modelled as two-armed bandits (where separate values are learned for reject and accept actions), but the situation is effectively a one-armed bandit (if one action is correct, the other is mistaken). It is unclear to what extent this limitation affects the current RL formulations. Can the authors justify/explain their reasoning for including these specific variants? The manuscript only states Q-values for reject actions, but what are the Q-values for accept actions? This is unclear.

In Experiment 2, only the preferred model is capable of generalization, so it is perhaps unsurprising that this model "wins." However, this does not strongly support the proposed learning mechanism, lacking a comparison with simpler generalizing mechanisms (see following comments).

(3) Conceptual Leap in Modelling Interpretation: The distinction between simple RL models and preference-inference models seems to hinge on the ability to generalize learning from one offer to another. Whereas in the RL models, learning occurs independently for each offer (hence no cross-offer generalization), preference inference allows for generalization between different offers. However, the paper does not explore "model-free" RL models that allow generalization based on the similarity of features of the offers (e.g., payment for the receiver, payment for the offer-giver, who benefits more). Such models are more parsimonious and could explain the results without invoking a theory of mind or any modelling of the teacher. In such model versions, a learner acquires a functional form that allows prediction of the teacher's feedback based on offer features (e.g., linear or quadratic weighting). Because feedback for an offer modulates the parameters of this function (feature weights), generalization occurs without necessarily evoking any sophisticated model of the other person. This leaves open the possibility that RL models could perform just as well or even outperform the preference learning model, casting doubt on the authors' conclusions.

Of note: even the behaviourists knew that when Little Albert was taught to fear rats, this fear generalized to rabbits. This could occur simply because rabbits are somewhat similar to rats. But this doesn't mean Little Albert had a sophisticated model of animals that he used to infer how they behave.

In their rebuttal letter, the authors acknowledge these possibilities, but the manuscript still does not explore or address alternative mechanisms.

(4) Limitations of the Preference-Inference Model: The preference-inference model struggles to capture key aspects of the data, such as the increase in rejection rates for 70:30 DI offers during the learning phase (e.g., Fig. 3A, AI+DI blue group). This is puzzling. Thinking about this, I realized the model makes quite strong, unintuitive predictions which are not examined. For example, if a subject begins the learning phase rejecting the 70:30 offer more than 50% of the time (meaning the starting guilt parameter is higher than 1.5), then, over learning, the tendency to reject will decrease to below 50% (the guilt parameter will be pulled down below 1.5). This is despite the fact that the teacher rejects 75% of the offers. In other words, as learning continues, learners will diverge from the teacher. On the other hand, if a participant begins learning by tending to accept this offer (guilt < 1.5), then during learning, they can increase their rejection rate but never above 50%. Thus, one can never fully converge on the teacher. I think this relates to the model's failure in accounting for the pattern mentioned above. I wonder if individuals actually abide by these strict predictions. In any case, these issues raise questions about the validity of the model as a representation of how individuals learn to align with a teacher's preferences (given that the model doesn't really allow for such an alignment).

In their rebuttal letter, the authors acknowledged these anomalies and stated that they were able to build a better model (where anomalies are mitigated, though not fully eliminated). But they still report the current model and do not develop/discuss alternatives. A more principled model may be a Bayesian model where participants learn a belief distribution (rather than point estimates) regarding the teacher's parameters.

(5) Statistical Analysis: The authors state in their rebuttal letter that they used the most flexible random effect structure in mixed-effects models. But this seems not to be the case in the model reported in Table SI3 (the very same model was used for other analyses too). Indeed, here it seems only intercepts are random effects. This left me confused about which models were used.

Author response:

The following is the authors’ response to the original reviews

Reviewer #1 (Public review):

Summary:

Zhang et al. addressed the question of whether advantageous and disadvantageous inequality aversion can be vicariously learned and generalized. Using an adapted version of the ultimatum game (UG), in three phases, participants first gave their own preference (baseline phase), then interacted with a "teacher" to learn their preference (learning phase), and finally were tested again on their own (transfer phase). The key measure is whether participants exhibited similar choice preferences (i.e., rejection rate and fairness rating) influenced by the learning phase, by contrasting their transfer phase and baseline phase. Through a series of statistical modeling and computational modeling, the authors reported that both advantageous and disadvantageous inequality aversion can indeed be learned (Study 1), and even be generalised (Study 2).

Strengths:

This study is very interesting, it directly adapted the lab's previous work on the observational learning effect on disadvantageous inequality aversion, to test both advantageous and disadvantageous inequality aversion in the current study. Social transmission of action, emotion, and attitude have started to be looked at recently, hence this research is timely. The use of computational modeling is mostly appropriate and motivated. Study 2, which examined the vicarious inequality aversion in conditions where feedback was never provided, is interesting and important to strengthen the reported effects. Both studies have proper justifications to determine the sample size.

Weaknesses:

Despite the strengths, a few conceptual aspects and analytical decisions have to be explained, justified, or clarified.

INTRODUCTION/CONCEPTUALIZATION

(1) Two terms seem to be interchangeable, which should not, in this work: vicarious/observational learning vs preference learning. For vicarious learning, individuals observe others' actions (and optionally also the corresponding consequence resulting directly from their own actions), whereas, for preference learning, individuals predict, or act on behalf of, the others' actions, and then receive feedback if that prediction is correct or not. For the current work, it seems that the experiment is more about preference learning and prediction, and less so about vicarious learning. The intro and set are heavily around vicarious learning, and later the use of vicarious learning and preference learning is rather mixed in the text. I think either tone down the focus on vicarious learning, or discuss how they are different. Some of the references here may be helpful: (Charpentier et al., Neuron, 2020; Olsson et al., Nature Reviews Neuroscience, 2020; Zhang & Glascher, Science Advances, 2020)

We are appreciative of the Reviewer for raising this question and providing the reference. In response to this comment we have elected to avoid, in most cases, use of the term ‘vicarious’ and instead focus the paper on learning of others’ preferences (without specific commitment to various/observational learning per se). These changes are reflected throughout all sections of the revised manuscript, and in the revised title. We believe this simplified terminology has improved the clarity of our contribution.

EXPERIMENTAL DESIGN

(2) For each offer type, the experiment "added a uniformly distributed noise in the range of (-10 ,10)". I wonder what this looks like? With only integers such as 25:75, or even with decimal points? More importantly, is it possible to have either 70:30 or 90:10 option, after adding the noise, to have generated an 80:20 split shown to the participants? If so, for the analyses later, when participants saw the 80:20 split, which condition did this trial belong to? 70:30 or 90:10? And is such noise added only to the learning phase, or also to the baseline/transfer phases? This requires some clarification.

We thank the Reviewer for pointing this out. The uniformly distributed noise was added to all three phases to make the proposers’ behavior more realistic. This added noise was rounded to integer numbers, constrained from -9 to 9, which means in both 70:30 and 90:10 offer types, an 80:20 split could not occur. We have made this feature of our design clear in the Method section Line 524 ~ 528:

“In all task phases, we added uniformly distributed noise to each trial’s offer (ranging from -9 to 9, inclusive, rounding to the nearest integer) such that the random amount added (or subtracted) from the Proposer’s share was subtracted (or added) to the Receiver’s share. We adopted this manipulation to make the proposers’ behavior appear more realistic. The orders of offers participants experienced were fully randomized within each experiment phase. ”

(3) For the offer conditions (90:10, 70:30, 50:50, 30:70, 10:90) - are they randomized? If so, how is it done? Is it randomized within each participant, and/or also across participants (such that each participant experienced different trial sequences)? This is important, as the order especially for the learning phase can largely impact the preference learning of the participants.

We agree with the Reviewer the order in which offers are experienced could be very important. The order of the conditions was randomized independently for each participant (i.e. each participant experienced different trial sequences). We made this point clear in the Methods part. Line 527 ~ 528:

“The orders of offers participants experienced were fully randomized within each experiment phase.”

STATISTICAL ANALYSIS & COMPUTATIONAL MODELING

(4) In Study 1 DI offer types (90:10, 70:30), the rejection rate for DI-AI averse looks consistently higher than that for DI averse (ie, the blue line is above the yellow line). Is this significant? If so, how come? Since this is a between-subject design, I would not anticipate such a result (especially for the baseline). Also, for the LME results (eg, Table S3), only interactions were reported but not the main results.

We thank the Reviewer for pointing out this feature of the results. Prompted by this comment, we compared the baseline rejection rates between two conditions for these two offer types, finding in Experiment 1 that rejection rates in the DI-AI-averse condition were significantly higher than in the DI-averse condition (DI-AI-averse vs. DI-averse; Offer 90:10, β = 0.13, p < 0.001, Offer 70:30, β = 0.09, p < 0.034). We agree with the Reviewer that there should, in principle, be no difference between the experiences of participants in these two conditions is identical in the Baseline phase. However, we did not observe these difference in baseline preferences in Experiment 2 (DI-AI-averse vs. DI-averse; Offer 90:10, β = 0.07, p < 0.100, Offer 70:30, β = 0.05, p < 0.193). On the basis of the inconsistency of this effect across studies we believe this is a spurious difference in preferences stemming from chance.

Regarding the LME results, the reason why only interaction terms are reported is due to the specification of the model and the rationale for testing.

Taking the model reported in Table S3 as an example—a logistic model which examines Baseline phase rejection rates as a function of offer level and condition—the between-subject conditions (DI-averse and DI-AI-averse) are represented by dummy-coded variables. Similarly, offer types were also dummy-coded, such that each of the five columns (90:10, 70:30, 50:50, 30:70, and 10:90) correspond corresponded to a particular offer type. This model specification yields ten interaction terms (i.e., fixed effects) of interest—for example, the “DI-averse × Offer 90:10” indicates baseline rejection rates for 90:10 offers in DI-averse condition. Thus, to compare rejection rates across specific offer types, we estimate and report linear contrasts between these resultant terms. We have clarified the nature of these reported tests in our revised Results—for example, line189-190: “linear contrasts; e.g. 90:10 vs 10:90, all Ps<0.001, see Table S3 for logistic regression coefficients for rejection rates).

Also in response to this comment that and a recommendation from Reviewer 2 (see below), we have revised our supplementary materials to make each model specification clearer as SI line 25:

“RejectionRate ~ 0 + (Disl + Advl):(Offer10 + Offer30 + Offer50 + Offer70 + Offer90) + (1|Subject)”

(5) I do not particularly find this analysis appealing: "we examined whether participants' changes in rejection rates between Transfer and Baseline, could be explained by the degree to which they vicariously learned, defined as the change in punishment rates between the first and last 5 trials of the Learning phase." Naturally, the participants' behavior in the first 5 trials in the learning phase will be similar to those in the baseline; and their behavior in the last 5 trials in the learning phase would echo those at the transfer phase. I think it would be stronger to link the preference learning results to the change between the baseline and transfer phase, eg, by looking at the difference between alpha (beta) at the end of the learning phase and the initial alpha (beta).

Thanks for pointing this out. Also, considering the comments from Reviewer 2 concerning the interpretation of this analysis, we have elected to remove this result from our revision.

(6) I wonder if data from the baseline and transfer phases can also be modeled, using a simple Fehr-Schimdt model. This way, the change in alpha/beta can also be examined between the baseline and transfer phase.

We agree with the Reviewer that a simplified F-S model could be used, in principle, to characterize Baseline and Transfer phase behavior, but it is our view that the rejection rates provide readers with the clearest (and simplest) picture of how participants are responding to inequity. Put another way, we believe that the added complexity of using (and explaining) a new model to characterize simple, steady-state choice behavior (within these phases) would not be justified or add appreciable insights about participants’ behavior.

(7) I quite liked Study 2 which tests the generalization effect, and I expected to see an adapted computational modeling to directly reflect this idea. Indeed, the authors wrote, "[...] given that this model [...] assumes the sort of generalization of preferences between offer types [...]". But where exactly did the preference learning model assume the generalization? In the methods, the modeling seems to be only about Study 1; did the authors advise their model to accommodate Study 2? The authors also ran simulation for the learning phase in Study 2 (Figure 6), and how did the preference update (if at all) for offers (90:10 and 10:90) where feedback was not given? Extending/Unpacking the computational modeling results for Study 2 will be very helpful for the paper.

We are appreciative of the Reviewer’s positive impression of Experiment 2. Upon reflection, we realize that our original submission was not clear about the modeling done in Experiment 2, and we should clarify here that we did also fit the Preference Inference model to this dataset. As in Experiment 1, this model assumes that the participants have a representation of the teacher’s preference as a Fehr-Schmidt form utility function and infer the Teacher’s Envy and Guilt parameters through learning. The model indicates that, on the basis of experience with the Teacher’s preferences on moderately unfair offers (i.e., offer 70:30 and offer 30:70), participants can successfully infer these guess of these two parameters, and in turn, compute Fehr-Schmidt utility to guide their decisions in the extreme unfair offers (i.e., offer 90:10 and offer 10:90).

In response to this comment, we have made this clearer in our Results (Line 377-382):

“Finally, following Experiment 1, we fit a series of computational models of Learning phase choice behavior, comparing the goodness-of-fit of the four best-fitting models from Experiment 1 (see Methods). As before, we found that the Preference Inference model provided the best fit of participants’ Learning Phase behavior (Figure S1a, Table S12). Given that this model is able to infer the Teacher’s underlying inequity-averse preferences (rather than learns offer-specific rejection preferences), it is unsurprising that this model best describes the generalization behavior observed in Experiment 2.”

and in our revised Methods (Line 551-553)

“We considered 6 computational models of Learning Phase choice behavior, which we fit to individual participants’ observed sequences of choices, in both Experiments 1 and 2, via Maximum Likelihood Estimation”

Reviewer #2 (Public review):

Summary:

This study investigates whether individuals can learn to adopt egalitarian norms that incur a personal monetary cost, such as rejecting offers that benefit them more than the giver (advantageous inequitable offers). While these behaviors are uncommon, two experiments demonstrate that individuals can learn to reject such offers through vicarious learning - by observing and acting in line with a "teacher" who follows these norms. The authors use computational modelling to argue that learners adopt these norms through a sophisticated process, inferring the latent structure of the teacher's preferences, akin to theory of mind.

Strengths:

This paper is well-written and tackles a critical topic relevant to social norms, morality, and justice. The findings, which show that individuals can adopt just and fair norms even at a personal cost, are promising. The study is well-situated in the literature, with clever experimental design and a computational approach that may offer insights into latent cognitive processes. Findings have potential implications for policymakers.

Weaknesses:

Note: in the text below, the "teacher" will refer to the agent from which a participant presumably receives feedback during the learning phase.

(1) Focus on Disadvantageous Inequity (DI): A significant portion of the paper focuses on responses to Disadvantageous Inequitable (DI) offers, which is confusing given the study's primary aim is to examine learning in response to Advantageous Inequitable (AI) offers. The inclusion of DI offers is not well-justified and distracts from the main focus. Furthermore, the experimental design seems, in principle, inadequate to test for the learning effects of DI offers. Because both teaching regimes considered were identical for DI offers the paradigm lacks a control condition to test for learning effects related to these offers. I can't see how an increase in rejection of DI offers (e.g., between baseline and generalization) can be interpreted as speaking to learning. There are various other potential reasons for an increase in rejection of DI offers even if individuals learn nothing from learning (e.g. if envy builds up during the experiment as one encounters more instances of disadvantageous fairness).

We are appreciative of the Reviewer’s insight here and for the opportunity to clarify our experimental logic. We included DI offers in order to 1) expose participants to the full spectrum of offer types, and avoid focusing participants exclusively upon AI offers, which might result in a demand characteristic and 2) to afford exploration of how learning dynamics might differ in DI context s—which was, to some extent, examined in our previous study (FeldmanHall, Otto, & Phelps, 2018)—versus AI contexts. Furthermore, as this work builds critically on our previous study, we reasoned that replicating these original findings (in the DI context) would be important for demonstrating the generality of the learning effects in the DI context across experimental settings. We now remark on this point in our revised Introduction Line 129 ~132:

“In addition, to mechanistically probe how punitive preferences are acquired in Adv-I and Dis-I contexts—in turn, assessing the replicability of our earlier study investigating punitive preference acquisition in the Dis context—we also characterize trial-by-trial acquisition of punitive behavior with computational models of choice.”

(2) Statistical Analysis: The analysis of the learning effects of AI offers is not fully convincing. The authors analyse changes in rejection rates within each learning condition rather than directly comparing the two. Finding a significant effect in one condition but not the other does not demonstrate that the learning regime is driving the effect. A direct comparison between conditions is necessary for establishing that there is a causal role for the learning regime.

We agree with the Reviewer and upon reflection, believe that direct comparisons between conditions would be helpful to support the claim that the different learning conditions are responsible for the observed learning effects. In brief, these specific tests buttress the idea that exposure to AI-averse preferences result in increases in AI punishment rates in the Transfer phase (over and above the rates observed for participants who were only exposed to DI-averse preferences).

Accordingly, our revision now reports statistics concerning the differences between conditions for AI offers in Experiment 1 (Line 198~ 207):

“Importantly, when comparing these changes between the two learning conditions, we observed significant differences in rejection rates for Adv-I offers: compared to exposure to a Teacher who rejected only Dis-I offers, participants exposed to a Teacher who rejected both Dis-I and Adv-I offers were more likely to reject Adv-I offers and rated these offers more unfair. This difference between conditions was evident in both 30:70 offers (Rejection rates: β(SE) = 0.10(0.04), p = 0.013; Fairness ratings: β(SE) = -0.86(0.17), p < 0.001) and 10:90 offers (Rejection rates: β(SE) = 0.15(0.04), p < 0.001, Fairness ratings: β(SE) = -1.04(0.17), p < 0.001). As a control, we also compared rejection rates and fairness rating changes between conditions in Dis-I offers (90:10 and 30:70) and Fair offers (i.e., 50:50) but observed no significant difference (all ps > 0.217), suggesting that observing an Adv-I-averse Teacher’s preferences did not influence participants’ behavior in response to Dis-I offers.”

Line 222 ~ 230:

“A mixed-effects logistic regression revealed a significant larger (positive) effect of trial number on rejection rates of Adv-I offers for the Adv-Dis-I-Averse condition compared to the Dis-I-Averse condition. This relative rejection rate increase was evident both in 30:70 offers (Table S7; β(SE) = -0.77(0.24), p < 0.001) and in 10:90 offers (β(SE) = -1.10(0.33), p < 0.001). In contrast, comparing Dis-I and Fairness offers when the Teacher showed the same tendency to reject, we found no significant difference between the two conditions (90:10 splits: β(SE)=-0.48(0.21),p=0.593;70:30 splits: β(SE)=-0.01(0.14),p=0.150; 50:50 splits: β(SE)=-0.00(0.21),p=0.086). In other words, participants by and large appeared to adjust their rejection choices in accordance with the Teacher’s feedback in an incremental fashion.”

And in Experiment 2 Line 333 ~ 345:

“Similar to what we observed in Experiment 1 (Figure 4a), Compared to the participants in the Dis-I-Averse Condition, participants in the Adv-I-Averse Condition increased their rates of rejection of extreme Adv-I offerers (i.e., 10:90) in the Transfer Phase, relative to the Baseline phase (β(SE) = -0.12(0.04), p < 0.004; Table S9), suggesting that participants’ learned (and adopted) Adv-I-averse preferences, generalized from one specific offer type (30:70) to an offer types for which they received no Teacher feedback (10:90). Examining extreme Dis-I offers where the Teacher exhibited identical preferences across the two learning conditions, we found no difference in the Changes of Rejection Rates from Baseline to Transfer phase between conditions (β(SE) = -0.05(0.04), p < 0.259). Mirroring the observed rejection rates (Figure 4b), relative to the Dis-I-Averse Condition, participants’ fairness ratings for extreme Adv-I offers increased more from the Baseline to Transfer phase in the Adv-Dis-I-Averse Condition than in the Dis-I-Averse condition (β(SE) = -0.97(0.18), p < 0.001), but, importantly, changes in fairness ratings for extreme Dis-I offers did not differ significantly between learning conditions (β(SE) = -0.06(0.18), p < 0.723)”

Line 361 ~ 368:

“Examining the time course of rejection rates in Adv-I-contexts during the Learning phase (Figure 5) revealed that participants learned over time to punish mildly unfair 30:70 offers, and these punishment preferences generalized to more extreme offers (10:90). Specifically, compared to the Dis-I-Averse Condition, in the Adv-Dis-I-Averse condition we observed a significant larger trend of increase in rejections rates for 10:90 (Adv-I) offers (Figure 5, β(SE) = -0.81(0.26), p < 0.002 mixed-effects logistic regression, see Table S10). Again, when comparing the rejection rate increase in the extremely Dis-I offers (90:10), we didn’t find significant difference between conditions (β(SE) = -0.25(0.19), p < 0.707).”

(3) Correlation Between Learning and Contagion Effects:

The authors argue that correlations between learning effects (changes in rejection rates during the learning phase) and contagion effects (changes between the generalization and baseline phases) support the idea that individuals who are better aligning their preferences with the teacher also give more consideration to the teacher's preferences later during generalization phase. This interpretation is not convincing. Such correlations could emerge even in the absence of learning, driven by temporal trends like increasing guilt or envy (or even by slow temporal fluctuations in these processes) on behalf of self or others. The reason is that the baseline phase is temporally closer to the beginning of the learning phase whereas the generalization phase is temporally closer to the end of the learning phase. Additionally, the interpretation of these effects seems flawed, as changes in rejection rates do not necessarily indicate closer alignment with the teacher's preferences. For example, if the teacher rejects an offer 75% of the time then a positive 5% learning effect may imply better matching the teacher if it reflects an increase in rejection rate from 65% to 70%, but it implies divergence from the teacher if it reflects an increase from 85% to 90%. For similar reasons, it is not clear that the contagion effects reflect how much a teacher's preferences are taken into account during generalization.

This comment is very similar to a previous comment made by Reviewer 1, who also called into question the interpretability of these correlations. In response to both of these comments we have elected to remove these analyses from our revision.

(4) Modeling Efforts: The modelling approach is underdeveloped. The identification of the "best model" lacks transparency, as no model-recovery results are provided, and fits for the losing models are not shown, leaving readers in the dark about where these models fail. Moreover, the reinforcement learning (RL) models used are overly simplistic, treating actions as independent when they are likely inversely related (for example, the feedback that the teacher would have rejected an offer provides feedback that rejection is "correct" but also that acceptance is "an error", and the later is not incorporated into the modelling). It is unclear if and to what extent this limits current RL formulations. There are also potentially important missing details about the models. Can the authors justify/explain the reasoning behind including these variants they consider? What are the initial Q-values? If these are not free parameters what are their values?

We are appreciative of the Reviewer for identifying these potentially unaddressed questions.

The RL models we consider in the present study are naïve models which, in our previous study (FeldmanHall, Otto, & Phelps, 2018), we found to capture important aspects of learning. While simplistic, we believed these models serve as a reasonable baseline for evaluating more complex models, such as the Preference Inference model. We have made this point more explicit in our revised Introduction, Line 129 ~ 132:

“In addition, to mechanistically probe how punitive preferences may be acquired in Adv-I and Dis-I contexts—in turn, assessing the replicability of our earlier study investigating punitive preference acquisition in the Dis-I context—we also characterize trial-by-trial acquisition of punitive behavior with computational models of choice.”

Again, following from our previous modeling of observational learning (FeldmanHall et al., 2018), we believe that the feedback the Teacher provides here is ideally suited to the RL formalism. In particular, when the teacher indicates that the participant’s choice is what they would have preferred, the model receives a reward of ‘1’ (e.g., the participant rejects and the Teacher indicates they would preferred rejection, resulting in a positive prediction error) otherwise, the model receives a reward of ‘0’ (e.g., the participant accepts and the Teacher indicates they would preferred rejection, resulting in a negative prediction error), indicating that the participant did not choose in accordance with the Teacher’s preferences. Through an error driven learning process, these models provide a naïve way of learning to act in accordance with the Teacher’s preferences.

Regarding the requested model details: When treating the initial values as free parameters (model 5), we set Q(reject, offertype) as free values in [0,1] and Q(accept,offertype) as 0.5. This setting can capture participants' initial tendency to reject or accept offers from this offer type. When the initial values are fixed, for all offer types we set Q(reject, offertype) = Q(accept,offertype) = 0.5. In practice, when the initial values are fixed, setting them to 0.5 or 0 doesn’t make much difference. We have clarified these points in our revised Methods, Line 275 ~ 576:

“We kept the initial values fixed in this model, that is Q₀(reject,offertype) =0.5, (offertype ∈ 90:10, 70:30, 50:50, 30:70, 10:90)”

And Line 582 ~ 584:

“Formally, this model treats Q₀(reject,offertype) =0.5, (offertype ∈ 90:10, 70:30, 50:50, 30:70, 10:90) as free parameters with values between 0 and 1.”

(5) Conceptual Leap in Modeling Interpretation: The distinction between simple RL models and preference-inference models seems to hinge on the ability to generalize learning from one offer to another. Whereas in the RL models learning occurs independently for each offer (hence to cross-offer generalization), preference inference allows for generalization between different offers. However, the paper does not explore RL models that allow generalization based on the similarity of features of the offers (e.g., payment for the receiver, payment for the offer-giver, who benefits more). Such models are more parsimonious and could explain the results without invoking a theory of mind or any modelling of the teacher. In such model versions, a learner learns a functional form that allows to predict the teacher's feedback based on said offer features (e.g., linear or quadratic form). Because feedback for an offer modulates the parameters of this function (feature weights) generalization occurs without necessarily evoking any sophisticated model of the other person. This leaves open the possibility that RL models could perform just as well or even show superiority over the preference learning model, casting doubt on the authors' conclusions. Of note: even the behaviourists knew that as Little Albert was taught to fear rats, this fear generalized to rabbits. This could occur simply because rabbits are somewhat similar to rats. But this doesn't mean little Alfred had a sophisticated model of animals he used to infer how they behave.

We are appreciative of the Reviewer for their suggestion of an alternative explanation for the observed generalization effects. Our understanding of the suggestion, put simply, put simply, is that an RL model could capture the observed generalization effects if the model were to learn and update a functional form of the Teacher’s rejection preferences using an RL-like algorithm. This idea is similar, conceptually to our account of preference learning whereby the learner has a representation of the teacher’s preferences. In our experiment the offer is in the range of [0-100], the crux of this idea is why the participants should take the functional form (either v-shaped or quadratic) with the minimum at 50. This is important because, at the beginning of the learning phase, the rejection rates are already v-shaped with 50 as its minimum. The participants do not need to adjust the minimum of this functional form. Thus, if we assume that the participants represent the teacher’s rejection rate as a v-shape function with a minimum at [50,50], then this very likely implies that the participants have a representation that the teacher has a preference for fairness. Above all, we agree that with suitable setup of the functional form, one could implement an RL model to capture the generalization effects, without presupposing an internal “model” of the teacher’s preferences.

However, there is another way of modeling the generalization effect by truly “model-free” similarity-based Reinforcement learning. In this approach, we do not assume any particular functional form of the teacher’s preferences, but rather, assumes that experience acquired in one offer type can be generalized to offers that are close (i.e., similar) to the original offer. Accordingly, we implement this idea using a simple RL model in which the action values for each offer type is updated by a learning rate that is scaled by the distance between that offer and the experienced offer (i.e., the offer that generated the prediction error). This learning rate is governed by a Gaussian distribution, similar to the case in the Gaussian process regression (cf. Chulz, Speekenbrink, & Krause, 2018). The initial value of the ‘Reject’ action, for each offer , is set to a free parameter between 0 and 1, and the initial value for the 'Accept’ action was set to 0.5. The results show that even though this model exhibits the trend of increasing rejection rates observed in the AI-DI punish condition, the initial preferences (i.e., starting point of learning) diverges markedly from the Learning phase behavior we observed in Experiment 1:

Author response image 1.

This demonstrated that the participant at least maintains a representation of the teacher’s preference at the beginning. That is, they have prior knowledge about the shape of this preference. We incorporated this property into the model, that is, we considered a new model that assumes v-shaped starting values for rejection with two parameters, alpha and beta, governing the slope of this v-shaped function (this starting value actually mimics the shape of the preference functions of the Fehr-Schmidt model). We found that this new model (which we term the “Model RL Sim Vstart”) provided a satisfactory qualitative fit of the Transfer phase learning curves in Experiment 1 (see below).

Author response image 2.

However, we didn’t adopt this model as the best model for the following reasons. First, this model yielded a larger AIC value (indicating worse quantitative fit) compared to our preference Inference model in both Experiments 1 and 2, likely owing to its increased complexity (5 free parameters versus 4 in the Preference Inference model). Accordingly, we believe that inclusion of this model in our revised submission would be more distracting than helpful on account of the added complexity of explaining and justifying these assumptions, and of course its comparatively poor goodness of fit (relative to the preference inference model).

(6) Limitations of the Preference-Inference Model: The preference-inference model struggles to capture key aspects of the data, such as the increase in rejection rates for 70:30 DI offers during the learning phase (e.g. Figure 3A, AI+DI blue group). This is puzzling.

Thinking about this I realized the model makes quite strong unintuitive predictions that are not examined. For example, if a subject begins the learning phase rejecting the 70:30 offer more than 50% of the time (meaning the starting guilt parameter is higher than 1.5), then overleaning the tendency to reject will decrease to below 50% (the guilt parameter will be pulled down below 1.5). This is despite the fact the teacher rejects 75% of the offers. In other words, as learning continues learners will diverge from the teacher. On the other hand, if a participant begins learning to tend to accept this offer (guilt < 1.5) then during learning they can increase their rejection rate but never above 50%. Thus one can never fully converge on the teacher. I think this relates to the model's failure in accounting for the pattern mentioned above. I wonder if individuals actually abide by these strict predictions. In any case, these issues raise questions about the validity of the model as a representation of how individuals learn to align with a teacher's preferences (given that the model doesn't really allow for such an alignment).

In response to this comment we explain our efforts to build a new model that might be able conceptually resolves the issue identified by the Reviewer.

The key intuition guiding the Preference inference model is a Bayesian account of learning which we aimed to further simplify. In this setting, a Bayesian learner maintains a representation of the teacher’s inequity aversion parameters and updates it according to the teacher’s (observed) behavior. Intuitively, the posterior distribution shifts to the likelihood of the teacher’s action. On this view, when the teacher rejects, for instance, an AI offer, the learner should assign a higher probability to larger values of the Guilt parameter, and in turn the learner should change their posterior estimate to better capture the teacher’s preferences.

In the current study, we simplified this idea, implementing this sort of learning using incremental “delta rule” updating (e.g. Equation 8 of the main text). Then the key question is to define the “teaching signal”. Assuming that the teacher rejects an offer 70:30, based on Bayesian reasoning, the teacher’s envy parameter (α) is more likely to exceed 1.5 (computed as 30/(50-30), per equation 7) than to be smaller than 1.5. Thus, 1.5, which is then used in equation 8 to update α, can be thought of as a teaching signal. We simply assumed that if the initial estimate is already greater than 1.5, which means the prior is consistent with the likelihood, no updating would occur. This assumption raises the question of how to set the learning rate range. In principle, an envy parameter that is larger than 1.5 should be the target of learning (i.e., the teaching signal), and thus our model definition allows the learning rate to be greater than 1, incorporating this possibility.

Our simplified preference inference model has already successfully captured some key aspects of the participants’ learning behavior. However, it may fail in the following case: assume that the participant has an initial estimate of 1.51 for the envy parameter (β). Let’s say this corresponds to a rejection rate of 60%. Thus, no matter how many times the teacher rejects the offer 70:30, the participant’s estimate of the envy parameter remains the same, but observing only one offer acceptance would decrease this estimate, and in turn, would decrease the model’s predicted rejection rate. We believe this is the anomalous behavior—in 70:30 offers—identified by the Reviewer which the model does not appear able to recreate participants’ in these offers.

This issue actually touches the core of our model specification, that is, the choosing of the teaching signal. As we chose 1.5 as the teaching signal—i.e. lower bound on whenever the teacher rejects or accepts an offer of 70:30, a very small deviation of 1.5 would fail one part of updating. One way to mitigate this problem would be to choose a lower bound for α greater than 1.5, such that when the Teacher rejects a 70:30 offer, we assign a number greater than 1.5 (by ‘hard-coding’ this into the model via modification of equation 7). One sensible candidate value could be the middle point between 1.5 and 10 (the maximum value of α per our model definition). Intuitively, the model of this setting could still pull up the value of α to 1.51 when the teacher rejects 70:30, thus alleviating (but not completely eliminating) the anomaly.

We fitted this modified Preference Inference model to the data from Experiment 1 (see Author response image 3 below) and found that even though this model has a smaller AIC (and thus better quantitative fit than the original Preference Inference model), it still doesn’t fully capture the participants’ behavior for 70:30 offers.

Author response image 3.

Accordingly, rather than revising our model to include an unprincipled ‘kludge’ to account for this minor anomaly in the model behavior, we have opted to report our original model in our revision as we still believe it parsimoniously captures our intuitions about preference learning and provides a better fit to the observed behavior than the other RL models considered in the present study.

Reviewer #1 (Recommendations for the authors):

(1) I do not particularly prefer the acronyms AI and DI for disadvantageous inequity and advantageous inequity. Although they have been used in the literature, not every single paper uses them. More importantly, AI these days has such a strong meaning of artificial intelligence, so when I was reading this, I'd need to very actively inhibit this interpretation. I believe for the readability for a wider readership of eLife, I would advise not to use AI/DI here, but rather use the full terms.

We thank the Reviewer for this suggestion. As the full spelling of the two terms are somewhat lengthy, and appear frequently in the figures, we have elected to change the abbreviations for disadvantageous inequity and advantageous inequity to Dis-I and Adv-I, respectively in the main text and the supplementary information. We still use AI/DI in the response letter to make the terminology consistent.

(2) Do "punishment rate" and "rejection rate" mean the same? If so, it would be helpful to stick with one single term, eg, rejection rate.

We thank the Reviewer for this suggestion. As these terms have the same meaning, we have opted to use the term “rejection rate” throughout the main text.

(3) For the linear mixed effect models, were other random effect structures also considered (eg, random slops of experimental conditions)? It might be worth considering a few model specifications and selecting the best one to explain the data.

Thanks for this comment. Following established best practices (Barr, Levy, Scheepers, & Tily, 2013) we have elected to use a maximal random effects structure, whereby all possible predictor variables in the fixed effects structure also appear in the random effects structure.

(4) For equation (4), the softmax temperature is denoted as tau, but later in the text, it is called gamma. Please make it consistent.

We are appreciative of the Reviewer’s attention to detail. We have corrected this error.

Reviewer #2 (Recommendations for the authors):

(1) Several Tables in SI are unclear. I wasn't clear if these report raw probabilities of coefficients of mixed models. For any mixed models, it would help to give the model specification (e.g., Walkins form) and explain how variables were coded.

We are appreciative of the Reviewer’s attention to detail. We have clarified, in the captions accompanying our supplemental regression tables, that these coefficients represent log-odds. Regretfully we are unaware of the “Walkins form” the Reviewer references (even after extensive searching of the scientific literature). However, in our new revision we do include lme4 model syntax in our supplemental information which we believe will be helpful for readers seeking replicate our model specification.

(2) In one of the models it was said that the guilt and envy parameters were bounded between 0-1 but this doesn't make sense and I think values outside this range were later reported.

We are again appreciative of the Reviewer’s attention to detail. This was an error we have corrected— the actual range is [0,10].

(3) It is unclear if the model parameters are recoverable.

In response to this comment our revision now reports a basic parameter recovery analysis for the winning Preference Inference model. This is reported in our revised Methods:

“Finally, to verify if the free parameters of the winning model (Preference Inference) are recoverable, we simulated 200 artificial subjects, based on the Learning Phase of Experiment 1, with free parameters randomly chosen (uniformly) from their defined ranges. We then employed the same model-fitting procedure as described above to estimate these parameter value, observing that parameters. We found that all parameters of the model can be recovered (see Figure S2).”

And scatter plots depicting these simulated (versus recovered) parameters are given in Figure S2 of our revised Supplementary Information:

(4) I was confused about what Figure S2 shows. The text says this is about correlating contagious effects for different offers but the captions speak about learning effects. This is an important aspect which is unclear.

We have removed this figure in response to both Reviewers’ comments about the limited insights that can be drawn on the basis of these correlations.

eLife Assessment

This important study provides solid evidence for new insights into the role of Type-1 nNOS interneurons in driving neuronal network activity and controlling vascular network dynamics in awake, head-fixed mice. The authors use an original strategy based on the ablation of Type-1 nNOS interneurons with local injection of saporin conjugated to a substance P analogue into the somatosensory cortex. They show that ablation of type I nNOS neurons has surprisingly little effect on neurovascular coupling, although it alters neural activity and vascular dynamics.

Reviewer #1 (Public review):

Turner et al. present an original approach to investigate the role of Type-1 nNOS interneurons in driving neuronal network activity and in controlling vascular network dynamics in awake head-fixed mice. Selective activation or suppression of Type-1 nNOS interneurons has previously been achieved using either chemogenetic, optogenetic or local pharmacology. Here, the authors took advantage of the fact that Type-1 nNOS interneurons are the only cortical cells that express the tachykinin receptor 1 to ablate them with a local injection of saporin conjugated to substance P (SP-SAP). SP-SAP causes cell death in 90 % of type1 nNOS interneurons without affecting microglia, astrocytes and neurons. The authors report that the ablation has no major effects on sleep or behavior. Refining the analysis by scoring neural and hemodynamic signals with electrode recordings, calcium signal imaging and wide field optical imaging, they observe that Type-1 nNOS interneuron ablation does not change the various phases of the sleep/wake cycle. However, it does reduce low-frequency neural activity, irrespective of the classification of arousal state. Analyzing neurovascular coupling using multiple approaches, they report small changes in resting-state neural-hemodynamic correlations across arousal states, primarily mediated by changes in neural activity. Finally, they show that nNOS type 1 interneurons play a role in controlling interhemispheric coherence and vasomotion.

In conclusion, these results are interesting, use state-of-the-art methods and are well supported by the data and their analysis. I have only a few comments on the stimulus-evoked haemodynamic responses that can be easily addressed:

Comments on revisions:

As I mentioned in my initial review, this study is important. In my opinion, it could be published as is. Nonetheless, I am still somewhat dissatisfied with the authors' responses to my earlier comments. I understand that the same animals were not used for both stimulation paradigms, which is unfortunate. Nonetheless, I would have appreciated it if the authors had provided a couple of experiments illustrating GCaMP7 signals during brief stimulation in their reply to the reviewers. I am still unconvinced by the authors' suggestion that the GCaMP7 signal would remain stable during removal of the vascular undershoot. Since the absence of the undershoot is notable, I anticipate that a significant part of the initial response to prolonged stimulation is influenced by processes that occur during the 0.1-second stimulation, processes that may involve a change in the bulk neuronal response.

In short, the data could support or refute the following statement: "Loss of type-I nNOS neurons drove minimal changes in the vasodilation elicited by brief stimulation..."

Reviewer #2 (Public review):

Summary:

This important study by Turner et al., examines the functional role of a sparse but unique population of neurons in the cortex that express Nitric oxide synthase (Nos1). To do this, they pharmacolologically ablate these neurons in focal region of whisker related primary somatosensory (S1) cortex using a saponin-Substance P conjugate. Using widefield and 2-photon microscopy, as well as field recordings, they examine the impact of this cell specific lesion on blood flow dynamics and neuronal population activity. Within primary somatosensory cortex after Nos1 ablation, they find changes in neural activity patterns, decreased delta band power, reduced sensory evoked changes in blood flow (specifically eliminates the sustained blood flow change after stimulation) and decreased vasomotion.

Strengths:

This was a technically challenging study and the experiments were executed in an expert manner. The manuscript was well written and I appreciated the cartoon summary diagrams included in each figure. The analysis was rigorous and appropriate. Their discovery that Nos1 neurons can have significant effects on blood flow dynamics and neural activity is quite novel that should seed many follow up, mechanistic experiments to explain this phenomenon. The conclusions were justified by the convincing data presented.

Weaknesses:

I did not find any major flaws with the study. I originally noted some potential issues with the authors' characterization of the lesion and its extent, but that has been resolved in the revised manuscript.

Comments on revisions:

The authors have thoughtfully addressed the relatively minor concerns I had originally raised. Congratulations to the authors for producing this important paper.

Author response:

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

Reviewer #1 (Public review):

Turner et al. present an original approach to investigate the role of Type-1 nNOS interneurons in driving neuronal network activity and in controlling vascular network dynamics in awake head-fixed mice. Selective activation or suppression of Type-1 nNOS interneurons has previously been achieved using either chemogenetic, optogenetic, or local pharmacology. Here, the authors took advantage of the fact that Type-1 nNOS interneurons are the only cortical cells that express the tachykinin receptor 1 to ablate them with a local injection of saporin conjugated to substance P (SP-SAP). SP-SAP causes cell death in 90 % of type1 nNOS interneurons without affecting microglia, astrocytes, and neurons. The authors report that the ablation has no major effects on sleep or behavior. Refining the analysis by scoring neural and hemodynamic signals with electrode recordings, calcium signal imaging, and wide-field optical imaging, the authors observe that Type-1 nNOS interneuron ablation does not change the various phases of the sleep/wake cycle. However, it does reduce low-frequency neural activity, irrespective of the classification of arousal state. Analyzing neurovascular coupling using multiple approaches, they report small changes in resting-state neural-hemodynamic correlations across arousal states, primarily mediated by changes in neural activity. Finally, they show that nNOS type 1 interneurons play a role in controlling interhemispheric coherence and vasomotion.

In conclusion, these results are interesting, use state-of-the-art methods, and are well supported by the data and their analysis. I have only a few comments on the stimulus-evoked haemodynamic responses, and these can be easily addressed.

We thank the reviewer for their positive comments on our work.

Reviewer #2 (Public review):

Summary:

This important study by Turner et al. examines the functional role of a sparse but unique population of neurons in the cortex that express Nitric oxide synthase (Nos1). To do this, they pharmacologically ablate these neurons in the focal region of whisker-related primary somatosensory (S1) cortex using a saponin-substance P conjugate. Using widefield and 2photon microscopy, as well as field recordings, they examine the impact of this cell-specific lesion on blood flow dynamics and neuronal population activity. Locally within the S1 cortex, they find changes in neural activity paFerns, decreased delta band power, and reduced sensory-evoked changes in blood flow (specifically eliminating the sustained blood flow change amer stimulation). Surprisingly, given the tiny fraction of cortical neurons removed by the lesion, they also find far-reaching effects on neural activity paFerns and blood volume oscillations between the cerebral hemispheres.

Strengths:

This was a technically challenging study and the experiments were executed in an expert manner. The manuscript was well wriFen and I appreciated the cartoon summary diagrams included in each figure. The analysis was rigorous and appropriate. Their discovery that Nos1 neurons can have far-reaching effects on blood flow dynamics and neural activity is quite novel and surprising (to me at least) and should seed many follow-up, mechanistic experiments to explain this phenomenon. The conclusions were justified by the convincing data presented.

Weaknesses:

I did not find any major flaws in the study. I have noted some potential issues with the authors' characterization of the lesion and its extent. The authors may want to re-analyse some of their data to further strengthen their conclusions. Lastly, some methodological information was missing, which should be addressed.

We thank the reviewer for their enthusiasm for our work.

Reviewer #3 (Public review):

The role of type-I nNOS neurons is not fully understood. The data presented in this paper addresses this gap through optical and electrophysiological recordings in adult mice (awake and asleep).

This manuscript reports on a study on type-I nNOS neurons in the somatosensory cortex of adult mice, from 3 to 9 months of age. Most data were acquired using a combination of IOS and electrophysiological recordings in awake and asleep mice. Pharmacological ablation of the type-I nNOS populations of cells led to decreased coherence in gamma band coupling between lem and right hemispheres; decreased ultra-low frequency coupling between blood volume in each hemisphere; decreased (superficial) vascular responses to sustained sensory stimulus and abolishment of the post-stimulus CBV undershoot. While the findings shed new light on the role of type-I nNOS neurons, the etiology of the discrepancies between current observations and literature observations is not clear and many potential explanations are put forth in the discussion.

We thank the reviewer for their comments.

Reviewer #1 (Recommendations for the authors):  

(1) Figure 3, Type-1 nNOS interneuron ablation has complex effects on neural and vascular responses to brief (.1s) and prolonged (5s) whisker stimulation. During 0.1 s stimulation, ablation of type 1 nNOS cells does not affect the early HbT response but only reduces the undershoot. What is the pan-neuronal calcium response? Is the peak enhanced, as might be expected from the removal of inhibition? The authors need to show the GCaMP7 trace obtained during this short stimulation.

Unfortunately, we did not perform brief stimulation experiments in GCaMP-expressing mice. As we did not see a clear difference in the amplitude of the stimulus-evoked response with our initial electrophysiology recordings (Fig. 3a), we suspected that an effect might be visible with longer duration stimuli and thus pivoted to a pulsed stimulation over the course of 5 seconds for the remaining cohorts. It would have been beneficial to interweave short-stimulus trials for a direct comparison between the complimentary experiments, but we did not do this.

During 5s stimulation, both the early and delayed calcium/vascular responses are reduced. Could the authors elaborate on this? Does this mean that increasing the duration of stimulation triggers one or more additional phenomena that are sensitive to the ablation of type 1 nNOS cells and mask what is triggered by the short stimulation? Are astrocytes involved? How do they interpret the early decrease in neuronal calcium?

As our findings show that ablation reduces the calcium/vascular response more prominently during prolonged stimulation, we do suspect that this is due to additional NO-dependent mechanisms or downstream responses. NO is modulator of neural activity, generally increasing excitability (Kara and Friedlander 1999, Smith and Otis 2003), so any manipulation that changes NO levels will change (likely decrease) the excitability of the network, potentially resulting in a smaller hemodynamic response to sensory stimulation secondary to this decrease. While short stimuli engage rapid neurovascular coupling mechanisms, longer duration (>1s) stimulation could introduce additional regulatory elements, such as astrocytes, that operate on a slower time scale. On the right, we show a comparison of the control groups ploFed together from Fig. 3a and 3b with vertical bars aligned to the peak. During the 5s stimulation, the time-to-peak is roughly 830 milliseconds later than the 0.1s stimulation, meaning it’s plausible that the signals don’t separate until later. Our interpretation is that the NVC mechanisms responsible for brief stimulus-evoked change are either NO-independent or are compensated for in the SSP-SAP group by other means due to the chronic nature of the ablation. 

We have added the following text to the Discussion (Line 368): “Loss of type-I nNOS neurons drove minimal changes in the vasodilation elicited by brief stimulation, but led to decreased vascular responses to sustained stimulation, suggesting that the early phase of neurovascular coupling is not mediated by these cells, consistent with the multiple known mechanisms for neurovascular coupling (AFwell et al 2010, Drew 2019, Hosford & Gourine 2019) acting through both neurons and astrocytes with multiple timescales (Le Gac et al 2025, Renden et al 2024, Schulz et al 2012, Tran et al 2018).”

Author response image 1.

(2) In Figures 4d and e, it is unclear to me why the authors use brief stimulation to analyze the relationship between HbT and neuronal activity (gamma power) and prolonged stimulation for the relationship between HbT and GCaMP7 signal. Could they compare the curves with both types of stimulation?

As discussed previously, we did not use the same stimulation parameters across cohorts. The mice with implanted electrodes received only brief stimulation, while those undergoing calcium imaging received longer duration stimulus. 

Reviewer #2 (Recommendations for the authors):

(1) Results, how far-reaching is the cell-specific ablation? Would it be possible to estimate the volume of the cortex where Nos1 cells are depleted based on histology? Were there signs of neuronal injury more remotely, for example, beading of dendrites?

We regularly see 1-2 mm in diameter of cell ablation within the somatosensory cortex of each animal, which is consistent with the spread of small molecules. Ribosome inactivating proteins like SAP are smaller than AAVs (~5 nm compared to ~25 nm in diameter) and thus diffuse slightly further. We observed no obvious indication of neuronal injury more remotely or in other brain regions, but we did not image or characterize dendritic beading, as this would require a sparse labeling of neurons to clearly see dendrites (NeuN only stains the cell body). Our histology shows no change in cell numbers. 

We have added the following text to the Results (Line 124): “Immunofluorescent labeling in mice injected with Blank-SAP showed labeling of nNOS-positive neurons near the injection site. In contrast, mice injected with SP-SAP showed a clear loss in nNOS-labeling, with a typical spread of 1-2 mm from the injection site, though nNOS-positive neurons both subcortically and in the entirety of the contralateral hemisphere remaining intact.”

(2) For histological analysis of cell counts amer the lesion, more information is needed. How was the region of interest for counting cells determined (eg. 500um radius from needle/pipeFe tract?) and of what volume was analysed?

The region of interest for both SSP-SAP and Blank SAP injections was a 1 mm diameter circle centered around the injection site and averaged across sections (typically 3-5 when available). In most animals, the SSP-SAP had a lateral spread greater than 500 microns and encompassed the entire depth of cortex (1-1.5 mm in SI, decreasing in the rostral to caudal direction). The counts within the 1 mm diameter ROI were averaged across sections and then converted into the cells per mm area as presented. Note the consistent decrease in type I nNOS cells seen across mice in Fig 1d, Fig S1b.

We have added the following text in the Materials & Methods (Line 507): “The region of interest for analysis of cell counts was determined based on the injection site for both SP-SAP and Blank SAP injections, with a 1 mm diameter circle centered around the injection site and averaged across 3-5 sections where available. In most animals, the SP-SAP had a lateral spread greater than 500 microns and encompassed the entire depth of cortex (1-1.5 mm in SI).”

(3) Based on Supplementary Figure 1, it appears that the Saponin conjugate not only depletes Nos neurons but also may affect vascular (endothelial perhaps) Nos expression. Some quantification of this effect and its extent may be insighIul in terms of ascribing the effects of the lesion directly on neurons vs indirectly and perhaps more far-reaching via vascular/endothelial NOS.

Thank you for this comment. While this is a possibility, while we have found that the high nNOS expression of type-I nnoos neurons makes NADPH diaphorase a good stain for detecting them, it is less useful for cell types that expres NOS at lower levels.  We have found that the absolute intensity of NADPH diaphorase staining is somewhat variable from section to section. Variability in overall NADPH diaphorase intensity is likely due to several factors, such as duration of staining, thickness of the section, and differences in PFA concentration within the tissue and between animals. As NADPH diaphorase staining is highly sensitive to amount PFA exposure, any small differences in processing could affect the intensity, and slight differences in perfusion quality and processing could account. A second, perhaps larger issue could be due to differences in the number of arteries (which will express NOS at much higher levels than veins, and thus will appear darker) in the section. We did not stain for smooth muscle and so cannot differentiate arteries and veins.  Any difference in vessel intensity could be due to random variations in the numbers of arteries/veins in the section. While we believe that this is a potentially interesting question, our histological experiments were not able to address it.

(4) The assessment for inflammation took place 1 month amer the lesion, but the imaging presumably occurred ~ 2 weeks amer the lesion. Note that it seemed somewhat ambiguous as to when approximately, the imaging, and electrophysiology experiments took place relative to the induction of the lesion. Presumably, some aspects of inflammation and disruption could have been missed, at the time when experiments were conducted, based on this disparity in assessment. The authors may want to raise this as a possible limitation.

We apologize for our unclear description of the timeline. We began imaging experiments at least 4 weeks amer ablation, the same time frame as when we performed our histological assays. 

We have added the following text to the Discussion (Line 379): “With imaging beginning four weeks amer ablation, there could be compensatory rewiring of local and/or network activity following type-I nNOS ablation, where other signaling pathways from the neurons to the vasculature become strengthened to compensate for the loss of vasodilatory signaling from the typeI nNOS neurons.”

(5) Results Figure 2, please define "P or delta P/P". Also, for Figure 2c-f, what do the black vertical ticks represent?

∆P/P is the change in the gamma-band power relative to the resting-state baseline, and black tick marks indicate binarized periods of vibrissae motion (‘whisking’). We have clarified this in Figure caption 2 (Line 174).

(6) Figure 3b-e, is there not an undershoot (eventually) amer 5s of stimulation that could be assessed? 

Previous work has shown that there is no undershoot in response to whisker stimulations of a few seconds (Drew, Shih, Kelinfeld, PNAS, 2011).  The undershoot for brief stimuli happens within ~2.5 s of the onset/cessation of the brief stimulation, this is clearly lacking in the response to the 5s stim (Fig 3).  The neurovascular coupling mechanisms recruited during the short stimulation are different than those recruited during the long stimulus, making a comparison of the undershoot between the two stimulation durations problematic. 

For Figures 3e and 6 how was surface arteriole diameter or vessel tone measured? 2P imaging of fluorescent dextran in plasma? Please add the experimental details of 2P imaging to the methods. Including some 2P images in the figures couldn't hurt to help the reader understand how these data were generated.

We have added details about our 2-photon imaging (FITC-dextran, full-width at half-maximum calculation for vessel diameter) as well as a trace and vessel image to Figure 2.

We have added the following text to the Materials & Methods (Line 477): “In two-photon experiments, mice were briefly anesthetized and retro-orbitally injected with 100 µL of 5% (weight/volume) fluorescein isothiocyanate–dextran (FITC) (FD150S, Sigma-Aldrich, St. Louis, MO) dissolved in sterile saline.”

We have added the following text to the Materials & Methods (Line 532): “A rectangular box was drawn around a straight, evenly-illuminated vessel segment and the pixel intensity was averaged along the long axis to calculate the vessel’s diameter from the full-width at half-maximum (https://github.com/DrewLab/Surface-Vessel-FWHM-Diameter; (Drew, Shih et al. 2011)).”

(7) Did the authors try stimulating other body parts (eg. limb) to estimate how specific the effects were, regionally? This is more of a curiosity question that the authors could comment on, I am not recommending new experiments.

We did measure changes in [HbT] in the FL/HL representation of SI during locomotion (Line 205), which is known to increase neural activity in the somatosensory cortex (Huo, Smith and Drew, Journal of Neuroscience, 2014; Zhang et al., Nature Communications 2019). We observed a similar but not statistically significant trend of decreased [HbT] in SP-SAP compared to control. This may have been due to the sphere of influence of the ablation being centered on the vibrissae representation and not having fully encompassed the limb representation. We agree with the referee that it would be interesting to characterize these effects on other sensory regions as well as brain regions associated with tasks such as learning and behavior.

(8) Regarding vasomotion experiments, are there no other components of this waveform that could be quantified beyond just variance? Amplitude, frequency? Maybe these don't add much but would be nice to see actual traces of the diameter fluctuations. Further, where exactly were widefield-based measures of vasomotion derived from? From some seed pixel or ~1mm ROI in the center of the whisker barrel cortex? Please clarify.

The reviewer’s point is well taken. We have added power spectra of the resting-state data which provides amplitude and frequency information. The integrated area under the curve of the power spectra is equal to the variance. Widefield-based measures of vasomotion were taken from the 1 mm ROI in the center of the whisker barrel cortex.

We have added the following text to the Materials & Methods (Line 560): “Variance during the resting-state for both ∆[HbT] and diameter signals (Fig. 7) was taken from resting-state events lasting ≥10 seconds in duration. Average ∆[HbT] from within the 1 mm ROI over the vibrissae representation of SI during each arousal state was taken with respect to awake resting baseline events ≥10 seconds in duration.” 

(9) On page 13, the title seems like a bit strong. The data show a change in variance but that does not necessarily mean a change in absolute amplitude. Also, I did not see any reports of absolute vessel widths between groups from 2P experiments so any difference in the sampling of larger vs smaller arterioles could have affected the variance (ie. % changes could be much larger in smaller arterioles).

We have updated the title of Figure 7 to specifically state power (which is equivalent to the variance) rather than amplitude (Line 331). We have also added absolute vessel widths to the Results (Line 340): “There was no difference in resting-state (baseline) diameter between the groups, with Blank-SAP having a diameter of 24.4 ± 7.5 μm and SP-SAP having a diameter of 23.0 ± 9.4 μm (Fest, p ti 0.61). “

(10) Big picture question. How could a manipulation that affects so few cells in 1 hemisphere (below 0.5% of total neurons in a region comprising 1-2% of the volume of one hemisphere) have such profound effects in both hemispheres? The authors suggest that some may have long-range interhemispheric projections, but that is presumably a fraction of the already small fraction of Nos1 neurons. Perhaps these neurons have specializing projections to subcortical brain nuclei (Nucleus Basilis, Raphe, Locus Coerulus, reticular thalamus, etc) that then project widely to exert this outsized effect? Has there not been a detailed anatomical characterization of their efferent projections to cortical and sub-cortical areas? This point could be raised in the discussion.

We apologize for the lack of clarity of our work in this point.  We would like to clarify that the only analysis showing a change in the unablated hemisphere being coherence/correlation analysis between the two hemispheres.  Other metrics (LFP power and CBV power spectra) do not change in the hemisphere contralateral to the injections site, as we show in data added in two supplementary figures (Fig. S4 and 7). The coherence/correlation is a measure of the correlated dynamics in the two hemispheres. For this metric to change, there only needs to be a change in the dynamics of one hemisphere relative to another.  If some aspects of the synchronization of neural and vascular dynamics across hemispheres are mediated by concurrent activation of type I nNOS neurons in both hemispheres, ablating them in one hemisphere will decrease synchrony. It is possible that type I nNOS neurons make some subcortical projections that were not reported in previous work (Tomioka 2005, Ruff 2024), but if these exist they are likely to be very small in number as they were not noted.  

We have added the text in the Results (Line 228): “In contrast to the observed reductions in LFP in the ablated hemisphere, we noted no gross changes in the power spectra of neural LFP in the unablated hemisphere (Fig. S7) or power of the cerebral blood volume fluctuations in either hemisphere (Fig. S4).”

Line 335): “The variance in ∆[HbT] during rest, a measure of vasomotion amplitude, was significantly reduced following type-I nNOS ablation (Fig. 7a), dropping from 40.9 ± 3.4 μM² in the Blank-SAP group (N ti 24, 12M/12F) to 23.3 ± 2.3 μM² in the SP-SAP group (N ti 24, 11M/13F) (GLME p ti 6.9×10^-5) with no significant di[erence in the unablated hemisphere (Fig. S7).”

Reviewer #3 (Recommendations for the authors):

(1)  The reporting would be greatly strengthened by following ARRIVE guidelines 2.0: https://arriveguidelines.org/: aFrition rates and source of aFrition, justification for the use of 119 (beyond just consistent with previous studies), etc.

We performed a power analysis prior to our study aiming to detect a physiologically-relevant effect size of (Cohen’s d) ti 1.3, or 1.3 standard deviations from the mean. Alpha and Power were set to the standard 0.05 and 0.80 respectively, requiring around 8 mice per group (SP-SAP, Blank, and for histology, naïve animals) for multiple independent groups (ephys, GCamp, histology). To potentially account for any aFrition due to failures in Type-I nNOS neuron ablation or other problems (such as electrode failure or window issues) we conservatively targeted a dozen mice for each group. Of mice that were imaged (1P/2P), two SP-SAP mice were removed from the dataset (24 SP-SAP remaining) post-histological analysis due to not showing ablation of nNOS neurons, an aFrition rate of approximately 8%.

We have added the following text to the Materials & Methods (Line 441): “Sample sizes are consistent with previous studies (Echagarruga et al 2020, Turner et al 2023, Turner et al 2020, Zhang et al 2021) and based on a power analysis requiring 8-10 mice per group (Cohen’s d ti 1.3, α ti 0.05, (1 - β) ti 0.800). Experimenters were not blind to experimental conditions or data analysis except for histological experiments. Two SP-SAP mice were removed from the imaging datasets (24 SP-SAP remaining) due to not showing ablation of nNOS neurons during post-histological analysis, an aFrition rate of approximately 8%.”

(2) Intro, line 38: Description of the importance of neurovascular coupling needs improvement. Coordinated haemodynamic activity is vital for maintaining neuronal health and the energy levels needed.

We have added a sentence to the introduction (Line 41): “Neurovascular coupling plays a critical role in supporting neuronal function, as tightly coordinated hemodynamic activity is essential for meeting energy metabolism and maintaining brain health (Iadecola et al 2023, Schaeffer & Iadecola 2021).“

(3) Given the wide range of mice ages, how was the age accounted for/its effects examined?

Previous work from our lab has shown that there is no change in hemodynamics responses in awake mice over a wide range of ages (2-18 months), so the age range we used (3 and 9 months of age) should not impact this.  

We have added the following text in the Results (Line 437): “Previous work from our lab has shown that the vasodilation elicited by whisker stimulation is the same in 2–4-month-old mice as in 18-month-old mice (BenneF, Zhang et al. 2024). As the age range used here is spanned by this time interval, we would not expect any age-related differences.”

(4) How was the susceptibility of low-frequency neuronal coupling signals to noise managed? How were the low-frequency bands results validated?

We are not sure what the referee is asking here. Our electrophysiology recordings were made differentially using stereotrodes with tips separated by ~100µm, which provides excellent common-mode rejection to noise and a localized LFP signal. Previous publications from our lab (Winder et al., Nature Neuroscience 2017; Turner et al., eLife2020) and others (Tu, Cramer, Zhang, eLife 2024) have repeatedly show that there is a very weak correlation between the power in the low frequency bands and hemodynamic signals, so our results are consistent with this previous work. 

(5) It would be helpful to demonstrate the selectivity of cell *death* (as opposed to survival) induced by SP-SAP injections via assessments using markers of cell death.

We agree that this would be helpful complement to our histological studies that show loss of type-I nNOS neurons, but no loss of other cells and minimal inflammation with SP-saporin injections.  However, we did not perform histology looking at cell death, only at surviving cells, given that we see no obvious inflammation or cells loss, which would be triggered by nonspecific cell death.  Previous work has established that saporin is cytotoxic and specific only to cell that internalize the saporin.   Internalization of saporin causes cell death via apoptosis (Bergamaschi, Perfe et al. 1996), and that the substance P receptor is internalized when the receptor is bound (Mantyh, Allen et al. 1995). Treatment of internalized saporin generates cellular debris that is phagocytosed by microglial, consistent with cell death (Seeger, Hartig et al. 1997). While it is possible that treatment of SP-saporin causes type 1 nNOS neurons to stop expressing nitric oxide synthase (which would make them disappear from our IHC staining), we think that this is unlikely given the literature shows internalized saporin is clearly cytotoxic. 

We have added the following text to the Results (Line 131): “It is unlikely that the disappearance of type-I nNOS neurons is because they stopped expressing nNOS, as internalized saporin is cytotoxic. Exposure to SP-conjugated saporin causes rapid internalization of the SP receptor-ligand complex (Mantyh, Allen et al. 1995), and internalized saporin causes cell death via apoptosis (Bergamaschi, Perfe et al. 1996). In the brain, the resulting cellular debris from saporin administration is then cleared by microglia phagocytosis (Seeger, Hartig et al. 1997).”

(6) Was the decrease in inter-hemispheric correlation associated with any changes to the corpus callosum?

We noted no gross changes to the structure of the corpus callosum in any of our histological reconstructions following SSPSAP administration, however, we did not specifically test for this. Again, as we note in our reply in reviewer 2, the decrease in interhemispheric synchronization does not imply that there are changes in the corpus callosum and could be mediated by the changes in neural activity in the hemisphere in which the Type-I nNOS neurons were ablated.

(7) How were automated cell counts validated?

Criteria used for automated cell counts were validated with comparisons of manual counting as described in previous literature. We have added additional text describing the process in the Materials & Methods (Line 510): “For total cell counts, a region of interest (ROI) was delineated, and cells were automatically quantified under matched criteria for size, circularity and intensity. Image threshold was adjusted until absolute value percentages were between 1-10% of the histogram density. The function Analyze Par-cles was then used to estimate the number of particles with a size of 100-99999 pixels^2 and a circularity between 0.3 and 1.0 (Dao, Suresh Nair et al. 2020, Smith, Anderson et al. 2020, Sicher, Starnes et al. 2023). Immunoreactivity was quantified as mean fluorescence intensity of the ROI (Pleil, Rinker et al. 2015).”

(8) Given the weighting of the vascular IOS readout to the superficial tissue, it is important to qualify the extent of the hemodynamic contrast, ie the limitations of this readout.

We have added the following text to the Discussion (Line 385): “Intrinsic optical signal readout is primarily weighted toward superficial tissue given the absorption and scaFering characteristics of the wavelengths used. While surface vessels are tightly coupled with neural activity, it is still a maFer of debate whether surface or intracortical vessels are a more reliable indicator of ongoing activity (Goense et al 2012; Huber et al 2015; Poplawsky & Kim 2014).” 

(9) Partial decreases observed through type-I iNOS neuronal ablation suggest other factors also play a role in regulating neural and vascular dynamics: data presented thus do *not* "indicate disruption of these neurons in diseases ranging from neurodegeneration to sleep disturbances," as currently stated. Please revise.

We agree with the reviewer. We have changed the abstract sentence to read (Line 30): “This demonstrates that a small population of nNOS-positive neurons are indispensable for regulating both neural and vascular dynamics in the whole brain, raising the possibility that loss of these neurons could contribute to the development of neurodegenerative diseases and sleep disturbances.”

eLife Assessment

This paper addresses a significant question regarding the low overlap between genetic discoveries for human complex diseases and those for gene expression by emphasizing the contribution of cell-type-specific chromatin accessibility QTLs. The analyses supporting the main claims are convincing, and the key conclusions are valuable and of interest to readers in the fields of human genetics and functional genomics.

Reviewer #1 (Public review):

Most human traits and common diseases are polygenic, influenced by numerous genetic variants across the genome. These variants are typically non-coding and likely function through gene regulatory mechanisms. To identify their target genes, one strategy is to examine if these variants are also found among genetic variants with detectable effects on gene expression levels, known as eQTLs. Surprisingly, this strategy has had limited success, and most disease variants are not identified as eQTLs, a puzzling observation recently referred to as "missing regulation".

In this work, Jeong and Bulyk aimed to better understand the reasons behind the gap between disease-associated variants and eQTLs. They focused on immune-related diseases and used lymphoblastoid cell lines (LCLs) as a surrogate for the cell types mediating the genetic effects. Their main hypothesis is that some variants without eQTL evidence might be identifiable by studying other molecular intermediates along the path from genotype to phenotype. They specifically focused on variants that affect chromatin accessibility, known as caQTLs, as a potential marker of regulatory activity.

The authors present data analyses supporting this hypothesis: several disease-associated variants are explained by caQTLs but not eQTLs. They further show that although caQTLs and eQTLs likely have largely overlapping underlying genetic variants, some variants are discovered only through one of these mapping strategies. Notably, they demonstrate that eQTL mapping is underpowered for gene-distal variants with small effects on gene expression, whereas caQTL mapping is not dependent on the distance to genes. Additionally, for some disease variants with caQTLs but no corresponding eQTLs in LCLs, they identify eQTLs in other cell types.

Altogether, Jeong and Bulyk convincingly demonstrate that for immune-related diseases, discovering the missing disease-eQTLs requires both larger eQTL studies and a broader range of cell types in expression assays. It remains to be seen what fractions of the missing disease-eQTLs will be discovered with either strategy and whether these results can be extended to other diseases or traits.

It should be noted that the problem of "missing regulation" has been investigated and discussed in several recent papers, notably Umans et al., Trends in Genetics 2021; Connally et al., eLife 2022; Mostafavi et al., Nat. Genet. 2023. The results reported by Jeong and Bulyk are not unexpected in light of this previous work (all of which they cite), but they add valuable empirical evidence that mostly aligns with the model and discussions presented in Mostafavi et al.

Reviewer #2 (Public review):

eQTLs have emerged as a method for interpreting GWAS signals. However, some GWAS signals are difficult to explain with eQTLs. In this paper, the authors demonstrated that caQTLs can explain these signals. This suggests that for GWAS signals to actually lead to disease phenotypes, they must be accessible in the chromatin. This implies that for GWAS signals to translate into disease phenotypes, they need to be accessible within the chromatin.

However, fundamentally, caQTLs, like GWAS, have the limitation of not being able to determine which genes mediate the influence on disease phenotypes. This limitation is consistent with the constraints observed in this study.

(1) Reproducibility / Methods. The concrete numbers provided in the authors' response (e.g., 20 YRI LCL ATAC‑seq samples used only for peak discovery; caQTL mapping restricted to 100 GBR LCLs; 99,320 ATAC peaks tested vs 14,872 genes for eQTL; 373 European RNA‑seq samples, with clarification of overlap) do not appear to be reflected in the Methods. These specifics should be incorporated directly into the Methods sections.

(2) Experimental evidence demonstrating transcription factor binding at representative caQTL peaks would strengthen causal interpretation of these loci.

(3) Tissue/cell‑type specificity of caQTLs: Prior work supports that chromatin‑level effects are broadly shared across cellular states, whereas expression effects are more context‑specific; thus, caQTLs are generally less "state‑specific" than eQTLs. However, this does not imply equivalence across distinct cell types: caQTLs derived from different cell types may yield different results, particularly where accessibility is cell‑type restricted.

Author response:

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

Reviewer #1 (Public Review): 

Most human traits and common diseases are polygenic, influenced by numerous genetic variants across the genome. These variants are typically non-coding and likely function through gene regulatory mechanisms. To identify their target genes, one strategy is to examine if these variants are also found among genetic variants with detectable effects on gene expression levels, known as eQTLs. Surprisingly, this strategy has had limited success, and most disease variants are not identified as eQTLs, a puzzling observation recently referred to as "missing regulation". 

In this work, Jeong and Bulyk aimed to better understand the reasons behind the gap between disease-associated variants and eQTLs. They focused on immune-related diseases and used lymphoblastoid cell lines (LCLs) as a surrogate for the cell types mediating the genetic effects. Their main hypothesis is that some variants without eQTL evidence might be identifiable by studying other molecular intermediates along the path from genotype to phenotype. They specifically focused on variants that affect chromatin accessibility, known as caQTLs, as a potential marker of regulatory activity. 

The authors present data analyses supporting this hypothesis: several disease-associated variants are explained by caQTLs but not eQTLs. They further show that although caQTLs and eQTLs likely have largely overlapping underlying genetic variants, some variants are discovered only through one of these mapping strategies. Notably, they demonstrate that eQTL mapping is underpowered for gene-distal variants with small effects on gene expression, whereas caQTL mapping is not dependent on the distance to genes. Additionally, for some disease variants with caQTLs but no corresponding eQTLs in LCLs, they identify eQTLs in other cell types. 

Altogether, Jeong and Bulyk convincingly demonstrate that for immune-related diseases, discovering the missing disease-eQTLs requires both larger eQTL studies and a broader range of cell types in expression assays. It remains to be seen what fractions of the missing diseaseeQTLs will be discovered with either strategy and whether these results can be extended to other diseases or traits. 

We thank the reviewer for their accurate summary of our study and positive review of our findings for immune-related diseases.

It should be noted that the problem of "missing regulation" has been investigated and discussed in several recent papers, notably Umans et al., Trends in Genetics 2021; Connally et al., eLife 2022; Mostafavi et al., Nat. Genet. 2023. The results reported by Jeong and Bulyk are not unexpected in light of this previous work (all of which they cite), but they add valuable empirical evidence that mostly aligns with the model and discussions presented in Mostafavi et al. 

We thank the reviewer for their positive review of our results and manuscript. As Reviewer #1 noted, whether our and others' observation extends to other diseases or traits is an open question. For instance, Figure 2b in Mostafavi et al., Nat. Genet. (2023) demonstrated that there was a spectrum of depletion of eQTLs and enrichment of GWAS signals in constrained genes across various tissues and traits, respectively. Therefore, gene expression constraint may play a larger or smaller role in different diseases or traits. That immune cell types and cell states are extremely diverse (Schmiedel et al., Cell (2018) and Calderon et al., Nat. Genet. (2019), just to name a few) likely adds to the complexity of gene regulation that contributes to immune-mediated disease.

Reviewer #2 (Public Review): 

Summary: 

eQTLs have emerged as a method for interpreting GWAS signals. However, some GWAS signals are difficult to explain with eQTLs. In this paper, the authors demonstrated that caQTLs can explain these signals. This suggests that for GWAS signals to actually lead to disease phenotypes, they must be accessible in the chromatin. This implies that for GWAS signals to translate into disease phenotypes, they need to be accessible within the chromatin. 

However, fundamentally, caQTLs, like GWAS, have the limitation of not being able to determine which genes mediate the influence on disease phenotypes. This limitation is consistent with the constraints observed in this study. 

We thank the reviewer for their accurate summary of our results.

(1) For reproducibility, details are necessary in the method section.

Details about adding YRI samples in ATAC-seq: For example, how many samples are there, and what is used among public data? There is LCL-derived iPSC and differentiated iPSC (cardiomyocytes) data, not LCL itself. How does this differ from LCL, and what is the rationale for including this data despite the differences?

Banovich et al., Genome Research (2018) (PMID: 29208628), who generated data using LCLderived iPSCs and differentiated iPSCs (cardiomyocytes), also generated ATAC-seq data from 20 YRI LCL samples. We analyzed those data to identify open chromatin regions (i.e., ATACseq peaks) in LCLs and merged the regions with open chromatin regions identified with 100 GBR LCL samples from two studies by Kumasaka et al. (Nature Genetics (2016)

PMID: 26656845 and Nature Genetics (2019) PMID: 30478436). However, we restricted the caQTL analysis to only the 100 GBR samples because of possible ancestry effects and batch effects. We attempted caQTL analysis with the 20 YRI samples as well, but the result was noisy, likely due to smaller sample size and lower read depth of the ATAC-seq data.

caQTL is described as having better power than eQTL despite having fewer samples. How does the number of ATAC peaks used in caQTL compare to the number of gene expressions used in eQTL?

The number of ATAC peaks used in caQTL (99,320) is ~6.7 times greater than the number of genes (14,872) used in the eQTL analysis. Therefore, there is a higher chance of detecting a significant caQTL signal and a significant colocalization signal than there is for eQTLs. However, we reasoned that since distal eQTLs are more easily detected as caQTLs and since increasing the sample size of eQTLs through meta-analysis uncovered additional eQTL colocalization at loci with caQTL colocalization only, colocalized caQTLs are likely capturing disease-relevant regulatory effects.

Details about RNA expression data: In the method section, it states that raw data (ERP001942) was accessed, and in data availability, processed data (E-GEUV-1) was used. These need to be consistent.

Thank you for pointing this out. We used the processed data from Expression Atlas (https://www.ebi.ac.uk/gxa/experiments/E-GEUV-1/Results), and that's what we meant by "We downloaded RNA expression level data of the LCL samples from the Expression Atlas." We have revised the “RNA expression data preparation” section in our manuscript to make the text clearer.

How many samples were used (the text states 373, but how was it reduced from the original 465, and the total genotype is said to be 493 samples while ATAC has n=100; what are the 20 others?), and it mentions European samples, but does this exclude YRI?

We thank the reviewer for pointing out these points of confusion. Our reported count of 493 samples included YRI samples with RNA-seq data or ATAC-seq data that we ultimately did not use for QTL analyses. There were 373 European samples with RNA-seq data that we used for eQTL analysis, and 100 GBR samples (including some that overlap with the 373 European samples) that we used for caQTL analysis. We have revised the text to clarify these points.

(2) Experimental results determining which TFs might bind to the representative signals of caQTL are required.

We agree that caQTL colocalization is just the start of elucidating the regulatory mechanism of a GWAS locus. Determining which TFs are bound and which TFs' binding is altered would be necessary to describe the causal regulatory mechanism. For this, we utilized the Cistrome database to search for TFs whose binding overlaps the colocalized caQTL peaks. We present the results of this analysis in Supplementary Table 3 and Supplementary Figure 4, both of which we have added in our revised manuscript. Overall, protein factors associated with active transcription, such as POL2RA, and several immune cell TFs, including RUNX3, SPI1, and RELA, were frequently detected in those peaks. Detecting these factors in most peaks supports the likelihood that the colocalized caQTL peaks are active cis-regulatory elements. These results are consistent with our observation of enriched caQTL-mediated heritability in regions with active histone marks (Figure 1).

(3) It is stated that caQTL is less tissue-specific compared to eQTL; would caQTL performed with ATAC-seq results from different cell types, yield similar results?

We thank the reviewer for the question. Calderon et al. (PMID: 31570894) observed that "most effects on allelic imbalance (of ATAC-seq) were shared regardless of lineage or condition". Yet, there were regions where a different cell type or state would show inaccessibility (Figure 4d in Calderon et al.). Thus, we expect that ATAC-seq results from different cell types (e.g., T cells, B cells, monocytes, etc.) would lead to additional caQTLs showing colocalization at cell-typespecific open chromatin. However, if a region is accessible in both cell types, caQTL may be detected in both. Moreover, Alasoo et al., Nature Genetics (2018) (PMID: 29379200) observed that “many disease-risk variants affect chromatin structure in a broad range of cellular states, but their effects on expression are highly context specific.” In both studies, the authors investigated immune cell types, and there could be different observations in non-immune cell types and other diseases and traits.

Reviewer #1 (Recommendations For The Authors): 

I think it would strengthen the paper to explore gene-level differences in the discovery of caQTLs and eQTLs. For example, complex disease-relevant genes, on average, have more/longer regulatory domains (as shown by Wang and Goldstein, AJHG 2020; Mostafavi et al., Nat. Genet. 2023). Therefore, it is plausible that for such genes, caQTLs are much more easily discoverable than eQTLs due to (i) a larger mutational target size for caQTLs, and (ii) dispersion of expression heritability across multiple domains, which hampers the discovery of eQTLs but not caQTLs, which are studied independently of other domains in the region. In other words, discovered caQTLs and eQTLs likely vary in terms of their distance to genes (as the authors report), as well as their target genes.

We thank the reviewer for the suggestion to explore gene-level differences. We expect that the effects of complex disease-relevant genes having more / longer regulatory domains, on average, to explain our observations. We agree on both of your points that there are many more regulatory elements that are captured as accessible regions than expressed genes and that genes often have multiple independent eQTLs leading to dispersion of heritability. The genelevel trend that we described was the distance of the regulatory element from the genes. Additional analyses would be a relevant future direction.

Also considering gene-level analysis, Mostafavi et al. show that the types of biases they report for eQTLs also apply to other molecular QTLs. It would be valuable to compare GWAS hits with versus without caQTL colocalization. Similarly, it would be insightful to compare GWAS hits with both colocalized caQTLs and eQTLs to GWAS hits with colocalized caQTLs but no eQTLs in any of the cell types. 

We thank the reviewer for the comment. Investigating for potential biases in the colocalized caQTL would be useful, but we considered it beyond the scope of this work. In terms of biological factors, we demonstrated through mediated heritability analyses that more accessible chromatin (based on ATAC-seq read coverage) and regions with active histone marks were enriched for autoimmune disease associations (Figure 1). Furthermore, as greater distance of the regulatory variant from the transcription start site significantly reduced the cis-heritability, we would expect that distance would play a major role, similar to Mostafavi et al.’s conclusions.

I don't think the argument for the role of natural selection contributing to the "missing regulation" is presented accurately. Specifically, large eQTLs acting on top trait-relevant genes are under stronger selection and thus, on average, segregate at lower frequencies. This makes them difficult to discover in eQTL assays. However, if not lost, they contribute as much, if not more, to trait heritability than weaker eQTLs at the same gene because their larger effects compensate for their lower frequency. At the most extreme, selection should have a "flattening" effect (e.g., see Simons et al., PLOS Biol 2018; O'Connor et al., AJHG 2019): weak and strong eQTLs at the same gene are expected to contribute equally to heritability. Therefore, the statement "Consequently, only weak eQTL variants, often in regions distal to the gene's promoter, may remain and affect traits" is not correct. If this turns out to be empirically true, other models, such as pleiotropic selection, need to explain it. 

We thank the reviewer for the correction. We agree with the comment and have revised the sentences in the introduction accordingly.

It is worth speculating why caQTLs may be more consistent across cell types than cis-eQTLs. Additionally, readers may infer from the paper that the focus should shift from eQTLs to caQTLs, which may not be the authors' intention. Perhaps these approaches are complementary: caQTLs can help with TSS-distal disease variants, while finding the target gene and regulatory context is more straightforward with eQTL colocalization. Addressing these points in the discussion will be helpful.

We appreciate the reviewer's suggestion to clarify the advantages of incorporating cis-eQTLs and caQTLs. Our argument is exactly as you put it, and we added a paragraph on this in the Discussion.

I believe the authors could do more to contextualize their findings within the existing literature on the subject, particularly Umans et al., Trends in Genetics 2021; Connally et al., eLife 2022; and Mostafavi et al., Nat. Genet. 2023. For instance, Umans et al. suggest that "if most standard eQTLs are generally benign, increasing sample size and adding more tissue types in an effort to identify even more standard eQTLs may not help us to explain many more disease risk mutations". Conversely, Mostafavi et al. argue for a multipronged approach, which appears more aligned with the authors' conclusions.

We followed the reviewer’s suggestion to place our work in the context of existing literature on this topic. Moreover, we clarified what our recommendations for future data generation are.

I thought Figures 1C-D were unclear. 

We added a sentence in the figure legend describing that stronger and more significant enrichment indicate that mediated heritability is concentrated in that subset.

Reviewer #2 (Recommendations For The Authors): 

Complete workflow figures for caQTL calling and eQTL calling are required. 

To improve clarity of the caQTL and eQTL calling workflow, we added Supplementary Figure 1.

eLife Assessment

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

Reviewer #1 (Public review):

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

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

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

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

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

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

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

Reviewer #2 (Public review):

Summary:

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

Strengths:

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

Weaknesses:

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

Auhtor response:

Public Reviews:

Reviewer #1 (Public review):

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

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

We are grateful for this astute remark. A comparison of gfDNA concentration among the diagnostic groups indicates a trend of increasing values as the diagnosis progresses toward malignancy. The observed values for the diagnostic groups are as follows:

Author response table 1.

The chart below presents the statistical analyses of the same diagnostic/tumor-stage groups (One-Way ANOVA followed by Tukey’s multiple comparison tests). It shows that gastric fluid gfDNA concentrations gradually increase with malignant progression. We observed that the initial tumor stages (T0 to T2) exhibit intermediate gfDNA levels, which in this group is significantly lower than in advanced disease (p = 0.0036), but not statistically different from non-neoplastic disease (p = 0.74).

Author response image 1.

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

We appreciate the attention to detail regarding the numbers analyzed in the manuscript. Importantly, the results are meaningful because the number of subjects in each group is comparable (T0-T2, N = 65; T3, N = 91; T4, N = 63). The mean gastric fluid gfDNA values (ng/µL) increase with disease stage (T0-T2: 15.12; T3-T4: 30.75), and both are higher than the mean gfDNA values observed in non-neoplastic disease (10.81 ng/µL for N+PD and 10.10 ng/µL for PN). These subject numbers in each diagnostic group accurately reflect real-world data from a tertiary cancer center.

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

Histopathological analyses were performed throughout the study not only for the initial diagnosis of tissue biopsies, but also for the classification of Lauren’s subtypes, tumor staging, and the assessment of the presence and extent of immune cell infiltrates. Regarding the time of disease onset, this variable is inherently unknown--by definition--at the time of a diagnostic EGD. While the prognosis definition is indeed straightforward, we believe that a simple, cost-effective, and practical approach is advantageous for patients across diverse clinical settings and is more likely to be effectively integrated into routine EGD practice.

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

We wish to reinforce that EGD, along with conventional histopathology, remains the gold standard for gastric cancer evaluation. EGD under sedation is routinely performed for diagnosis, and the collection of gastric fluids for gfDNA evaluation does not affect patient comfort. Thus, while gfDNA analysis was evidently not intended as a diagnostic EGD and biopsy replacement, it may provide added prognostic value to this exam.

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

We are grateful for these comments and apologize for the clerical oversight. All figures, tables, titles and figure legends have now been double-checked.

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

We assume that the unusual wording remark regarding “overall logicality” pertains to the rationale and/or reasoning of this investigational study. Our working hypothesis was that during neoplastic disease progression, tumor cells continuously proliferate and, depending on various factors, attract immune cell infiltrates. Consequently, both tumor cells and immune cells (as well as tumor-derived DNA) are released into the fluids surrounding the tumor at its various locations, including blood, urine, saliva, gastric fluids, and others. Thus, increases in DNA levels within some of these fluids have been documented and are clinically meaningful. The concurrent observation of elevated gastric fluid gfDNA levels and immune cell infiltration supports the hypothesis that increased gfDNA—which may originate not only from tumor cells but also from immune cells—could be associated with better prognosis, as suggested by this study of a large real-world patient cohort.

In summary, we thank Reviewer #1 for his time and effort in a constructive critique of our work.

Reviewer #2 (Public review):

Summary:

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

Strengths:

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

Reviewer #2 has conceptually grasped the overall rationale of the study quite well, and we are grateful for their assessment and comprehensive summary of our findings.

Weaknesses:

(1) The study has several notable weaknesses. The association between high gfDNA levels and better survival contradicts conventional expectations and raises concerns about the biological interpretation of the findings.

We agree that this would be the case if the gfDNA was derived solely from tumor cells. However, the findings presented here suggest that a fraction of this DNA would be indeed derived from infiltrating immune cells. The precise determination of the origin of this increased gfDNA remains to be achieved in future follow-up studies, and these are planned to be evaluated soon, by applying DNA- and RNA-sequencing methodologies and deconvolution analyses.

(2) The diagnostic performance of gfDNA alone was only moderate, and the study did not explore potential improvements through combination with established biomarkers. Methodological limitations include a lack of control for pre-analytical variables, the absence of longitudinal data, and imbalanced group sizes, which may affect the robustness and generalizability of the results.

Reviewer #2 is correct that this investigational study was not designed to assess the diagnostic potential of gfDNA. Instead, its primary contribution is to provide useful prognostic information. In this regard, we have not yet explored combining gfDNA with other clinically well-established diagnostic biomarkers. We do acknowledge this current limitation as a logical follow-up that must be investigated in the near future.

Moreover, we collected a substantial number of pre-analytical variables within the limitations of a study involving over 1,000 subjects. Longitudinal samples and data were not analyzed here, as our aim was to evaluate prognostic value at diagnosis. Although the groups are imbalanced, this accurately reflects the real-world population of a large endoscopy center within a dedicated cancer facility. Subjects were invited to participate and enter the study before sedation for the diagnostic EGD procedure; thus, samples were collected prospectively from all consenting individuals.

Finally, to maintain a large, unbiased cohort, we did not attempt to balance the groups, allowing analysis of samples and data from all patients with compatible diagnoses (please see Results: Patient groups and diagnoses).

(3) Additionally, key methodological details were insufficiently reported, and the ROC analysis lacked comprehensive performance metrics, limiting the study's clinical applicability.

We are grateful for this useful suggestion. In the current version, each ROC curve (Supplementary Figures 1A and 1B) now includes the top 10 gfDNA thresholds, along with their corresponding sensitivity and specificity values (please see Suppl. Table 1). The thresholds are ordered from-best-to-worst based on the classic Youden’s J statistic, as follows:

Youden Index = specificity + sensitivity – 1 [Youden WJ. Index for rating diagnostic tests. Cancer 3:32-35, 1950. PMID: 15405679]. We have made an effort to provide all the key methodological details requested, but we would be glad to add further information upon specific request.

eLife Assessment

This study reports important findings about the nature of feedback to primary visual cortex (V1) during object recognition. The state-of-the-art functional MRI evidence for the main claims is solid, and the combination of high-resolution fMRI with MEG yields significant insight into neural mechanisms. The findings presented here are relevant to a number of scientific fields such as object recognition, categorisation and predictive coding.

Reviewer #1 (Public review):

This study examines the spatiotemporal properties of feedback signals in the human brain during an object discrimination task. Using 7T fMRI and MEG, the authors show that task-relevant object category information can be decoded from both deep and superficial layers of V1, originating from occipito-temporal and posterior parietal cortices. In contrast, task-irrelevant category feedback does not appear in V1, even when the same objects are foveally presented. Low-level orientation information, however, is decodable from V1 regardless of task relevance and is supported by recurrence with occipito-temporal regions. These findings suggest that category decoding in V1 depends on task-driven feedback rather than feedforward visual features.

Strengths

This study leverages two advanced neuroimaging modalities attempting to connect object recognition across cortical layer and whole-brain levels. The revised manuscript strengthens the connection between the fMRI and MEG components.<br /> It also demonstrates that a peripheral object discrimination task is effective for isolating feedforward and feedback signals using 7T fMRI.<br /> It is particularly notable that no low-level features were fed back to V1's superficial layers in the peripheral object discrimination task. The authors further show that high- and low-level feedback to the foveal V1 are comparable in strength, supporting the idea that the superficial layer in V1 selectively represents task-relevant content.

Weaknesses

One alternative explanation for the absence of task-irrelevant category decoding in the foveal task could be that feedback enhancement may be required to decode complex features from V1 (compared to a coarse orientation feature). It would be informative to test whether the findings hold if the categorical boundary were defined through a low level feature other than orientation (e.g., frequency) (e.g. Ester, Sprague and Serences, 2020).

I would like to echo the concerns raised by the other reviewer regarding multiple comparisons correction. It is important to apply correction procedures, especially given the number of statistical tests performed across brain regions where strict a priori hypotheses are unlikely. In the case of cluster-based statistics, the manuscript should clearly specify both the cluster-forming threshold and the significance threshold used for comparing true cluster masses to the shuffled distribution.

Conclusion

Overall, the results support the study's conclusions. This work addresses a timely question in object categorization and predictive coding-specifically, how feedback signals vary in content and timing across cortical layers.

Reviewer #2 (Public review):

Summary:

This manuscript reports high-resolution functional MRI data and MEG data revealing additional mechanistic information about an established paradigm studying how foveal regions of primary visual cortex (V1) are involved in processing peripheral visual stimuli. Because of the retinotopic organization of V1, peripheral stimuli should not evoke responses in the regions of V1 that represent stimuli in the center of the visual field (the fovea). However, functional MRI responses in foveal regions do reflect the characteristics of peripheral visual stimuli - this is a surprising finding first reported in 2008. The present study uses fMRI data with sub-millimeter resolution to study the how responses at different depths in the foveal gray matter do or don't reflect peripheral object characteristics during 2 different tasks: one in which observers needed to make detailed judgments about object identity, and one in which observers needed to make more coarse judgments about object orientation. FMRI results reveal interesting and informative patterns in these two conditions. A follow-on MEG study yields information about the timing of these responses. Put together, the findings settle some questions in the field and add new information about the nature of visual feedback to V1.

Strengths:

(1) Rigorous and appropriate use of "laminar fMRI" techniques.

(2) The introduction does an excellent job of contextualizing the work.

(3) Control experiments and analyses are designed and implemented well

Weaknesses:

(1) The use of the term "low order" to describe object orientation is potentially confusing. During review, the authors considered this issue and responded that they would continue with the use of the term low-order to describe object orientation because a low-pass spatial frequency filter would provide object orientation information. This is certainly a reasonable perspective; nonetheless, this reviewer thinks spatial frequencies that low are not readily represented by neurons in early visual cortex and it is common to use "low-order" to refer to features extracted in early visual areas, so I think this causes confusion.

(2) The methods contain a nice description of the methods for "correcting the vascular-related signals". I'm guessing this is the method that removed, e.g., 22% of foveal voxels (previous paragraph), but it's not entirely clear whether the voxel selection methods described in the "correcting the vascular-related signals" are describing the same processing step referred to in the previous paragraph as "a portion of voxels was removed based on large vein distribution".

(3) It is quite difficult to perform laminar analyses across multiple visual areas because distortion compensation is not perfect and registration of functional to anatomical data will always be a bit better in some places and a bit worse in others. An ideal manuscript would include some images showing registration quality in V1, LOC, and IPS regions for a few different participants, or include some kind of quality metric indicating the confidence in depth assignments in different regions.

(4) For the decoding analysis, it would be helpful to have more information about how samples were defined for each condition -- were the beta values for entire blocks used as samples for each condition, or were separate timepoints during a block used in the SVM as repeated samples for each condition?

Author response:

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

Reviewer #1 (Public review):

(1.1) The authors argue that low-level features in a feedback format could be decoded only from deep layers of V1 (and not superficial layers) during a perceptual categorization task. However, previous studies (Bergman et al., 2024; Iamshchinina et al., 2021) demonstrated that low-level features in the form of feedback can be decoded from both superficial and deep layers. While this result could be due to perceptual task or highly predictable orientation feature (orientation was kept the same throughout the experimental block), an alternative explanation is a weaker representation of orientation in the feedback (even before splitting by layers there is only a trend towards significance; also granger causality for orientation information in MEG part is lower than that for category in peripheral categorization task), because it is orthogonal to the task demand. It would be helpful if the authors added a statistical comparison of the strength of category and orientation representations in each layer and across the layers.

We agree that the strength of feedback information is related to task demand. Specifically, we would like to highlight the relationship between task demand and feedback information in the superficial layer. Previous studies have shown that foveal feedback information is observed only when the task requires the identity information of the peripheral objects (Williams et al., 2008; Fan et al., 2016; Yu and Shim, 2016). In this study, we found that the deep layer represented both orientation and categorical feedback information, while the superficial layer only represented categorical information. This suggests that feedback information in the superficial layer may be related to (or enhanced by) the task demands. In other words, if the experimental design required participants to discriminate orientation rather than object identity, we would expect stronger orientation information in foveal V1 and significant decoding performance of orientation feedback information in the superficial layer of foveal V1. This assumption is consistent with the anatomical connections of the superficial layer, which not only receives feedback connections but also sends outputs to higher-level regions for further processing. This is also consistent with Iamshchinina et al.’s observation that, when orientation information had to be mentally rotated and reported (i.e., task-relevant), it was observed in both the superficial and deep layers of V1. Bergmann et al. observed illusory color information in the superficial layer of V1, which may reflect a combination of lateral propagation and feedback mechanisms in the superficial layer that support visual filling-in phenomena. We have revised the discussion in the manuscript: In other words, if the experimental design required participants to discriminate orientation rather than object identity, we would expect stronger orientation information in foveal V1 and significant decoding performance of orientation feedback information in the superficial layer of foveal V1. Recent studies (Iamshchinina et al., 2021; Bergman et al., 2024) have also highlighted the relationship between feedback information and neural representations in V1 superficial layer.

To further demonstrate the laminar profiles of low- and high-order information, we have re-analyzed the data and added more fine-scale laminar profiles with statistical comparisons in the revised manuscript. The results again showed significant neural decoding performances in the deep layer of both category and orientation information, and only significant decoding performances of category information in the superficial layer.

(1.2) The authors argue that category feedback is not driven by low-level confounding features embedded in the stimuli. They demonstrate the ability to decode orientations, particularly well represented by V1, in the absence of category discrimination. However, the orientation is not a category-discriminating feature in this task. It could be that the category-discriminating features cannot be as well decoded from V1 activity patterns as orientations. Also, there are a number of these category discriminating features and it is unclear if it is a variation in their representational strength or merely the absence of the task-driven enhancement that preempts category decoding in V1 during the foveal task. In other words, I am not sure whether, if orientation was a category-specific feature (sharpies are always horizontal and smoothies are vertical), there would still be no category decoding.

The low-order features mentioned in the manuscript refer to visual information encoded intrinsically in V1, independent of task demands. In the foveal experiment, the task is to discriminate the color of fixation, which is unrelated to the category or orientation of the object stimuli. The results showed that only orientation information could be decoded from foveal V1. This indicates that low-order information, such as orientation, is strongly and automatically encoded in V1, even when it is irrelevant to the task. Meanwhile, category information could not be decoded, indicating that category information relies on feedback signals driven by attention or the task to the objects, both of which are absent in the fixation task. Other evidence indicates that category feedback is not driven by low-level features intrinsically encoded in V1. First, the laminar profiles of these two types of feedback information differ considerably (see response to 1.1). Second, only category feedback information was correlated with behavioral performance (MEG experiment). These findings demonstrate that category feedback information is task-driven and differs from the automatically encoded low-order information in foveal V1. The reviewer expressed some uncertainty that, whether “if orientation was a category-specific feature (sharpies are always horizontal and smoothies are vertical), there would still be no category decoding”. Our data showed that orientation could be automatically decoded in V1, regardless of task demand. Thus, if orientation was a category-specific feature in the foveal task (i.e., sharpies are always horizontal and smoothies are always vertical), category decoding would be successful in V1. However, in this scenario, the orientation and other shape features are not independent, thus preventing us to find out whether non-orientation shape features could be decoded in V1.  

Reviewer #2 (Public review):

(2.1) While not necessarily a weakness, I do not fully agree with the description of the 2 kinds of feedback information as "low-order" and "high-order". I understand the motivation to do this - orientation is typically considered a low-level visual feature. But when it's the orientation of an entire object, not a single edge, orientation can only be defined after the elements of the object are grouped. Also, the discrimination between spikies and smoothies requires detecting the orientations of particular edges that form the identifying features. To my mind, it would make more sense to refer to discrimination of object orientation as "coarse" feature discrimination, and orientation of object identity as "fine" feature discrimination. Thus, the sentence on line 83, for example, would read "Interestingly, feedback with fine and coarse feature information exhibits different laminar profiles.".

We agree that the object orientation (invariant to object category or identity) is defined on a larger spatial scale than the local orientation features such as local edges, however, in this sense, the object orientation is a coarse feature. In contrast, the category-defining information is mainly contributed by the local shape information (i.e., little cubes vs. bumps), which is more fine-scale information. One way to look at this difference is that the object orientation information is mainly carried by low-spatial frequency information and will survive low-pass filtering, hence “coarse”; while the object category information would largely be lost if the objects underwent low-pass spatial filtering.

We believe the labeling words “low-order” and “high-order” are consistent with the typical use of these terms in the literature, referring to features intrinsically encoded in early visual cortex vs. in high level object sensitive cortical regions. The more important aspects of our results are in their differential engagement in feedforward vs. feedback processing, with low-order features automatically represented in the early visual cortex during feedforward processing while high-order features represented due to feedback processing. Results from the foveal fMRI experiment (Exp. 2) strongly support this assumption that, when objects were presented at the fovea and the task was a fixation color task irrelevant to object information, foveal V1 could only represent orientation information, not category information. Notably, there was a dramatic difference in decoding performance in foveal V1 between Exp.1 and Exp.2, which ruled out the argument that both orientation and category information were driven by local edge information represented in V1.

(2.2) Figure 2 and text on lines 185, and 186: it is difficult to interpret/understand the findings in foveal ROIs for the foveal control task without knowing how big the ROI was. Foveal regions of V1 are grossly expanded by cortical magnification, such that the central half-degree can occupy several centimeters across the cortical surface. Without information on the spatial extent of the foveal ROI compared to the object size, we can't know whether the ROI included voxels whose population receptive fields were expected to include the edges of the objects.

The ROI of foveal V1 was defined using data from independent localizer runs. In each localizer run, flashing checkerboards of the same size as the objects in the task runs were presented at the fovea or in the periphery. The ROI of foveal V1 was identified as the voxels responsive to the foveal checkerboards. In other words, The ROI of foveal V1 included the voxels whose population receptive fields covered the entire object in the foveal visual field.

We included a figure in the revised manuscript comparing the activation maps induced by the foveal object stimulus in the task runs with the ROI coverage defined by the localizer runs. 

(2.3) Line 143 and ROI section of the methods: in order for the reader to understand how robust the responses and analyses are, voxel counts should be provided for the ROIs that were defined, as well as for the number (fraction) of voxels excluded due to either high beta weights or low signal intensity (lines 505-511).

In the revised manuscript, we have included the number of voxels in each ROI and the criteria for voxel selection:

For each ROI, the number of voxels depended on the size of the activated region, as estimated from the localizer data. The numbers are as follows: foveal V1, 2185 ± 389; peripheral V1, 1294± 215; LOC, 3451 ± 863; and pIPS, 5154 ± 1517. To avoid the signals of large vessels, a portion of voxels was removed based on the distribution of large vessels: V1 foveal, 22.5% ± 6.6%; V1 peripheral, 6.8% ± 3.9%; LOC, 16.1% ± 8.1% ; and pIPS, 5.1% ± 3.2%. For the decoding analysis, the top 500 responsive voxels in each ROI were selected to balance the voxel numbers across different ROIs for training and testing the decoder.

(2.4) I wasn't able to find mention of how multiple-comparisons corrections were performed for either the MEG or fMRI data (except for one Holm-Bonferonni correction in Figure S1), so it's unclear whether the reported p-values are corrected.

For the fMRI results, there is strong evidence showing that feedback information is sent to the foveal V1 during a peripheral object task (Williams et al., 2008; Fan et al., 2016; Yu and Shim, 2016). In addition, anatomical and functional evidence shows that the superficial and deep layers of V1 receive feedback information during visual processing. Therefore, in the current study, we specifically examined two types of feedback information in the superficial and deep layers of foveal V1, and did not apply multiple-comparison correction to the decoding results.

Regarding the MEG results, since we did not have a strong prior about when feedback information would arrive in the foveal V1, a cluster-based permutation method was used to correct for multiple comparisons in each time course. Specifically, for each time point, the sign of the effect for each participant was randomly flipped 50000 times to obtain the null hypothesis distribution for each time point. Clusters were defined as continuous significant time points in the real and flipped time series, and the effects in each cluster were summed to create a cluster-based effect. The most significant cluster-based effect in each flipped time series was then used to generate the corrected null hypothesis distribution.

We included these clarifications in Significance testing part of the revised manuscript.

Reviewer #1 (Recommendations for the authors):

It would be helpful if the authors could elaborate more on the fMRI decoding results in higher-order visual areas in the Discussion (there are recent studies also investigating higher-order visual areas (Carricarte et al., 2024) and associative areas (Degutis et al., 2024)) and relate it to the MEG information transmission results between the areas overlapping with the regions recorded in the fMRI part of the study.

We have discussed the fMRI decoding results in the LOC and IPS in the revised manuscript: 

In the current study, fMRI signals from early visual cortex and two high-level brain regions (LOC and pIPS) were recorded. Neural dynamics of these regions were extracted from MEG signals. Decoding analyses based on fMRI and MEG signals consistently showed that object category information could be decoded from both regions. These findings raise an important question:  Further Granger causality analysis indicates that the feedback information in foveal V1 was mainly driven by signals from the LOC. Layer-specific analysis showed that category information could be decoded in the middle and superficial layers of the LOC. A reasonable interpretation of this result is that feedforward information from the early visual cortex was received by the LOC’s middle layer, then the category information was generated and fed back to foveal V1 through the LOC’s superficial layer. A recent study (Carricarte et al., 2024) found that, in object selective regions in temporal cortex, the deep layer showed the strongest fMRI responses during an imagery task. Together, the results suggest that the deep and superficial layers correspond to different feedback mechanisms. It is worth noting that other cortical regions may also generate feedback signals to the early visual cortex. The current study did not have simultaneously recorded fMRI signals from the prefrontal cortex, but it has been shown that feedback signals can be traced back to the prefrontal cortex during complex cognitive tasks, such as working memory (Finn et al., 2019; Degutis et al., 2024). Further fMRI studies with submillimeter resolution and whole-brain coverage are needed to test other potential feedback pathways during object processing.

The behavioral performance seems quite low (67%), could authors explain the reasons for it?

We designed the object stimuli to be difficult to distinguish on purpose. Some of our pilot data showed that the more involved the participants were in the peripheral object task, the easier the foveal feedback information was to decoded. It is reasonable to assume that if the peripheral objects were easily distinguishable, the feedback mechanism may not be fully recruited during object processing. Furthermore, since we were decoding category and orientation information rather than identity information, the difficulty of distinguishing two objects from the same category and with the same orientation would not affect the decoding of category and orientation information in the neural signals.

Reviewer #2 (Recommendations for the authors):

(1) Line 52: the meaning of the sentence starting with "However, ..." is not entirely clear. Maybe the word "while" is missing after the first comma?

(2) Line 224. If I'm understanding the rationale for the MEG analysis correctly, it was not possible to localize foveal regions, but the cross-location decoding analysis was used to approximate the strength and timing of feedback information. If this is the case, "neural representations in the foveal region" were not extracted.

(3) Figure 4. The key information is too small to see. The lines indicating where decoding performance was significant are quite thin but very important, and the text next to them indicating onset times of significant decoding is in such a small font size I needed to zoom in to 300% to read it (yes, my eyes are getting old and tired). Increasing the font size used to represent key information would be nice.

(4) Figure 4 caption. Line 270 describes the line color in the plots as yellow, but that color is decidedly orange to my eye.

(5) Line 340/341: Papers that define and describe feedback-receptive fields seem important to cite here:

Keller, A. J., Roth, M. M., & Scanziani, M. (2020). Feedback generates a second receptive field in neurons of the visual cortex. Nature, 582(7813), 545-549.

Kirchberger, L., Mukherjee, S., Self, M. W., & Roelfsema, P. R. (2023). Contextual drive of neuronal responses in mouse V1 in the absence of feedforward input. Science advances, 9(3), eadd2498.

(6) Lines 346-350: this sentence seems to have some missing or misused words, because the syntax isn't intact.

(7) Line 367: supports should be support.

We thank the reviewers for the comments and have corrected them in the manuscript.

eLife Assessment

This important study identifies a plant-derived metabolite, betulin, as an effective natural insecticide against aphids and uncovers its specific molecular target. The evidence is compelling, combining greenhouse and field efficacy trials with rigorous molecular, genetic, and electrophysiological approaches that converge on a conserved binding site in the aphid GABA receptor. While additional work is needed to fully assess potential off-target effects and ecological safety, the study provides a strong mechanistic foundation. These findings will be of interest to researchers in plant biology, chemical ecology, and sustainable pest management.

Reviewer #1 (Public review):

Wang, Junxiu et al. investigated the underlying molecular mechanisms of the insecticidal activity of betulin against the peach aphid, Myzus persicae. There are two important findings described in this manuscript: (a) betulin inhibits the gene expression of GABA receptor in the aphid, and (b) betulin binds to the GABA receptor protein, acting as an inhibitor. The first finding is supported by RNA-Seq and RNAi, and the second one is convinced with MST and electrophysiological assays. Further investigations on the betulin binding site on the receptor protein provided a fundamental discovery that T228 is the key amino acid residue for its affinity, thereby acting as an inhibitor, backed up by site-directed mutagenesis of the heterologously-expressed receptor in E. coli and by CRISPR-genome editing in Drosophila.

Comments on revisions:

All of my review comments have been addressed, and the manuscript has been revised accordingly.

Reviewer #2 (Public review):

Summary:

This important study shows that betulin from wild peach trees disrupts neural signaling in aphids by targeting a conserved site in the insect GABA receptor. The authors present a nicely integrated set of molecular, physiological, and genetic experiments to establish the compound's species-specific mode of action. While the mechanistic evidence is solid, the manuscript would benefit from a broader discussion of evolutionary conservation and potential off-target ecological effects.

Strengths:

The main strengths of the study lie in its mechanistic clarity and experimental rigor. The identification of a betulin-binding single threonine residue was supported by (1) site-directed mutagenesis and (2) functional assays. These experiments strongly support the specificity of action. Furthermore, the use of comparative analyses between aphids and fruit flies demonstrates an important effort to explore species specificity, and the integration of quantitative data further enhances the robustness of the conclusions.

Comments on revisions:

The revision satisfactorily addresses my concerns on evolutionary context, methodological clarity, and ecological risk.

Author response:

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

Reviewer #1 (Public review): 

Wang, Junxiu et al. investigated the underlying molecular mechanisms of the insecticidal activity of betulin against the peach aphid, Myzus persicae. There are two important findings described in this manuscript: (a) betulin inhibits the gene expression of GABA receptor in the aphid, and (b) betulin binds to the GABA receptor protein, acting as an inhibitor. The first finding is supported by RNA-Seq and RNAi, and the second one is convinced with MST and electrophysiological assays. Further investigations on the betulin binding site on the receptor protein provided a fundamental discovery that T228 is the key amino acid residue for its affinity, thereby acting as an inhibitor, backed up by site-directed mutagenesis of the heterologously-expressed receptor in E. coli and by CRISPR-genome editing in Drosophila.

Although the manuscript does have strengths in principle, the weaknesses do exist: the manuscript would benefit from more comprehensive analyses to fully support its key claims in the manuscript. In particular:

(1) The Western blotting results in Figure 5A & B appear to support the claim that betulin inhibits GABR gene expression (L26), as a decrease in target protein levels is often indicative of suppressed gene expression. The result description for Figure 5A & B is found in L312-L316, within Section 3.6 ("Responses of MpGABR to betulin"), where MST and voltage-clamp assays are also presented. It seems the observed decrease in MpGABR protein content is due to gene downregulation, rather than a direct receptor protein-betulin interaction. However, this interpretation lacks discussion or analysis in either the corresponding results section or the Discussion. In contrast, Figures 5C-F are specifically designed to illustrate protein-betulin interactions. Presenting Figure 5A & B alongside these panels might lead to confusion, as they support distinct claims (gene expression vs. protein binding/inhibition). Therefore, I recommend moving Figure 5A & B either to the end of Figure 3 or to a separate figure altogether to improve clarity and logical flow. A minor point in the Western blotting experiment is that although GAPDH was used as a reference protein, there is no explanation in the corresponding M&M section.

We thank the reviewer for the concise and accurate summary and appreciate the constructive feedback on the article’s strengths and weaknesses.

(A) According to your suggestion, the original Figure 5A and B have been inserted into Figure 3, following Figure 3D. The original Figure 3E-I has been saved as a new figure, to illustrate the RNAi assay.

(b) “GAPDH was used as a reference protein” has been supplied in the M&M section, see

Line 209.

(2) The description of the electrophysiological recording experiment is unclear regarding the use of GABA. I didn't realize that GABA, the true ligand of the GABA receptor, was used in this inhibition experiment until I reached the Results section (L321), which states, "In the presence of only GABA, a fast inward current was generated." Crucially, no details are provided on the experiment itself, including how GABA was applied (e.g., concentration, duration, whether GABA was treated, followed by betulin, or vice versa). This information is essential for reproducibility. Please ensure these details are thoroughly described in the corresponding M&M section.

We thank the reviewer for the valuable comments.

(a) Detailed information on how to apply GABA has been added to the corresponding M&M section (Lines 260-263): After 3 days of incubation, the oocytes were used for electrophysiological recording. GABA was dissolved in 1 × Ringer's solution to prepare 100 µM GABA solution. Subsequently, the 100 µM GABA solutions containing different concentrations of betulin (0, 5, 10, 20, 40, 80, 160, 320 µM) were used to perfuse the oocytes.

(b) Additionally, we also checked other contents of M&M section to ensure that sufficient detail has been supplied.

(3) The phylogenetic analysis, particularly concerning Figures 4 and 6B, needs significant attention for clarity and representativeness. First, your claim that MpGABR is only closely related to CAI6365831.1 (L305-L310) is inconsistent with the provided phylogenetic tree, which shows MpGABR as equally close to Metopolophium dirhodum (XP_060864885.1) and Acyrthosiphon pisum (XP_008183008.2). Therefore, singling out only Macrosiphum euphorbiae (CAI6365831.1) is not supported by the data. Second, the representation of various insect orders is insufficient. All 11 sequences in the Hemiptera category (in both Figure 4 and Figure 6B) are exclusively from the Aphididae family. This small subset cannot represent the highly diverse Order Hemiptera. Consequently, statements like "only THR228 was conserved in Hemiptera" (L338), "The results of the sequence alignment revealed that only THR228 was conserved in Hemiptera" (L430), or "THR228... is highly conserved in Hemiptera" (L486) are not adequately supported. Third, similar concerns apply to the Diptera order, which includes 10 Drosophila and 2 mosquito samples (not diverse or representative enough), and likely to other orders as well. Thereby, the Figure 6B alignment should be revised accordingly to reflect a more accurate representation or to clarify the scope of the analysis. Fourth, there's a discrepancy in the phylogenetic method used: the M&M section (L156) states that MEGA7, ClustalW, and the neighbor-joining method were used, while the Figure 4 caption mentions that MEGA X, MUSCLE, and the Maximum likelihood method were employed. This inconsistency needs to be clarified and made consistent throughout the manuscript. Fifth, I have significant concerns about the phylogenetic tree itself (Figure 4). A small glitch was observed at the Danaus plexippus node, which raises suspicion regarding potential manipulation after tree construction. More critically, the tree, especially within Coleoptera, does not appear to be clearly resolved. I am highly concerned about whether all included sequences are true GABR orthologs or if the dataset includes partial or related sequences that could distort the phylogeny. Finally, for Figure 6B, both protein (XP_) and nucleotide (XM_) sequences were mix used. I recommend using the protein sequences instead of nucleotide sequences in this figure panel, as protein sequences are more directly informative.

We thank the reviewer for the careful reading and valuable comments.

(a) Firstly, according to your comments, phylogenetic analysis has been re-performed with more represent species from each Order (Fig. 5 and Fig. 7B). The results revealed that only THR228 was conserved across 11 species in the Aphididae family of Hemiptera. Therefore, the expressions like "only THR228 was conserved in Hemiptera" have been revised to “among the four residues, only THR228 was conserved across 11 species in the Aphididae family of Hemiptera” (Line 106, Line 369, Line 477, and Lines 563-564).

(b) We have modified the description of Fig. 5 (the original Fig. 4): MpGABR  (XP_022173711.1) was found to be genetically closely related to CAI6365831.1 from Macrosiphum euphorbiae, XP 008183008.2 from Acyrthosiphon pisum, and XP 060864885.1 from Metopolophium dirhodum (Fig. 5 and Table S6). See Lines 342-346.

(c) Phylogenetic analysis was performed using MEGA7 with multiple amino acid sequence alignment (ClustalW) and the neighbor-joining method. We have revised the Fig. 5 (the original Fig. 4) caption to make it accurate and consistent throughout the manuscript.

(d) We are sorry about the small glitch at the Danaus plexippus node. Actually, after the phylogenetic tree was constructed, it was imported in Adobe Illustration for coloring and classification annotation. There may have been operational errors during the process of resizing the image, resulting in the occurrence of the small glitch. Besides, the unclear clustering of Coleoptera may be due to improper regulation of distance (pixels) of branch from nodes. Again, thanks for your careful reading. We have rebuilt the phylogenetic tree.

(e) Based on your suggestion, the sequence IDs have been unified as the protein sequence IDs (Fig. 5, Fig. 7B and Table S6)

(4) The Discussion section requires significant revision to provide a more insightful and interpretative analysis of the results. Currently, much of the section primarily restates findings rather than offering deeper discussion. For instance, L409-L419 restate the results, followed by the short sentence "Collectively, these results suggest that betulin may have insecticidal effects on aphids by inhibiting MpGABR expression". It could be further expanded to make it beneficial to elaborate on proposed mechanisms by which gene expression might be suppressed, including any potential transcription factors involved. In contrast, while L422-L442 also initially summarize results, the subsequent paragraph (L445-L472) effectively discusses the potential mechanisms of inhibitory action and how mortality is triggered, which is a good model for other parts of the section. However, all the discussion ends up with a short statement, "implying that betulin acts as a CA of MpGABR" (L472), which appears to be a leap. The inference that betulin acts as a competitive antagonist (CA) is solely based on the location of its extracellular binding site, which does not exactly overlap with the GABA binding site. It needs stronger justification or actually requires further experimental validation. The authors should consider rephrasing this statement to acknowledge the need for additional studies to definitively confirm this mechanism of action.

We appreciate the reviewer's careful reading and valuable feedback, which will certainly enhance the quality of our manuscript.

(a) Possible reasons for the effect of betulin on MpGABR expression have been discussed in our manuscript (Lines 455-466): The regulation of gene expression is sophisticated and delicate (Pope and Medzhitov 2018). The regulatory network controlling GABR expression remains unclear. In adult rats, epileptic seizures has been reported to increase the levels of brain-derived neurotrophic factor (BDNF), which in turn prompted the transcription factors CREB and ICER to reduce the gene expression of the GABR α1 subunit (Lund et al. 2008). In Drosophila, it has been demonstrated that WIDE AWAKE, which regulated the onset of sleep, interacted with the GABR and upregulated its expression level (Liu et al. 2014). In Drosophila brain, circular RNA circ_sxc was found to inhibit the expression of miR-87-3p in the brain through sponge adsorption, thereby regulating the expression of neurotransmitter receptor ligand proteins, including GABR, and ensuring the normal function of synaptic signal transmission in brain neurons (Li et al. 2024). However, it remains unclear how betulin reduces the expression of MpGABR, and further research is needed.

(b) In the Discussion section, we acknowledged the need for further research to ultimately confirm the mechanism by which betulin competes with GABA for binding to MpGABR (Lines 532-535): Although the mechanism by which betulin competes with GABA for binding to MpGABR requires further experimental validation, our work may have provided a novel target for developing insecticides.

(c) Besides, we have added the discussion of the sensitivity of GABA receptor to betulin in Discussion section (Lines 491-501): Studies on key amino acids that are crucial for GABR function has primarily focused on transmembrane regions. For instance, based on the mutational research and Drosophila GABR modeling approach, multiple key amino acids were identified as insecticide targets in the transmembrane domain (Nakao and Banba 2021). Guo et al. proposed that amino acid substitutions in the transmembrane domain 2 contribute to terpenoid insensitivity during plant-insect coevolution (Guo et al. 2023). However, these studies have neglected the extracellular domain. Our study signified that betulin targets the THR228 site in the extracellular domain of MpGABR, which is conserved only in the Aphididae family. Therefore, betulin is speculated to be a specific insecticidal substance evolved by plants in response to aphid infestation. Besides, further verification is needed to determine whether betulin is toxic to other insect species.

(d) Furthermore, the discussion of potential ecological risks of deploying betulin as a bioinsecticide has been elaborated in our manuscript (Lines 538-553): The development of bioinsecticides should not only focus on the toxic effects of active substance on target organisms, but also on their influence on the ecosystem (Haddi et al. 2020). Although our results indicate that betulin has specific toxicity to aphids, previous studies have reported that betulin and its derivatives had effects on Plutella xylostella L. (Huang et al. 2025), Aedes aegypti (de Almeida Teles et al. 2024), and Drosophila melanogaster (Lee and Min 2024). Therefore, further research is needed to determine whether there are other insecticidal mechanisms or off target effects of betulin. Additionally, betulin exhibits a wide range of pharmacological activities (Amiri et al. 2020), which have been used to treat various diseases, such as cancer (Lv 2023), glioblastoma (Li et al. 2022), inflammation (Szlasa et al. 2023) and hyperlipidemia (Tang et al. 2011). Before applying betulin in the field, it is necessary to fully verify and consider whether betulin has any impact on farmers' health. Furthermore, will betulin cause residue or diffusion in the process of field application? Will long-term application promote the evolution of resistance to aphids or other insects? These issues also need further experimental verification. In summary, before any field application, further research is needed on the environmental behavior, degradation process, and safety of betulin.

Reviewer #2 (Public review):

Summary:

This important study shows that betulin from wild peach trees disrupts neural signaling in aphids by targeting a conserved site in the insect GABA receptor. The authors present a nicely integrated set of molecular, physiological, and genetic experiments to establish the compound's species-specific mode of action. While the mechanistic evidence is solid, the manuscript would benefit from a broader discussion of evolutionary conservation and

potential off-target ecological effects.

Strengths:

The main strengths of the study lie in its mechanistic clarity and experimental rigor. The identification of a betulin-binding single threonine residue was supported by (1) site-directed mutagenesis and (2) functional assays. These experiments strongly support the specificity of action. Furthermore, the use of comparative analyses between aphids and fruit flies demonstrates an important effort to explore species specificity, and the integration of quantitative data further enhances the robustness of the conclusions.

Weaknesses:

There are several important limitations that need to be addressed. The manuscript does not explore whether the observed sensitivity to betulin reflects a broadly conserved feature of GABA receptors across animal lineages or a more lineage-specific adaptation. This evolutionary context is crucial for understanding the broader significance of the findings.

In addition, while the compound's aphicidal effect is well established, the potential for off-target effects in non-target organisms - especially vertebrates - remains unaddressed, despite prior evidence that betulin interacts with mammalian GABAa receptors. There is little discussion on the ecological or environmental safety of exogenous betulin application, such as persistence, degradation, or exposure risks.

We sincerely thank the reviewer for the time and effort dedicated to our manuscript's detailed review and assessment. The revision suggestions were constructive, and we have provided a point-by-point response to address them.

(a) Briefly introduce the evolutionary conservation of GABA receptors has been added in the Introduction (Lines 90-98): Previous study has proposed that vertebrate and human GABR genes maintain a broad and conservative gene clustering pattern, while in invertebrates, this pattern is missing, indicating that these gene clusters formed early in vertebrate evolution and were established after diverging from invertebrates. Notably, invertebrates each possess a unique GABR gene pair, which are homologous with human GABR α and β subunits, suggesting that the existing GABR gene cluster evolved from an ancestral α - β subunit gene pair (Tsang et al. 2006). During the coevolution of plants and insects, the duplications and amino acid substitutions in GABR may be beneficial for the adaptation to insecticides and terpenoid compounds (Guo et al. 2023).

(b) We have added the discussion of the sensitivity of GABA receptor to betulin in Discussion section (Lines 491-501): Studies on key amino acids that are crucial for GABR function has primarily focused on transmembrane regions. For instance, based on the mutational research and Drosophila GABR modeling approach, multiple key amino acids were identified as insecticide targets in the transmembrane domain (Nakao and Banba 2021). Guo et al. proposed that amino acid substitutions in the transmembrane domain 2 contribute to terpenoid insensitivity during plant-insect coevolution (Guo et al. 2023). However, these studies have neglected the extracellular domain. Our study signified that betulin targets the THR228 site in the extracellular domain of MpGABR, which is conserved only in the Aphididae family. Therefore, betulin is speculated to be a specific insecticidal substance evolved by plants in response to aphid infestation. Besides, further verification is needed to determine whether betulin is toxic to other insect species.

(c) The discussion of potential ecological risks of deploying betulin as a bioinsecticide has been elaborated in our manuscript (Lines 538-553): The development of bioinsecticides should not only focus on the toxic effects of active substance on target organisms, but also on their influence on the ecosystem (Haddi et al. 2020). Although our results indicate that betulin has specific toxicity to aphids, previous studies have reported that betulin and its derivatives had effects on Plutella xylostella L. (Huang et al. 2025), Aedes aegypti (de Almeida Teles et al. 2024), and Drosophila melanogaster (Lee and Min 2024). Therefore, further research is needed to determine whether there are other insecticidal mechanisms or off target effects of betulin. Additionally, betulin exhibits a wide range of pharmacological activities (Amiri et al. 2020), which have been used to treat various diseases, such as cancer (Lv 2023), glioblastoma (Li et al. 2022), inflammation (Szlasa et al. 2023) and hyperlipidemia (Tang et al. 2011). Before applying betulin in the field, it is necessary to fully verify and consider whether betulin has any impact on farmers' health. Furthermore, will betulin cause residue or diffusion in the process of field application? Will long-term application promote the evolution of resistance to aphids or other insects? These issues also need further experimental verification. In summary, before any field application, further research is needed on the environmental behavior, degradation process, and safety of betulin.

Reviewer #1 (Recommendations for the authors):

(1) L28 Provide the full name of MST.

Thanks for your suggestion. The full name of MST, microscale thermophoresis, has been supplied.

(2) L87 in the Order Hemiptera.

Thanks for your suggestion. Corrected.

(3) L99 "Leaf bioassay" would be better to differentiate the greenhouse and field bioassays.

Thanks for your suggestion. Corrected.

(4) L104 It should be 7 doses, including the "0 mg/mL" control.

Thanks for your suggestion. Corrected.

(5) L104 Since the LC50 of pymetrozine is 1.0612 mg/mL, a wider range of doses should have been tested compared to the dose range of betulin.

Thanks for your comment.

(a) Firstly, seven doses (0, 0.0625, 0.125, 0.25, 0.5, 1, and 2 mgmL^-1) were set to calculate the LC50 of betulin and pymetrozine. Since the LC50 values of betulin and pymetrozine are 0.1641 and 1.0612 mgmL^–1, respectively, which are within the set range, indicating that the set dose range is reasonable and the LC50 values of betulin and pymetrozine are reliable.

(b) To compare the control effects of betulin and pymetrozine against M. persicae, LC50 of betulin (0.1641 mgmL^-1) and pymetrozine (1.0612 mgmL^-1) were used to treat M. persicae.

(6) L109 Greenhouse and field bioassays.

Thanks for your suggestion. Corrected.

(7) L112 Tween-80 and acetone in L103. Keep the order consistent throughout the manuscript.

Thanks for your suggestion. Corrected.

(8) L122 Mortality was recorded at 1, 5, 9, and 14 days after treatment. Revise the other similar mistakes throughout the manuscript (e.g. L250, L254, L255, L256, L259, etc.).

Thanks for your suggestion. Corrected.

(9) L126 apterous instead of wingless (keep a consistent expression).

Thanks for your suggestion. Corrected.

(10) L138 Primer Premier?

Thanks for your comment. Corrected.

(11) L141 Add RPS18 primers in Table S2.

Thanks for your comment. Corrected.

(12) L155 MEGA7 vs. MEGAX (as described in the Figure 4 caption).

Thanks for your comment. Corrected.

(13) L156 NJ method vs. ML method (as described in the Figure 4 caption).

Thanks for your comment. Corrected.

(14) L157 2.7. RNAi assay (Remove "In vitro" and re-number the following M&M sections accordingly).

Thanks for your comment. Corrected.

(15) L163 Add dsGFP primers in Table S2.