Author response:

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

We thank you for sending our manuscript for the second round of review.  We are encouraged by the comments from reviewer #2 that our supplementary work on naïve T cells and antibody blockade work satisfied their previous concerns and is important for our work.

The Editors raised concerns that we have shared preliminary data on Nrn1 and AMPAR double knockout mice.  We apologize for our enthusiasm for these studies.  Because of the publication model by eLife, we shared that data not because we needed to persuade the reviewer for publication purposes but rather to agree with the reviewer that the molecular target of Nrn1 is important, and we are progressing in understanding this subject.

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

To Reviewer #1:

Thank you for your thorough review and comments on our work, which you described as “the role of neuritin in T cell biology studied here is new and interesting.”.  We have summarized your comments into two categories: biology and investigation approach, experimental rigor, and data presentation.

Biology and Investigation approach comments:

(1) Questions regarding the T cell anergy model:

Major point “(4) Figure 1E-H. The authors assume that this immunization protocol induces anergic cells, but they provide no experimental evidence for this. It would be useful to show that T cells are indeed anergic in this model, especially those that are OVA-specific. The lack of IL-2 production by Cltr cells could be explained by the presence of fewer OVA-specific cells, rather than by an anergic status.”

T cell anergy is a well-established concept first described by Schwartz’s group. It refers to the hyporesponsive T cell functional state in antigen-experienced CD4 T cells (Chappert and Schwartz, 2010; Fathman and Lineberry, 2007; Jenkins and Schwartz, 1987; Quill and Schwartz, 1987).  Anergic T cells are characterized by their inability to expand and to produce IL2 upon subsequent antigen re-challenge. In this paper, we have borrowed the existing in vivo T cell anergy induction model used by Mueller’s group for T cell anergy induction (Vanasek et al., 2006).  Specifically, Thy1.1+ Ctrl or Nrn1-/- TCR transgenic OTII cells were co-transferred with the congenically marked Thy1.2+ WT polyclonal Treg cells into TCR-/- mice.  After anergy induction, the congenically marked TCR transgenic T cells were recovered by sorting based on Thy1.1+ congenic marker, and subsequently re-stimulation ex vivo with OVA323-339 peptide. We evaluated the T cell anergic state based on OTII cell expansion in vivo and IL2 production upon OVA323-339 restimulation ex vivo.  

“The authors assume that this immunization protocol induces anergic cells, but they provide no experimental evidence for this.”

Because the anergy model by Mueller's group is well established (Vanasek et al., 2006), we did not feel that additional effort was required to validate this model as the reviewer suggested. Moreover, the limited IL2 production among the control cells upon restimulation confirms the validity of this model.

“The lack of IL-2 production by Cltr cells could be explained by the presence of fewer OVAspecific cells, rather than by an anergic status”.

Cells from Ctrl and Nrn1-/- mice on a homogeneous TCR transgenic (OTII) background were used in these experiments. The possibility that substantial variability of TCR expression or different expression levels of the transgenic TCR could have impacted IL2 production rather than anergy induction is unlikely.

Overall, we used this in vivo anergy model to evaluate the Nrn1-/- T cell functional state in comparison to Ctrl cells under the anergy induction condition following the evaluation of Nrn1 expression, particularly in anergic T cells.  Through studies using this anergy model, we observed a significant change in Treg induction among OTII cells. We decided to pursue the role of Nrn1 in Treg cell development and function rather than the biology of T cell anergy as evidenced by subsequent experiments.

Minor points “(6) On which markers are anergic cells sorted for RNAseq analysis?”

Cells were sorted out based on their congenic marker marking Ctrl or Nrn1-/- OTII cells transferred into the host mice.  We did not specifically isolate anergic cells for sequencing.

(2) Question regarding the validity of iTreg differentiation model.

Major point: “(5) Figure 2A-C and Figure 3. The use of iTregs to try to understand what is happening in vivo is problematic. iTregs are cells that have probably no equivalent in vivo, and so may have no physiological relevance. In any case, they are different from pTreg cells generated in vivo. Working with pTreg may be challenging, that is why I would suggest generating data with purified nTreg. Moreover, it was shown in the article of Gonzalez-Figueroa 2021 that Nrn1-/- nTreg retained a normal suppressive function, which would not be what is concluded by the authors of this manuscript. Moreover, we do not even know what the % of Foxp3 cells is in the iTreg used (after differentiation and 20h of re-stimulation) and whether this % is the same between Ctlr and Nrn1 KO cells.”.

We thank Reviewer #1 for their feedback. While it is true that iTregs made in vitro and in vivo generated pTregs display several distinctions (e. g., differences in Foxp3 expression stability, for example), we strongly disagree with this statement by Revieweer#1 “The use of iTregs to try to understand what is happening in vivo is problematic. iTregs are cells that have probably no equivalent in vivo, and so may have no physiological relevance.”  The induced Treg cell (iTreg) model was established over 20 years ago (Chen et al., 2003; Zheng et al., 2002), and the model is widely adopted with over 2000 citations. Further, it has been instrumental in understanding different aspects of regulatory T cell biology (Hurrell et al., 2022; John et al., 2022; Schmitt and Williams, 2013; Sugiura et al., 2022).   

Because we have observed reduced pTreg generation in vivo, we choose to use the in vitro iTreg model system to understand the mechanistic changes involved in Treg cell differentiation and function, specifically, neuritin’s role in this process. We have made no claim that iTreg cell biology is identical to pTreg generated in vivo or nTreg cells. However, the iTreg culture system has proved to be a good in vitro system for deciphering molecular events involved in complex processes. As such, it remains a commonly used approach by many research groups in the Treg cell field (Hurrell et al., 2022; John et al., 2022; Sugiura et al., 2022). Moreover, applying the iTreg in vitro culture system has been instrumental in helping us identify the cell electrical state change in Nrn1-/- CD4 cells and revealed the biological link between Nrn1 and the ionotropic AMPA receptor (AMPAR), which we will discuss in the subsequent discussion. It is technically challenging to use nTreg cells for T cell electrical state studies due to their heterogeneous nature from development in an in vivo environment and the effect of manipulation during the nTreg cell isolation process, which can both affect the T cell electrical state.   

“Moreover, it was shown in the article of Gonzalez-Figueroa 2021 that Nrn1-/- nTreg retained a normal suppressive function, which would not be what is concluded by the authors of this manuscript.” 

We have also carried out nTreg studies in vitro in addition to iTreg cells. Similar to Gonzalez-Figueroa et al.'s findings, we did not observe differences in suppression function between Nrn1-/- and WT nTreg using the in vitro suppression assay. However, Nrn1-/- nTreg cells revealed reduced suppression function in vivo (Fig. 2D-L). In fact, Gonzalez-Figueroa et al. observed reduced plasma cell formation after OVA immunization in Treg-specific Nrn1-/- mice, implicating reduced suppression from Nrn1-/- follicular regulatory T (Tfr) cells. Thus, our observation of the reduced suppression function of Nrn1-/- nTreg toward effector T cell expansion, as presented in Fig. 2D-L, does not contradict the results from Gonzalez-Figueroa et al. Rather, the conclusions of these two studies agree that Nrn1 can play important roles in immune suppression observable in vivo that are not captured readily by the in vitro suppression assay.

“Moreover, we do not even know what the % of Foxp3 cells is in the iTreg used (after differentiation and 20h of re-stimulation) and whether this % is the same between Ctlr and Nrn1 KO cells.”

We have stated in the manuscript on page 7 line 208 that “Similar proportions of Foxp3+ cells were observed in Nrn1-/- and Ctrl cells under the iTreg culture condition, suggesting that Nrn1 deficiency does not significantly impact Foxp3+ cell differentiation”. In the revised manuscript, we will include the data on the proportion of Foxp3+ cells before iTreg restimulation.

(3) Confirmation of transcriptomic data regarding amino acids or electrolytes transport change

Minor point“(3) Would not it be possible to perform experiments showing the ability of cells to transport amino acids or electrolytes across the plasma membrane? This would be a more interesting demonstration than transcriptomic data.”

We appreciate Review# 1’s suggestion regarding “perform experiments showing the ability of cells to transport amino acids or electrolytes across the plasma membrane”.  We have indeed already performed such experiments corroborating the transcriptomics data on differential amino acid and nutrient transporter expression. Specifically, we loaded either iTreg or Th0 cells with membrane potential (MP) dye and measured MP level change after adding the complete set of amino acids (complete AA).  Upon entry, the charge carried by AAs may transiently affect cell membrane potential. Different AA transporter expression patterns may show different MP change patterns upon AA entry, as we showed in Author response image 1. We observed reduced MP change in Nrn1-/- iTreg compared to the Ctrl, whereas in the context of Th0 cells, Nrn1-/- showed enhanced MP change than the Ctrl. We can certainly include these data in the revised manuscript.

Author response image 1.

Membrane potential change induced by amino acids entry. a. Nrn1-/- or WT iTreg cells loaded with MP dye and MP change was measured upon the addition of a complete set of AAs. b. Nrn1-/- or WT Th0 cells loaded with MP dye and MP change was measured upon the addition of a complete set of AAs.

(4) EAE experiment data assessment

Minor point ”(5) Figure 5F. How are cells re-stimulated? If polyclonal stimulation is used, the experiment is not interesting because the analysis is done with lymph node cells. This analysis should either be performed with cells from the CNS or with MOG restimulation with lymph node cells.”

In the EAE study, the Nrn1-/- mice exhibit similar disease onset but a protracted non-resolving disease phenotype compared to the WT control mice.  Several reasons may contribute to this phenotype: 1. Enhanced T effector cell infiltration/persistence in the central nervous system (CNS); 2. Reduced Treg cell-mediated suppression to the T effector cells in the CNS; 3. Protracted non-resolving inflammation at the immunization site has the potential to continue sending T effector cells into CNS, contributing to persistent inflammation. Based on this reasoning, we examined the infiltrating T effector cell number and Treg cell proportion in the CNS.  We also restimulated cells from draining lymph nodes close to the inflammation site, looking for evidence of persistent inflammation.  When mice were harvested around day 16 after immunization, the inflammation at the local draining lymph node should be at the contraction stage.  We stimulated cells with PMA and ionomycin intended to observe all potential T effector cells involved in the draining lymph node rather than only MOG antigen-specific cells.  We disagree with Reviewer #1’s assumption that “This analysis should either be performed with cells from the CNS or with MOG restimulation with lymph node cells.”. We think the experimental approach we have taken has been appropriately tailored to the biological questions we intended to answer.

Experimental rigor and data presentation.

(1) data labeling and additional supporting data

Major points

(2) The authors use Nrn1+/+ and Nrn1+/- cells indiscriminately as control cells on the basis of similar biology between Nrn1+/+ and Nrn1+/- cells at homeostasis. However, it is quite possible that the Nrn1+/- cells have a phenotype in situations of in vitro activation or in vivo inflammation (cancer, EAE). It would be important to discriminate Nrn1+/- and Nrn1+/+ cells in the data or to show that both cell types have the same phenotype in these conditions too.

(3) Figure 1A-D. Since the authors are using the Nrp1 KO mice, it would be important to confirm the specificity of the anti-Nrn1 mAb by FACS. Once verified, it would be important to add FACS results with this mAb in Figures 1A-C to have single-cell and quantitative data as well.

Minor points  

(1) Line 119, 120 of the text. It is said that one of the most up-regulated genes in anergic cells is Nrn1 but the data is not shown.

(2) For all figures showing %, the titles of the Y axes are written in an odd way. For example, it is written "Foxp3% CD4". It would be more conventional and clearer to write "% Foxp3+ / CD4+" or "% Foxp3+ among CD4+".

(4) For certain staining (Figure 3E, H) it would be important to show the raw data, in addition to MFI or % values.

We can adapt the labeling and provide additional data, including Nrn1 staining on Treg cells and flow graphs for pmTOR and pS6 staining (Fig. 3H), as requested by Reviewer #1.

(2) Experimental rigor:

General comments:

“However, it is disappointing that reading this manuscript leaves an impression of incomplete work done too quickly.”

We were discouraged to receive the comment, “this manuscript leaves an impression of incomplete work done too quickly.” Our study of this novel molecule began without any existing biological tools such as antibodies, knockout mice, etc.  Over the past several years, we have established our own antibodies for Nrn1 detection, obtained and characterized Nrn1 knockout mice, and utilized multiple approaches to identify the molecular mechanism of Nrn1 function. Through the use of the in vitro iTreg system described in this manuscript, we identified the association of Nrn1 deficiency with cell electrical state change, potentially connected to AMPAR function. We have further corroborated our findings by generating Nrn1 and AMPAR T cell specific double knockout mice and confirmed that T cell specific AMPAR deletion could abrogate the phenotype caused by the Nrn1 deficiency (see Support Figure 2).  We did not include the double knockout data in the current manuscript because AMPAR function has not yet been studied thoroughly in T cell biology, and we feel this topic warrants examination in its own right.  However, the unpublished data support the finding that Nrn1 modulates the T cell electrical state and, consequently, metabolism, ultimately influencing tolerance and immunity.  In its current form, the manuscript represents the first characterization of the novel molecule Nrn1 in anergic cells, Tregs, and effector T cells. While this work has led to several exciting additional questions, we disagree that the novel characterization we have presented Is incomplete. We feel that our present data set, which squarely highlights Nrn1’s role as an important immune regulator while shedding unprecedented light on the molecular events involved, will be of considerable interest to a broad field of researchers.

“Multiple models have been used, but none has been studied thoroughly enough to provide really conclusive and unambiguous data. For example, 5 different models were used to study T cells in vivo. It would have been preferable to use fewer, but to go further in the study of mechanisms.”

We have indeed used multiple in vivo models to reveal Nrn1's function in Treg differentiation, Treg suppression function, T effector cell differentiation and function, and the overall impact on autoimmune disease. Because the impact of ion channel function is often context-dependent, we examined the biological outcome of Nrn1 deficiency in several in vivo contexts.  We would appreciate it if Reviewer#1 would provide a specific example, given the Nrn1 phenotype, of how to proceed deeper to investigate the electrical change in the in vivo models.

“Major points

(1) A real weakness of this work is the fact that in most of the results shown, there are few biological replicates with differences that are often small between Ctrl and Nrn1 -/-. The systematic use of student's t-test may lead to thinking that the differences are significant, which is often misleading given the small number of samples, which makes it impossible to know whether the distributions are Gaussian and whether a parametric test can be used. RNAseq bulk data are based on biological duplicates, which is open to criticism.”

We respectfully disagree with Reviewer #1 on the question of statistical power and significance to our work. We have used 5-8 mice/group for each in vivo model and 3-4 technical replicates for the in vitro studies, with a minimum of 2-3 replicate experiments. These group sizes and replication numbers are in line with those seen in high-impact publications. While some differences between Ctrl and Nrn1-/- appear small, they have significant biological consequences, as evidenced by the various Nrn1-/- in vivo phenotypes. Furthermore, we believe we have subjected our data to the appropriate statistical tests to ensure rigorous analysis and representation of our findings.

To Reviewer #2.

We thank Reviewer #2 for the careful review of the manuscript. We especially appreciate the comments that “The characterizations of T cell Nrn1 expression both in vitro and in vivo are comprehensive and convincing. The in vivo functional studies of anergy development, Treg suppression, and EAE development are also well done to strengthen the notion that Nrn1 is an important regulator of CD4 responsiveness.”

“The major weakness of this study stems from a lack of a clear molecular mechanism involving Nrn1. “  

We fully understand this comment from Reviewer #2. The main mechanism we identified contributing to the functional defect of Nrn1-/- T cells involves novel effects on the electric and metabolic state of the cells. Although we referenced neuronal studies that indicate Nrn1 is the auxiliary protein for the ionotropic AMPA-type glutamate receptor (AMPAR) and may affect AMPAR function, we did not provide any evidence in this manuscript as the topic requires further in-depth study.   

For the benefit of this discussion, we include our preliminary Nrn1 and AMPAR double knockout data (Author response image 2), which indicates that abrogating AMPAR expression can compensate for the defect caused by Nrn1 deficiency in vitro and in vivo. This preliminary data supports the notion that Nrn1 modulates AMPAR function, which causes changes in T cell electric and metabolic state, influencing T cell differentiation and function.  

Author response image 2.

Deletion of AMPAR expression in T cells compensates for the defect caused by Nrn1 deficiency. Nrn1-/- mice were crossed with T cell-specific AMPAR knockout mice (AMPARfl/flCD4Cre+) mice. The following mice were generated and used in the experiment: T cell specific AMPAR-knockout and Nrn1 knockout mice (AKONKO), Nrn1 knockout mice (AWTNKO), Ctrl mice (AWTNWT). a. Deletion of AMPAR compensates for the iTreg cell defect observed in Nrn1-/- CD4 cells. iTreg live cell proportion, cell number, and Ki67 expression among Foxp3+ cells 3 days after aCD3 restimulation. b. Deletion of AMPAR in T cells abrogates the enhanced autoimmune response in Nrn1-/- Mouse in the EAE disease model. Mouse relative weight change and disease score progression after EAE disease induction.  

Ion channels can influence cell metabolism through multiple means (Vaeth and Feske, 2018; Wang et al., 2020). First, ion channels are involved in maintaining cell resting membrane potential. This electrical potential difference across the cell membrane is essential for various cellular processes, including metabolism (Abdul Kadir et al., 2018; Blackiston et al., 2009; Nagy et al., 2018; Yu et al., 2022). Second, ion channels facilitate the movement of ions across cell membranes. These ions are essential for various metabolic processes. For example, ions like calcium (Ca2+), potassium (K+), and sodium (Na+) play crucial roles in signaling pathways that regulate metabolism (Kahlfuss et al., 2020). Third, ion channel activity can influence cellular energy balance due to ATP consumption associated with ion transport to maintain ion balances (Erecińska and Dagani, 1990; Gerkau et al., 2019). This, in turn, can impact processes like ATP production, which is central to cellular metabolism. Thus, ion channel expression and function determine the cell’s bioelectric state and contribute to cell metabolism (Levin, 2021).

Because the AMPAR function has not been thoroughly studied using a genetic approach in T cells, we do not intend to include the double knockout data in this manuscript before fully characterizing the T cell-specific AMPAR knockout mice.  

“Although the biochemical and informatics studies are well-performed, it is my opinion that these results are inconclusive in part due to the absence of key "naive" control groups. This limits my ability to understand the significance of these data.

Specifically, studies of the electrical and metabolic state of Nrn1-/- inducible Treg cells (iTregs) would benefit from similar data collected from wild-type and Nrn1-/- naive CD4 T cells.”

We appreciate the reviewer’s comments. This comment reflects two concerns in data interpretation:

(1) Are Nrn1-/- naïve T cells fundamentally different from WT cells? Does this fundamental difference contribute to the observed electrical and metabolic phenotype in iTreg or Th0 cells? This is a very good question we will perform the experiments as the reviewer suggested. While Nrn1 is expressed at a basal (low) level in naïve T cells, deletion of Nrn1 may cause changes in naïve T cell phenotype.   

(2) Is the Nrn1-/- phenotype caused by Nrn1 functional deficiency or due to the secondary effect of Nrn1 deletion, such as non-physiological cell membrane structure changes?

We have done the following experiment to address this concern.  We have cultured WT T cells in the presence of Nrn1 antibody and compared the outcome with Nrn1-/- iTreg cells (Figure 3-figure supplement 2D,E,F). WT iTreg cells under antibody blockade exhibited similar changes as Nrn1-/- iTreg cells, confirming the physiological relevance of the Nrn1-/- phenotype.

Manuscript Revision based on the Reviewer’s suggestions:

Reviewer #1:

Major points (3) Figure 1A-D. Since the authors are using the Nrp1 KO mice, it would be important to confirm the specificity of the anti-Nrn1 mAb by FACS. 

Following the suggestion by Reviewer#1, We have included the Nrn1 Ab staining on activated Nrn1-/- CD4 cells in Figure 1D. We have also added the staining of cell surface Nrn1 on Treg cells in Figure 1-figure supplement 1D.

Major point: (5) “Moreover, we do not even know what the % of Foxp3 cells is in the iTreg used (after differentiation and 20h of re-stimulation) and whether this % is the same between Ctlr and Nrn1 KO cells.”

In the revised manuscript, we have included the proportion of Foxp3+ cells among Nrn1-/- and ctrl iTreg cells developed under the iTreg culture condition in Figure 2A.

Minor points  

(2) For all figures showing %, the titles of the Y axes are written in an odd way. For example, it is written "Foxp3% CD4". It would be more conventional and clearer to write "% Foxp3+ / CD4+" or "% Foxp3+ among CD4+".

Following reviewer#1’s suggestion, we have changed the Y-axis label in all the relevant figures.

(3) Would not it be possible to perform experiments showing the ability of cells to transport amino acids or electrolytes across the plasma membrane? This would be a more interesting demonstration than transcriptomic data.”

We appreciate Review# 1’s suggestion regarding “perform experiments showing the ability of cells to transport amino acids or electrolytes across the plasma membrane”.  We have used AAinduced cellular MP changes to confirm differential AA transporter expression patterns and their impact on cellular MP levels.  The data are included in the revised manuscript in Figure 3H and Figure 4K.

(4) For certain staining (Figure 3E, H) it would be important to show the raw data, in addition to MFI or % values.

We appreciated Reviewer #1’s suggestion and have included the histogram staining data for Figure 3E. We have moved the original Figure 3H to the supplemental figure and included the histogram staining data in Figure 3-figure supplement 1C.  Similarly, we have included the histogram staining data in Figure 4-figure supplement 1C.

Reviewer#2:

“Although the biochemical and informatics studies are well-performed, it is my opinion that these results are inconclusive in part due to the absence of key "naive" control groups. This limits my ability to understand the significance of these data.

Specifically, studies of the electrical and metabolic state of Nrn1-/- inducible Treg cells (iTregs) would benefit from similar data collected from wild-type and Nrn1-/- naive CD4 T cells.”

We greatly appreciate Reviewer#2’s suggestion and have carried out experiments on naïve CD4 cells derived from Nrn1-/- and WT mice. We have compared membrane potential, AA-induced MP change between Nrn1-/- and WT naïve T cells, and the metabolic state of Nrn1-/- and WT naïve T cells by carrying out glucose stress tests and mitochondria stress tests using a seahorse assay.  Moreover, to investigate whether the phenotype revealed in Nrn1-/- CD4 cells was caused by a secondary effect of cell membrane structure change due to Nrn1 deletion, we carried out Nrn1 antibody blockade in WT CD4 cells and investigated the phenotypic change. These new results are included in Figure 3-figure supplement 2.

Reference:

Abdul Kadir, L., M. Stacey, and R. Barrett-Jolley. 2018. Emerging Roles of the Membrane Potential: Action Beyond the Action Potential. Front Physiol 9:1661.

Blackiston, D.J., K.A. McLaughlin, and M. Levin. 2009. Bioelectric controls of cell proliferation: ion channels, membrane voltage and the cell cycle. Cell Cycle 8:3527-3536.

Chappert, P., and R.H. Schwartz. 2010. Induction of T cell anergy: integration of environmental cues and infectious tolerance. Current opinion in immunology 22:552-559.

Chen, W., W. Jin, N. Hardegen, K.J. Lei, L. Li, N. Marinos, G. McGrady, and S.M. Wahl. 2003. Conversion of peripheral CD4+CD25- naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. The Journal of experimental medicine 198:1875-1886.

Erecińska, M., and F. Dagani. 1990. Relationships between the neuronal sodium/potassium pump and energy metabolism. Effects of K+, Na+, and adenosine triphosphate in isolated brain synaptosomes. J Gen Physiol 95:591-616.

Fathman, C.G., and N.B. Lineberry. 2007. Molecular mechanisms of CD4+ T-cell anergy. Nat Rev Immunol 7:599-609.

Gerkau, N.J., R. Lerchundi, J.S.E. Nelson, M. Lantermann, J. Meyer, J. Hirrlinger, and C.R. Rose. 2019. Relation between activity-induced intracellular sodium transients and ATP dynamics in mouse hippocampal neurons. The Journal of physiology 597:5687-5705.

Hurrell, B.P., D.G. Helou, E. Howard, J.D. Painter, P. Shafiei-Jahani, A.H. Sharpe, and O. Akbari. 2022. PD-L2 controls peripherally induced regulatory T cells by maintaining metabolic activity and Foxp3 stability. Nature communications 13:5118.

Jenkins, M.K., and R.H. Schwartz. 1987. Antigen presentation by chemically modified splenocytes induces antigen-specific T cell unresponsiveness in vitro and in vivo. The Journal of experimental medicine 165:302-319.

John, P., M.C. Pulanco, P.M. Galbo, Jr., Y. Wei, K.C. Ohaegbulam, D. Zheng, and X. Zang. 2022. The immune checkpoint B7x expands tumor-infiltrating Tregs and promotes resistance to anti-CTLA-4 therapy. Nature communications 13:2506.

Kahlfuss, S., U. Kaufmann, A.R. Concepcion, L. Noyer, D. Raphael, M. Vaeth, J. Yang, P. Pancholi, M. Maus, J. Muller, L. Kozhaya, A. Khodadadi-Jamayran, Z. Sun, P. Shaw, D. Unutmaz, P.B. Stathopulos, C. Feist, S.B. Cameron, S.E. Turvey, and S. Feske. 2020. STIM1-mediated calcium influx controls antifungal immunity and the metabolic function of nonpathogenic Th17 cells. EMBO molecular medicine 12:e11592.

Levin, M. 2021. Bioelectric signaling: Reprogrammable circuits underlying embryogenesis, regeneration, and cancer. Cell 184:1971-1989.

Nagy, E., G. Mocsar, V. Sebestyen, J. Volko, F. Papp, K. Toth, S. Damjanovich, G. Panyi, T.A. Waldmann, A. Bodnar, and G. Vamosi. 2018. Membrane Potential Distinctly Modulates Mobility and Signaling of IL-2 and IL-15 Receptors in T Cells. Biophys J 114:2473-2482.

Quill, H., and R.H. Schwartz. 1987. Stimulation of normal inducer T cell clones with antigen presented by purified Ia molecules in planar lipid membranes: specific induction of a long-lived state of proliferative nonresponsiveness. Journal of immunology (Baltimore, Md. : 1950) 138:3704-3712.

Schmitt, E.G., and C.B. Williams. 2013. Generation and function of induced regulatory T cells. Frontiers in immunology 4:152.

Sugiura, A., G. Andrejeva, K. Voss, D.R. Heintzman, X. Xu, M.Z. Madden, X. Ye, K.L. Beier, N.U. Chowdhury, M.M. Wolf, A.C. Young, D.L. Greenwood, A.E. Sewell, S.K. Shahi, S.N. Freedman, A.M. Cameron, P. Foerch, T. Bourne, J.C. Garcia-Canaveras, J. Karijolich, D.C. Newcomb, A.K. Mangalam, J.D. Rabinowitz, and J.C. Rathmell. 2022. MTHFD2 is a metabolic checkpoint controlling effector and regulatory T cell fate and function. Immunity 55:65-81.e69.

Vaeth, M., and S. Feske. 2018. Ion channelopathies of the immune system. Current opinion in immunology 52:39-50.

Vanasek, T.L., S.L. Nandiwada, M.K. Jenkins, and D.L. Mueller. 2006. CD25+Foxp3+ regulatory T cells facilitate CD4+ T cell clonal anergy induction during the recovery from lymphopenia. Journal of immunology (Baltimore, Md. : 1950) 176:5880-5889.

Wang, Y., A. Tao, M. Vaeth, and S. Feske. 2020. Calcium regulation of T cell metabolism. Current opinion in physiology 17:207-223.

Yu, W., Z. Wang, X. Yu, Y. Zhao, Z. Xie, K. Zhang, Z. Chi, S. Chen, T. Xu, D. Jiang, X. Guo, M. Li, J. Zhang, H. Fang, D. Yang, Y. Guo, X. Yang, X. Zhang, Y. Wu, W. Yang, and D. Wang. 2022. Kir2.1-mediated membrane potential promotes nutrient acquisition and inflammation through regulation of nutrient transporters. Nature communications 13:3544.

Zheng, S.G., J.D. Gray, K. Ohtsuka, S. Yamagiwa, and D.A. Horwitz. 2002. Generation ex vivo of TGF-beta-producing regulatory T cells from CD4+CD25- precursors. Journal of immunology (Baltimore, Md. : 1950) 169:4183-4189.

Author Response

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

Public Reviews:

Reviewer #1 (Public Review):

Koumoundourou et al., identify a pathway downstream of Bcl11b that controls synapse morphology and plasticity of hippocampal mossy fiber synapses. Using an elegant combination of in vivo, ex vivo, and in vitro approaches, the authors build on their previous work that indicated C1ql2 as a functional target of Bcl11b (De Bruyckere et al., 2018). Here, they examine the functional implications of C1ql2 at MF synapses in Bcl11b cKO mice and following C1ql2 shRNA. The authors find that Bcl11b KO and shRNA against C1ql2 significantly reduces the recruitment of synaptic vesicles and impairs LTP at MF synapses. Importantly, the authors test a role for the previously identified C1ql2 binding partner, exon 25b-containing Nrxn3 (Matsuda et al., 2016), as relevant at MF synapses to maintain synaptic vesicle recruitment. To test this, the authors developed a K262E C1ql2 mutant that disrupts binding to Nrxn3. Curiously, while Bcl11b KO and C1ql2 KD largely phenocopy (reduced vesicle recruitment and impaired LTP), only vesicle recruitment is dependent on C1ql2-Nrxn3 interactions. These findings provide new insight into the functional role of C1ql2 at MF synapses. While the authors convincingly demonstrate a role for C1ql2-Nrxn3(25b+) interaction for vesicle recruitment and a Nrxn3(25b+)independent role for C1ql2 in LTP, the underlying mechanisms remain inconclusive. Additionally, a discussion of how these findings relate to previous work on C1ql2 at mossy fiber synapses and how the findings contribute to the biology of Nrxn3 would increase the interpretability of this work.

As suggested by reviewer #1, we extended our discussion of previous work on C1ql2 and additionally discussed the biology of Nrxn3 and how our work relates to it. Moreover, we extended our mechanistic analysis of how Bcl11b/C1ql2/Nrxn3 pathway controls synaptic vesicle recruitment as well as LTP (please see also response to reviewer #2 points 5 and 8 and reviewer #3 point 4 of public reviews below for detailed discussion).

Reviewer #2 (Public Review):

This manuscript describes experiments that further investigate the actions of the transcription factor Bcl11b in regulating mossy fiber (MF) synapses in the hippocampus. Prior work from the same group had demonstrated that loss of Bcl11b results in loss of MF synapses as well as a decrease in LTP. Here the authors focus on a target of Bcl11b a secreted synaptic organizer C1ql2 which is almost completely lost in Bcl11b KO. Viral reintroduction of C1ql2 rescues the synaptic phenotypes, whereas direct KD of C1ql2 recapitulates the Bcl1 phenotype. C1ql2 itself interacts directly with Nrxn3 and replacement with a binding deficient mutant C1q was not able to rescue the Bcl11b KO phenotype. Overall there are some interesting observations in the study, however there are also some concerns about the measures and interpretation of data.

The authors state that they used a differential transcriptomic analysis to screen for candidate targets of Bcl11b, yet they do not present any details of this screen. This should be included and at the very least a table of all DE genes included. It is likely that many other genes are also regulated by Bcl11b so it would be important to the reader to see the rationale for focusing attention on C1ql2 in this study.

The transcriptome analysis mentioned in our manuscript was published in detail in our previous study (De Bruyckere et al., 2018), including chromatin-immunoprecipitation that revealed C1ql2 as a direct transcriptional target of Bcl11b. Upon revision of the manuscript, we made sure that this was clearly stated within the main text module to avoid future confusion. In the same publication (De Bruyckere et al., 2018), we discuss in detail several identified candidate genes such as Sema5b, Ptgs2, Pdyn and Penk as putative effectors of Bcl11b in the structural and functional integrity of MFS. C1ql2 has been previously demonstrated to be almost exclusively expressed in DG neurons and localized to the MFS.

There it bridges the pre- and post-synaptic sides through interaction with Nrxn3 and KAR subunits, respectively, and regulates synaptic function (Matsuda et al., 2016). Taken together, C1ql2 was a very good candidate to study as a potential effector downstream of Bcl11b in the maintenance of MFS structure and function. However, as our data reveal, not all Bcl11b mutant phenotypes were rescued by C1ql2 (see supplementary figures 2d-f of revised manuscript). We expect additional candidate genes, identified in our transcriptomic screen, to act downstream of Bcl11b in the control of MFS.

All viral-mediated expression uses AAVs which are known to ablate neurogenesis in the DG (Johnston DOI: 10.7554/eLife.59291) through the ITR regions and leads to hyperexcitability of the dentate. While it is not clear how this would impact the measurements the authors make in MF-CA3 synapses, this should be acknowledged as a potential caveat in this study.

We agree with reviewer #2 and are aware that it has been demonstrated that AAV-mediated gene expression ablates neurogenesis in the DG. To avoid potential interference of the AAVs with the interpretability of our phenotypes, we made sure during the design of the study that all of our control groups were treated in the same way as our groups of interest, and were, thus, injected with control AAVs. Moreover, the observed phenotypes were first described in Bcl11b mutants that were not injected with AVVs (De Bruyckere et al., 2018). Finally, we thoroughly examined the individual components of the proposed mechanism (rescue of C1ql2 expression, over-expression of C1ql3 and introduction of mutant C1ql2 in Bcl11b cKOs, KD of C1ql2 in WT mice, and Nrxn123 cKO) and reached similar conclusions. Together, this strongly supports that the observed phenotypes occur as a result of the physiological function of the proteins involved in the described mechanism and not due to interference of the AAVs with these biological processes. We have now addressed this point in the main text module of the revised ms.

The authors claim that the viral re-introduction "restored C1ql2 protein expression to control levels. This is misleading given that the mean of the data is 2.5x the control (Figure 1d and also see Figure 6c). The low n and large variance are a problem for these data. Moreover, they are marked ns but the authors should report p values for these. At the least, this likely large overexpression and variability should be acknowledged. In addition, the use of clipped bands on Western blots should be avoided. Please show the complete protein gel in primary figures of supplemental information.

We agree with reviewer #2 that C1ql2 expression after its re-introduction in Bcl11b cKO mice was higher compared to controls and that this should be taken into consideration for proper interpretation of the data. To address this, based also on the suggestion of reviewer #3 point 1 below, we overexpressed C1ql2 in DG neurons of control animals. We found no changes in synaptic vesicle organization upon C1ql2 over-expression compared to controls. This further supports that the observed effect upon rescue of C1ql2 expression in Bcl11b cKOs is due to the physiological function of C1ql2 and not as result of the overexpression. These data are included in supplementary figure 2g-j and are described in detail in the results part of the revised manuscript.

Additionally, we looked at the effects of C1ql2 overexpression in Bcl11b cKO DGN on basal synaptic transmission. We plotted fEPSP slopes versus fiber volley amplitudes, measured in slices from rescue animals, as we had previously done for the control and Bcl11b cKO (Author response image 1a). Although regression analysis revealed a trend towards steeper slopes in the rescue mice (Author response image 1a and b), the observation did not prove to be statistically significant, indicating that C1ql2 overexpression in Bcl11b cKO animals does not strongly alter basal synaptic transmission at MFS. Overall, our previous and new findings support that the observed effects of the C1ql2 rescue are not caused by the artificially elevated levels of C1ql2, as compared to controls, but are rather a result of the physiological function of C1ql2.

Following the suggestion of reviewer #2 all western blot clipped bands were exchanged for images of the full blot. This includes figures 1c, 4c, 6b and supplementary figure 2g of the revised manuscript. P-value for Figure 1d has now been included.

Author response image 1.

C1ql2 reintroduction in Bcl11b cKO DGN does not significantly alter basal synaptic transmission at mossy fiber-CA3 synapses. a Input-output curves generated by plotting fEPSP slope against fiber volley amplitude at increasing stimulation intensities. b Quantification of regression line slopes for input-output curves for all three conditions. Control+EGFP, 35 slices from 16 mice; Bcl11b cKO+EGFP, 32 slices from 14 mice; Bcl11b cKO+EGFP-2A-C1ql2, 22 slices from 11 mice. The data are presented as means, error bars represent SEM. Kruskal-Wallis test (non-parametric ANOVA) followed by Dunn’s post hoc pairwise comparisons. p=0.106; ns, not significant.

Measurement of EM micrographs: As prior work suggested that MF synapse structure is disrupted the authors should report active zone length as this may itself affect "synapse score" defined by the number of vesicles docked. More concerning is that the example KO micrographs seem to have lost all the densely clustered synaptic vesicles that are away from the AZ in normal MF synapses e.g. compare control and KO terminals in Fig 2a or 6f or 7f. These terminals look aberrant and suggest that the important measure is not what is docked but what is present in the terminal cytoplasm that normally makes up the reserve pool. This needs to be addressed with further analysis and modifications to the manuscript.

As requested by reviewer #2 we analyzed and reported in the revised manuscript the active zone length. We found that the active zone length remained unchanged in all conditions (control/Bcl11b cKO/C1ql2 rescue, WT/C1ql2 KD, control/K262E and control/Nrxn123 cKO), strengthening our results that the described Bcl11b/C1ql2/Nrxn3 mechanism is involved in the recruitment of synaptic vesicles. These data have been included in supplementary figures 2c, 4h, 5f and 6g and are described in the results part of the revised manuscript.

We want to clarify that the synapse score is not defined by the number of docked vesicles to the plasma membrane. The synapse score, which is described in great detail in our materials and methods part and has been previously published (De Bruyckere et al., 2018), rates MFS based on the number of synaptic vesicles and their distance from the active zone and was designed according to previously described properties of the vesicle pools at the MFS. The EM micrographs refer to the general misdistribution of SV in the proximity of MFS. Upon revision of the manuscript, we made sure that this was clearly stated in the main text module to avoid further confusion.

The study also presents correlated changes in MF LTP in Bcl11b KO which are rescued by C1ql2 expression. It is not clear whether the structural and functional deficits are causally linked and this should be made clearer in the manuscript. It is also not apparent why this functional measure was chosen as it is unlikely that C1ql2 plays a direct role in presynaptic plasticity mechanisms that are through a cAMP/ PKA pathway and likely disrupted LTP is due to dysfunctional synapses rather than a specific LTP effect.

The inclusion of functional experiments in this and our previous study (de Bruyckere et al., 2018) was first and foremost intended to determine whether the structural alterations observed at MFB disrupt MFS signaling. From the signaling properties we tested, basal synaptic transmission (this study) and short-term potentiation (de Bruyckere et al., 2018) were unaltered by Bcl11b KO, whereas MF LTP was found to be abolished (de Bruyckere et al., 2018). Indeed, because MF LTP largely depends on presynaptic mechanisms, including the redistribution of the readily releasable pool and recruitment of new active zones (Orlando et al., 2021; Vandael et al., 2020), it appears to be particularly sensitive to the specific structural changes we observed. We therefore believe that it is valuable information that MF LTP is affected in Bcl11b cKO animals - it conveys a direct proof for the functional importance of the observed morphological alterations, while basic transmission remains largely normal. Furthermore, it subsequently provided a functional marker for testing whether the reintroduction of C1ql2 in Bcl11b cKO animals or the KD of C1ql2 in WT animals can functionally recapitulate the control or the Bcl11b KO phenotype, respectively.

We fully agree with the reviewer that C1ql2 is unlikely to directly participate in the cAMP/PKA pathway and that the ablation of C1ql2 likely disrupts MF LTP through an alternative mode of action. Our original wording in the paragraph describing the results of the forskolin-induced LTP experiment might have overstressed the importance of the cAMP pathway. We have now rephrased that paragraph to better describe the main idea behind the forskolin experiment, namely to circumvent the initial Ca2+ influx in order to test whether deficient presynaptic Ca2+ channel/KAR signaling might be responsible for the loss of LTP in Bcl11b cKO. The results are strongly indicative of a downstream mechanism and further investigation is needed to determine the specific mechanisms by which C1ql2 regulates MFLTP, especially in light of the result that C1ql2.K262E rescued LTP, while it was unable to rescue the SV recruitment at the MF presynapse. This raises the possibility that C1ql2 can influence MF-LTP through additional, yet uncharacterized mechanisms, independent of SV recruitment. As such, a causal link between the structural and functional deficits remains tentative and we have now emphasized that point by adding a respective sentence to the discussion of our revised manuscript. Nevertheless, we again want to stress that the main rationale behind the LTP experiments was to assess the functional significance of structural changes at MFS and not to elucidate the mechanisms by which MF LTP is established.

The authors should consider measures that might support the role of Bcl11b targets in SV recruitment during the depletion of synapses or measurements of the readily releasable pool size that would complement their findings in structural studies.

We fully agree that functional measurements of the readily releasable pool (RRP) size would be a valuable addition to the reported redistribution of SV in structural studies. We have, in fact, attempted to use high-frequency stimulus trains in both field and single-cell recordings (details on single-cell experiments are described in the response to point 8) to evaluate potential differences in RRP size between the control and Bcl11b KO (Figure for reviewers 2a and b). Under both recording conditions we see a trend towards lower values of the intersection between a regression line of late responses and the y-axis. This could be taken as an indication of slightly smaller RRP size in Bcl11b mutant animals compared to controls. However, due to several technical reasons we are extremely cautious about drawing such far-reaching conclusions based on these data. At most, they suffice to conclude that the availability of release-ready vesicles in the KO is likely not dramatically smaller than in the control.

The primary issue with using high-frequency stimulus trains for RRP measurements at MFS is the particularly low initial release probability (Pr) at these synapses. This means that a large number of stimulations is required to deplete the RRP. As the RRP is constantly replenished, it remains unclear when steady state responses are reached (reviewed by Kaeser and Regehr, 2017). This is clearly visible in our single-cell recordings (Author response image 2b), which were additionally complicated by prominent asynchronous release at later stages of the stimulus train and by a large variability in the shapes of cumulative amplitude curves between cells. In contrast, while the cumulative amplitude curves for field potential recordings do reach a steady state (Author response image 2a), field potential recordings in this context are not a reliable substitute for single cell or, in the case of MFB, singlebouton recordings. Postsynaptic cells in field potential recordings are not clamped, meaning that the massive release of glutamate due to continuous stimulation depolarizes the postsynaptic cells and reduces the driving force for Na+, irrespective of depletion of the RRP. This is supported by the fact that we consistently observed a recovery of fEPSP amplitudes later in the trains where RRP had presumably been maximally depleted. In summary, high-frequency stimulus trains at the field potential level are not a valid and established technique for estimating RRP size at MFS.

Specialized laboratories have used highly advanced techniques, such as paired recordings between individual MFB and postsynaptic CA3 pyramidal cells, to estimate the RRP size of MFB (Vandael et al., 2020). These approaches are outside the scope of our present study which, while elucidating functional changes following Bcl11b depletion and C1ql2 rescue, does not aim to provide a high-end biophysical analysis of the presynaptic mechanisms involved.

Author response image 2.

Estimation of RRP size using high-frequency stimulus trains at mossy fiber-CA3 synapses. a Results from field potential recordings. Cumulative fEPSP amplitude in response to a train of 40 stimuli at 100 Hz. All subsequent peak amplitudes were normalized to the amplitude of the first peak. Data points corresponding to putative steady state responses were fit with linear regression (RRP size is indirectly reflected by the intersection of the regression line with the yaxis). Control+EGFP, 6 slices from 5 mice; Bcl11b cKO+EGFP, 6 slices from 3 mice. b Results from single-cell recordings. Cumulative EPSC amplitude in response to a train of 15 stimuli at 50 Hz. The last four stimuli were fit with linear regression. Control, 5 cells from 4 mice; Bcl11b cKO, 3 cells from 3 mice. Note the shallow onset of response amplitudes and the subsequent frequency potentiation. Due to the resulting increase in slope at higher stimulus numbers, intersection with the y-axis occurs at negative values. The differences shown were not found to be statistically significant; unpaired t-test or Mann-Whitney U-test.

Bcl11b KO reduces the number of synapses, yet the I-O curve reported in Supp Fig 2 is not changed. How is that possible? This should be explained.

We agree with reviewer #2– this apparent discrepancy has indeed struck us as a counterintuitive result. It might be that synapses that are preferentially eliminated in Bcl11b cKO are predominantly silent or have weak coupling strength, such that their loss has only a minimal effect on basal synaptic transmission. Although perplexing, the result is fully supported by our single-cell data which shows no significant differences in MF EPSC amplitudes recorded from CA3 pyramidal cells between controls and Bcl11b mutants (Author response image 3; please see the response below for details and also our response to Reviewer #1 question 2).

Matsuda et al DOI: 10.1016/j.neuron.2016.04.001 previously reported that C1ql2 organizes MF synapses by aligning postsynaptic kainate receptors with presynaptic elements. As this may have consequences for the functional properties of MF synapses including their plasticity, the authors should report whether they see deficient postsynaptic glutamate receptor signaling in the Bcl11b KO and rescue in the C1ql2 re-expression.

We agree that the study by Matsuda et al. is of key importance for our present work. Although MF LTP is governed by presynaptic mechanisms and we previously did not see differences in short-term plasticity between the control and Bcl11b cKO (De Bruyckere et al., 2018), the clustering of postsynaptic kainate receptors by C1ql2 is indeed an important detail that could potentially alter synaptic signaling at MFS in Bcl11b KO. We, therefore, re-analyzed previously recorded single-cell data by performing a kinetic analysis on MF EPSCs recorded from CA3 pyramidal cells in control and Bcl11b cKO mice (Figure for reviewers 3a) to evaluate postsynaptic AMPA and kainate receptor responses in both conditions. We took advantage of the fact that AMPA receptors deactivate roughly 10 times faster than kainate receptors, allowing the contributions of the two receptors to mossy fiber EPSCs to be separated (Castillo et al., 1997 and reviewed by Lerma, 2003). We fit the decay phase of the second (larger) EPSC evoked by paired-pulse stimulation with a double exponential function, yielding a fast and a slow component, which roughly correspond to the fractional currents evoked by AMPA and kainate receptors, respectively. Analysis of both fast and slow time constants and the corresponding fractional amplitudes revealed no significant differences between controls and Bcl11b mutants (Figure for reviewers 3e-h), indicating that both AMPA and kainate receptor signaling is unaffected by the ablation of C1ql2 following Bcl11b KO.

Importantly, MF EPSC amplitudes evoked by the first and the second pulse (Author response image 3b), paired-pulse facilitation (Author response image 3c) and failure rates (Author response image 3d) were all comparable between controls and Bcl11b mutants. These results further corroborate our observations from field recordings that basal synaptic transmission at MFS is unaltered by Bcl11b KO.

We note that the results from single cell recordings regarding basal synaptic transmission merely confirm the observations from field potential recordings, and that the attempted measurement of RRP size at the single cell level was not successful. Thus, our single-cell data do not add new information about the mechanisms underlying the effects of Bcl11b-deficiency and we therefore decided not to report these data in the manuscript.

Author response image 3.

Basal synaptic transmission at mossy fiber-CA3 synapses is unaltered in Bcl11b cKO mice. a Representative average trace (20 sweeps) recorded from CA3 pyramidal cells in control and Bcl11b cKO mice at minimal stimulation conditions, showing EPSCs in response to paired-pulse stimulation (PPS) at an interstimulus interval of 40 ms. The signal is almost entirely blocked by the application of 2 μM DCG-IV (red). b Quantification of MF EPSC amplitudes in response to PPS for both the first and the second pulse. c Ratio between the amplitude of the second over the first EPSC. d Percentage of stimulation events resulting in no detectable EPSCs for the first pulse. Events <5 pA were considered as noise. e Fast decay time constant obtained by fitting the average second EPSC with the following double exponential function: I(t)=Afaste−t/τfast+Aslowe−t/τslow+C, where I is the recorded current amplitude after time t, Afast and Aslow represent fractional current amplitudes decaying with the fast (τfast) and slow (τslow) time constant, respectively, and C is the offset. Starting from the peak of the EPSC, the first 200 ms of the decaying trace were used for fitting. f Fractional current amplitude decaying with the fast time constant. g-h Slow decay time constant and fractional current amplitude decaying with the slow time constant. For all figures: Control, 8 cells from 4 mice; Bcl11b cKO, 8 cells from 6 mice. All data are presented as means, error bars indicate SEM. None of the differences shown were found to be statistically significant; Mann-Whitney U-test for nonnormally and unpaired t-test for normally distributed data.

Reviewer #3 (Public Review):

Overall, this is a strong manuscript that uses multiple current techniques to provide specific mechanistic insight into prior discoveries of the contributions of the Bcl11b transcription factor to mossy fiber synapses of dentate gyrus granule cells. The authors employ an adult deletion of Bcl11b via Tamoxifen-inducible Cre and use immunohistochemical, electron microscopy, and electrophysiological studies of synaptic plasticity, together with viral rescue of C1ql2, a direct transcriptional target of Bcl11b or Nrxn3, to construct a molecular cascade downstream of Bcl11b for DG mossy fiber synapse development. They find that C1ql2 re-expression in Bcl11b cKOs can rescue the synaptic vesicle docking phenotype and the impairments in MF-LTP of these mutants. They also show that C1ql2 knockdown in DG neurons can phenocopy the vesicle docking and plasticity phenotypes of the Bcl11b cKO. They also use artificial synapse formation assays to suggest that C1ql2 functions together with a specific Nrxn3 splice isoform in mediating MF axon development, extending these data with a C1ql2-K262E mutant that purports to specifically disrupt interactions with Nrxn3. All of the molecules involved in this cascade are disease-associated and this study provides an excellent blueprint for uncovering downstream mediators of transcription factor disruption. Together this makes this work of great interest to the field. Strengths are the sophisticated use of viral replacement and multi-level phenotypic analysis while weaknesses include the linkage of C1ql2 with a specific Nrxn3 splice variant in mediating these effects.

Here is an appraisal of the main claims and conclusions:

1) C1ql2 is a downstream target of Bcl11b which mediates the synaptic vesicle recruitment and synaptic plasticity phenotypes seen in these cKOs. This is supported by the clear rescue phenotypes of synapse anatomy (Fig.2) and MF synaptic plasticity (Fig.3). One weakness here is the absence of a control assessing over-expression phenotypes of C1ql2. It's clear from Fig.1D that viral rescue is often greater than WT expression (totally expected). In the case where you are trying to suppress a LoF phenotype, it is important to make sure that enhanced expression of C1ql2 in a WT background does not cause your rescue phenotype. A strong overexpression phenotype in WT would weaken the claim that C1ql2 is the main mediator of the Bcl11b phenotype for MF synapse phenotypes.

As suggested by reviewer #3, we carried out C1ql2 over-expression experiments in control animals. We show that the over-expression of C1ql2 in the DG of control animals had no effect on the synaptic vesicle organization in the proximity of MFS. This further supports that the observed effect upon rescue of C1ql2 expression in Bcl11b cKOs is due to the physiological function of C1ql2 and not a result of the artificial overexpression. These data are now included in supplementary figure 2g-j and are described in detail in the results part of the revised manuscript. Please also see response to point 3 of reviewer #2.

2) Knockdown of C1ql2 via 4 shRNAs is sufficient to produce the synaptic vesicle recruitment and MFLTP phenotypes. This is supported by clear effects in the shRNA-C1ql2 groups as compared to nonsense-EGFP controls. One concern (particularly given the use of 4 distinct shRNAs) is the potential for off-target effects, which is best controlled for by a rescue experiment with RNA insensitive C1ql2 cDNA as opposed to nonsense sequences, which may not elicit the same off-target effects.

We agree with reviewer #3 that the usage of shRNAs could potentially create unexpected off-target effects and that the introduction of a shRNA-insensitive C1ql2 in parallel to the expression on the shRNA cassette would be a very effective control experiment. However, the suggested experiment would require an additional 6 months (2 months for AAV production, 2-3 months from animal injection to sacrifice and 1-2 months for EM imaging/analysis and LTP measurements) and a high number of additional animals (minimum 8 for EM and 8 for LTP measurements). We note here, that before the production of the shRNA-C1ql2 and the shRNA-NS, the individual sequences were systematically checked for off-target bindings on the murine exome with up to two mismatches and presented with no other target except the proposed (C1ql2 for shRNA-C1ql2 and no target for shRNA-NS). Taking into consideration our in-silico analysis, we feel that the interpretation of our findings is valid without this (very reasonable) additional control experiment.

3) C1ql2 interacts with Nrxn3(25b+) to facilitate MF terminal SV clustering. This claim is theoretically supported by the HEK cell artificial synapse formation assay (Fig.5), the inability of the K262-C1ql2 mutation to rescue the Bcl11b phenotype (Fig.6), and the altered localization of C1ql2 in the Nrxn1-3 deletion mice (Fig.7). Each of these lines of experimental evidence has caveats that should be acknowledged and addressed. Given the hypothesis that C1ql2 and Nrxn3b(25b) are expressed in DG neurons and work together, the heterologous co-culture experiment seems strange. Up till now, the authors are looking at pre-synaptic function of C1ql2 since they are re-expressing it in DGNs. The phenotypes they are seeing are also pre-synaptic and/or consistent with pre-synaptic dysfunction. In Fig.5, they are testing whether C1ql2 can induce pre-synaptic differentiation in trans, i.e. theoretically being released from the 293 cells "post-synaptically". But the post-synaptic ligands (Nlgn1 and and GluKs) are not present in the 293 cells, so a heterologous synapse assay doesn't really make sense here. The effect that the authors are seeing likely reflects the fact that C1ql2 and Nrxn3 do bind to each other, so C1ql2 is acting as an artificial post-synaptic ligand, in that it can cluster Nrxn3 which in turn clusters synaptic vesicles. But this does not test the model that the authors propose (i.e. C1ql2 and Nrxn3 are both expressed in MF terminals). Perhaps a heterologous assay where GluK2 is put into HEK cells and the C1ql2 and Nrxn3 are simultaneously or individually manipulated in DG neurons?

C1ql2 is expressed by DG neurons and is then secreted in the MFS synaptic cleft, while Nrxn3, that is also expressed by DG neurons, is anchored at the presynaptic side. In our work we used the well established co-culture system assay and cultured HEK293 cells secreting C1ql2 (an IgK secretion sequence was inserted at the N-terminus of C1ql2) together with hippocampal neurons expressing Nrxn3(25b+). We used the HEK293 cells as a delivery system of secreted C1ql2 to the neurons to create regions of high concentration of C1ql2. By interfering with the C1ql2-Nrxn3 interaction in this system either by expression of the non-binding mutant C1ql2 variant in the HEK cells or by manipulating Nrxn expression in the neurons, we could show that C1ql2 binding to Nrxn3(25b+) is necessary for the accumulation of vGlut1. However, we did not examine and do not claim within our manuscript that the interaction between C1ql2 and Nrxn3(25b+) induces presynaptic differentiation. Our experiment only aimed to analyze the ability of C1ql2 to cluster SV through interaction with Nrxn3. Moreover, by not expressing potential postsynaptic interaction partners of C1ql2 in our system, we could show that C1ql2 controls SV recruitment through a purely presynaptic mechanism. Co-culturing GluK2-expressing HEK cells with simultaneous manipulation of C1ql2 and/or Nrxn3 in neurons would not allow us to appropriately answer our scientific question, but rather focus on the potential synaptogenic function of the Nrxn3/C1ql2/GluK2 complex and the role of the postsynaptic ligand in it. Thus, we feel that the proposed experiment, while very interesting in characterization of additional putative functions of C1ql2, may not provide additional information for the point we were addressing. In the revised manuscript we tried to make the aim and methodological approach of this set of experiments more clear.

4) K262-C1ql2 mutation blocks the normal rescue through a Nrxn3(25b) mechanism (Fig.6). The strength of this experiment rests upon the specificity of this mutation for disrupting Nrxn3b binding (presynaptic) as opposed to any of the known postsynaptic C1ql2 ligands such as GluK2. While this is not relevant for interpreting the heterologous assay (Fig.5), it is relevant for the in vivo phenotypes in Fig.6. Similar approaches as employed in this paper can test whether binding to other known postsynaptic targets is altered by this point mutation.

It has been previously shown that C1ql2 together with C1ql3 recruit postsynaptic GluK2 at the MFS. However, loss of just C1ql2 did not affect the recruitment of GluK2, which was disrupted only upon loss of both C1ql2 and C1ql3 (Matsuda et al., 2018). In our study we demonstrate a purely presynaptic function of C1ql2 through Nrxn3 in the synaptic vesicle recruitment. This function is independent of C1ql3, as C1ql3 expression is unchanged in all of our models and its over-expression did not compensate for C1ql2 functions (Fig. 2, 3a-c). Our in vitro experiments also reveal that C1ql2 can recruit both Nrxn3 and vGlut1 in the absence of any known postsynaptic C1ql2 partner (KARs and BAI3; Fig.5; please also see response above). Furthermore, we have now performed a kinetic analysis on single-cell data which we had previously collected to evaluate postsynaptic AMPA and kainate receptor responses in both the control and Bcl11b KO. Our analysis reveals no significant differences in postsynaptic current kinetics, making it unlikely that AMPA and kainate receptor signaling is altered upon the loss of C1ql2 following Bcl11b cKO (Author response image 3e-h; please also see our response to reviewer #2 point 8). Thus, we have no experimental evidence supporting the idea that a loss of interaction between C1ql2.K262E and GluK2 would interfere with the examined phenotype. However, to exclude that the K262E mutation disrupts interaction between C1ql2 and GluK2, we performed co-immunoprecipitation from protein lysate of HEK293 cells expressing GluK2myc-flag and GFP-C1ql2 or GluK2-myc-flag and GFP-K262E and could show that both C1ql2 and K262E had GluK2 bound when precipitated. These data are included in supplementary figure 5k of the revised manuscript.

5) Altered localization of C1ql2 in Nrxn1-3 cKOs. These data are presented to suggest that Nrx3(25b) is important for localizing C1ql2 to the SL of CA3. Weaknesses of this data include both the lack of Nrxn specificity in the triple a/b KOs as well as the profound effects of Nrxn LoF on the total levels of C1ql2 protein. Some measure that isn't biased by this large difference in C1ql2 levels should be attempted (something like in Fig.1F).

We acknowledge that the lack of specificity in the Nrxn123 model makes it difficult to interpret our data. We have now examined the mRNA levels of Nrxn1 and Nrxn2 upon stereotaxic injection of Cre in the DG of Nrxn123flox/flox animals and found that Nrxn1 was only mildly reduced. At the same time Nrxn2 showed a tendency for reduction that was not significant (data included in supplementary figure 6a of revised manuscript). Only Nrxn3 expression was strongly suppressed. Of course, this does not exclude that the mild reduction of Nrxn1 and Nrxn2 interferes with the C1ql2 localization at the MFS. We further examined the mRNA levels of C1ql2 in control and Nrxn123 mutants to ensure that the observed changes in C1ql2 protein levels at the MFS are not due to reduced mRNA expression and found no changes (data are included in supplementary figure 6b of the revised manuscript), suggesting that overall protein C1ql2 expression is normal.

The reduced C1ql2 fluorescence intensity at the MFS was first observed when non-binding C1ql2 variant K262E was introduced to Bcl11b cKO mice that lack endogenous C1ql2 (Fig.6). In these experiments, we found that despite the overall high protein levels of C1ql2.K262E in the hippocampus (Fig. 6c), its fluorescence intensity at the SL was significantly reduced compared to WT C1ql2 (Fig. 6d-e). The remaining signal of the C1ql2.K262E at the SL was equally distributed and in a punctate form, similar to WT C1ql2. Together, this suggests that loss of C1ql2-Nrxn3 interaction interferes with the localization of C1ql2 at the MFS, but not with the expression of C1ql2. Of course, this does not exclude that other mechanisms are involved in the synaptic localization of C1ql2, beyond the interaction with Nrxn3, as both the mutant C1ql2 in Bcl11b cKO and the endogenous C1ql2 in Nrxn123 cKOs show residual immunofluorescence at the SL. Further studies are required to determine how C1ql2-Nrxn3 interaction regulates C1ql2 localization at the MFS.

Reviewer #1 (Recommendations For The Authors):

In addition to addressing the comments below, this study would benefit significantly from providing insight and discussion into the relevant potential postsynaptic signaling components controlled exclusively by C1ql2 (postsynaptic kainate receptors and the BAI family of proteins).

We have now performed a kinetic analysis on single-cell data that we had previously collected to evaluate postsynaptic AMPA and kainate receptor responses in both the control and Bcl11b cKO. Our analysis reveals no significant differences in postsynaptic current kinetics, making it unlikely that AMPA and kainate receptor signaling differ between controls and upon the loss of C1ql2 following Bcl11b cKO (Author response image 3e-h; please also see our response to Reviewer #2 point 8). This agrees with previous findings that C1ql2 regulates postsynaptic GluK2 recruitment together with C1ql3 and only loss of both C1ql2 and C1ql3 results in a disruption of KAR signaling (Matsuda et al., 2018). In our study we demonstrate a purely presynaptic function of C1ql2 through Nrxn3 in the synaptic vesicle recruitment. This function is independent of C1ql3, as C1ql3 expression is unchanged in all of our models and its over-expression did not compensate for C1ql2 functions (Fig. 2, 3a-c). Our in vitro experiments also reveal that C1ql2 can recruit both Nrxn3 and vGlut1 in the absence of any known postsynaptic C1ql2 partner (KARs and BAI3; Fig.5; please also see our response to reviewer #3 point 4 above). We believe that further studies are needed to fully understand both the pre- and the postsynaptic functions of C1ql2. Because the focus of this manuscript was on the role of the C1ql2-Nrxn3 interaction and our investigation on postsynaptic functions of C1ql2 was incomplete, we did not include our findings on postsynaptic current kinetics in our revised manuscript. However, we increased the discussion on the known postsynaptic partners of C1ql2 in the revised manuscript to increase the interpretability of our results.

Major Comments:

The authors demonstrate that the ultrastructural properties of presynaptic boutons are altered after Bcl11b KO and C1ql2 KD. However, whether C1ql2 functions as part of a tripartite complex and the identity of the postsynaptic receptor (BAI, KAR) should be examined.

Matsuda and colleagues have nicely demonstrated in their 2016 (Neuron) study that C1ql2 is part of a tripartite complex with presynaptic Nrxn3 and postsynaptic KARs. Moreover, they demonstrated that C1ql2, together with C1ql3, recruit postsynaptic KARs at the MFS, while the KO of just C1ql2 did not affect the KAR localization. In our study we demonstrate a purely presynaptic function of C1ql2 through Nrxn3 in the synaptic vesicle recruitment. This function is independent of C1ql3, as C1ql3 expression is unchanged in all of our models and its over-expression did not compensate for C1ql2 functions (Fig. 2, 3a-c). Our in vitro experiments also reveal that C1ql2 is able to recruit both Nrxn3 and vGlut1 in the absence of any known postsynaptic C1ql2 partner (Fig. 5; please also see our response to reviewer #3 point 4 above). Moreover, we were able to show that the SV recruitment depends on C1ql2 interaction with Nrxn3 through the expression of a non-binding C1ql2 (Fig. 6) that retains the ability to interact with GluK2 (supplementary figure 5k of revised manuscript) or by KO of Nrxns (Fig. 7). Furthermore, we have now performed a kinetic analysis on single-cell data which we had previously collected to evaluate postsynaptic AMPA and kainate receptor responses in both the control and Bcl11b cKO. Our analysis reveals no significant differences in postsynaptic current kinetics, making it unlikely that AMPA and kainate receptor signaling differ between controls and Bcl11b mutants (Author response image 3e-h; please also see our response to Reviewer #2 question 8). Together, we have no experimental evidence so far that would support that the postsynaptic partners of C1ql2 are involved in the observed phenotype. While it would be very interesting to characterize the postsynaptic partners of C1ql2 in depth, we feel this would be beyond the scope of the present study.

Figure 1f: For a more comprehensive understanding of the Bcl11b KO phenotype and the potential role for C1ql2 on MF synapse number, a complete quantification of vGlut1 and Homer1 for all conditions (Supplement Figure 2e) should be included in the main text.

In our study we focused on the role of C1ql2 in the structural and functional integrity of the MFS downstream of Bcl11b. Bcl11b ablation leads to several phenotypes in the MFS that have been thoroughly described in our previous study (De Bruyckere et al., 2018). As expected, re-expression of C1ql2 only partially rescued these phenotypes, with full recovery of the SV recruitment (Fig. 2) and of the LTP (Fig. 3), but had no effect on the reduced numbers of MFS nor the structural complexity of the MFB created by the Bcl11b KO (supplementary figure 2d-f of revised manuscript). We understand that including the quantification of vGlut1 and Homer1 co-localization in the main figures would help with a better understanding of the Bcl11b mutant phenotype. However, in our manuscript we investigate C1ql2 as an effector of Bcl11b and thus we focus on its functions in SV recruitment and LTP. As we did not find a link between C1ql2 and the number of MFS/MFB upon re-expression of C1ql2 in Bcl11b cKO or now also in C1ql2 KD (see response to comment #4 below), we believe it is more suitable to present these data in the supplement.

Figure 3/4: Given the striking reduction in the numbers of synapses (Supplement Figure 2e) and docked vesicles (Figure 2d) in the Bcl11b KO and C1ql2 KD (Figure 4e-f), it is extremely surprising that basal synaptic transmission is unaffected (Supplement Figure 2g). The authors should determine the EPSP input-output relationship following C1ql2 KD and measure EPSPs following trains of stimuli at various high frequencies.

We fully acknowledge that this is an unexpected result. It is, however, well feasible that the modest displacement of SV fails to noticeably influence basal synaptic transmission. This would be the case, for example, if only a low number of vesicles are released by single stimuli, in line with the very low initial Pr at MFS. In contrast, the reduction in synapse numbers in the Bcl11b mutant might indeed be expected to reflect in the input-output relationship. It is possible, however, that synapses that are preferentially eliminated in Bcl11b cKO are predominantly silent or have weak coupling strength, such that their loss has only a minimal effect on basal synaptic transmission. Finally, we cannot exclude compensatory mechanisms (homeostatic plasticity) at the remaining synapses. A detailed analysis of these potential mechanisms would be a whole project in its own right.

As additional information, we can say that the largely unchanged input-output-relation in Bcl11b cKO is also present in the single-cell level data (Author response image 3; details on single-cell experiments are described in the response to Reviewer #2 point 8).

As suggested by the reviewer, we have now additionally analyzed the input-output relationship following C1ql2 KD and again did not observe any significant difference between control and KD animals. We have incorporated the respective input-output curves into the revised manuscript under Supplementary figure 3c-d.

Figure 4: Does C1ql2 shRNA also reduce the number of MFBs? This should be tested to further identify C1ql2-dependent and independent functions.

As requested by reviewer #1 we quantified the number of MFBs upon C1ql2 KD. We show that C1ql2 KD in WT animals does not alter the number of MFBs. The data are presented in supplementary figure 4d of the revised manuscript. Re-expression of C1ql2 in Bcl11b cKO did not rescue the loss of MFS created by the Bcl11b mutation. Moreover, C1ql2 re-expression did not rescue the complexity of the MFB ultrastructure perturbed by the Bcl11b ablation. Together, this suggests that Bcl11b regulates MFs maintenance through additional C1ql2-independent pathways. In our previously published work (De Bruyckere et al., 2018) we identified and discussed in detail several candidate genes such as Sema5b, Ptgs2, Pdyn and Penk as putative effectors of Bcl11b in the structural and functional integrity of MFS (please also see response to reviewer #2- point 1 of public reviews).

Figure 5: Clarification is required regarding the experimental design of the HEK/Neuron co-culture: 1. C1ql2 is a secreted soluble protein - how is the protein anchored to the HEK cell membrane to recruit Nrxn3(25b+) binding and, subsequently, vGlut1?

C1ql2 was secreted by the HEK293 cells through an IgK signaling peptide at the N-terminus of C1ql2. The high concentration of C1ql2 close to the secretion site together with the sparse coculturing of the HEK293 cells on the neurons allows for the quantification of accumulation of neuronal proteins. We have now described the experimental conditions in greater detail in the main text module of the revised manuscript

2) Why are the neurons transfected and not infected? Transfection efficiency of neurons with lipofectamine is usually poor (1-5%; Karra et al., 2010), while infection of neurons with lentiviruses or AAVs encoding cDNAs routinely are >90% efficient. Thus, interpretation of the recruitment assays may be influenced by the density of neurons transfected near a HEK cell.

We agree with reviewer #1 that viral infection of the neurons would have been a more effective way of expressing our constructs. However, due to safety allowances in the used facility and time limitation at the time of conception of this set of experiments, a lipofectamine transfection was chosen.

However, as all of our examined groups were handled in the same way and multiple cells from three independent experiments were examined for each experimental set, we believe that possible biases introduced by the transfection efficiency have been eliminated and thus have trust in our interpretation of these results.

3) Surface labeling of HEK cells for wild-type C1ql2 and K262 C1ql2 would be helpful to assess the trafficking of the mutant.

We recognize that potential changes to the trafficking of C1ql2 caused by the K262E mutation would be important to characterize, in light of the reduced localization of the mutant protein at the SL in the in vivo experiments (Fig. 6e). In our culture system, C1ql2 and K262E were secreted by the HEK cells through insertion of an IgK signaling peptide at the N-terminus of the myc-tagged C1ql2/K262E. Thus, trafficking analysis on this system would not be informative, as the system is highly artificial compared to the in vivo model. Further studies are needed to characterize C1ql2 trafficking in neurons to understand how C1ql2-Nrxn3 interaction regulates the localization of C1ql2. However, labeling of the myc-tag in C1ql2 or K262E expressing HEK cells of the co-culture model reveals a similar signal for the two proteins (Fig. 5a,c). Nrxn-null mutation in neurons co-cultured with C1ql2-expressing HEK cells disrupted C1ql2 mediated vGlut1 accumulation in the neurons. Selective expression of Nrxn3(25b) in the Nrxn-null neurons restored vGlut1 clustering was (Fig. 5e-f). Together, these data suggest that it is the interaction between C1ql2 and Nrxn3 that drives the accumulation of vGlut1.

Figure 6: Bcl11b KO should also be included in 6f-h.

As suggested by reviewer #1, we included the Bcl11b cKO in figures 6f-h and in corresponding supplementary figures 5c-j.

Figure 7b: What is the abundance of mRNA for Nrxn1 and Nrxn2 as well as the abundance of Nrxns after EGFP-Cre injection into DG?

We addressed this point raised by reviewer #1 by quantifying the relative mRNA levels of Nrxn1 and Nrxn2 via qPCR upon Nrxn123 mutation induction with EGFP-Cre injection. We have now examined the mRNA levels of Nrxn1 and Nrxn2 upon stereotaxic injection of Cre in the DG of Nrxn123flox/flox animals and found that Nrxn1 was only mildly reduced. At the same time Nrxn2 showed a tendency for reduction that was not significant. The data are presented in supplementary figure 6a of the revised maunscript.

Minor Comments for readability:

Synapse score is referred to frequently in the text and should be defined within the text for clarification.

'n' numbers should be better defined in the figure legends. For example, for protein expression analysis in 1c, n=3. Is this a biological or technical triplicate? For electrophysiology (e.g. 3c), does "n=7" reflect the number of animals or the number of slices? n/N (slices/animals) should be presented.

Figure 7a: Should the diagrams of the cre viruses be EGFP-Inactive or active Cre and not CRE-EGFP as shown in the diagram?

Figure 7b: the region used for the inset should be identified in the larger image.

All minor points have been fixed in the revised manuscript according to the suggestions.

Reviewer #3 (Recommendations For The Authors):

-Please describe the 'synapse score' somewhere in the text - it is too prominently featured to not have a clear description of what it is.

The description of the synapse score has been included in the main text module of the revised manuscript.

-The claim that Bcl11b controls SV recruitment "specifically" through C1ql2 is a bit stronger than is warranted by the data. Particularly given that C1ql2 is expressed at 2.5X control levels in their rescue experiments. See pt.2

Please see response to reviewer #3 point 1 of public reviews. To address this, we over-expressed C1ql2 in control animals and found no changes in the synaptic vesicle distribution (supplementary figure 2g-j of revised manuscript). This supports that the observed rescue of synaptic vesicle recruitment by re-expression of C1ql2 is due to its physiological function and not due to the artificially elevated protein levels. Of course, we cannot exclude the possibility that other, C1ql2-independent, mechanisms also contribute to the SV recruitment downstream of Bcl11b. Our data from the C1ql2 rescue, C1ql2 KD, the in vitro experiments and the interruption of C1ql2-Nrxn3 in vivo, strongly suggest C1ql2 to be an important regulator of SV recruitment.

-Does Bcl11b regulate Nrxn3 expression? Considering the apparent loss of C1ql2 expression in the Nrxn KO mice, this is an important detail.

We agree with reviewer #3 that this is an important point. We have previously done differential transcriptomics from DG neurons of Bcl11b cKOs compared to controls and did not find Nrxn3 among the differentially expressed genes. To further validate this, we now quantified the Nrxn3 mRNA levels via qPCR in Bcl11b cKOs compared to controls and found no differences. These data are included in supplementary figure 5a of the revised manuscript.

-It appears that C1ql2 expression is much lower in the Nrxn123 KO mice. Since the authors are trying to test whether Nrxn3 is required for the correct targeting of C1ql2, this is a confounding factor. We can't really tell if what we are seeing is a "mistargeting" of C1ql2, loss of expression, or both. If the authors did a similar analysis to what they did in Figure 1 where they looked at the synaptic localization of C1ql2 (and quantified it) that could provide more evidence to support or refute the "mistargeting" claim.

Please also see response to reviewer #3 point 5 of public reviews. To exclude that reduction of fluorescence intensity of C1ql2 at the SL in Nrxn123 KO mice is due to loss of C1ql2 expression, we examined the mRNA levels of C1ql2 in control and Nrxn123 mutants and found no changes (data are included in supplementary figure 6b of the revised manuscript), suggesting that C1ql2 gene expression is normal. The reduced C1ql2 fluorescence intensity at the MFS was first observed when non-binding C1ql2 variant K262E was introduced to Bcl11b cKO mice that lack endogenous C1ql2 (Fig.6). In these experiments, we found that despite the overall high protein levels of C1ql2.K262E in the hippocampus (Fig. 6c), its fluorescence intensity at the SL was significantly reduced compared to WT C1ql2 (Fig. 6d-e). The remaining C1ql2.K262E signal in the SL was equally distributed and in a punctate form, similar to WT C1ql2. Together, this indicates that the loss of C1ql2-Nrxn3 interaction interferes with the localization of C1ql2 along the MFS, but not with expression of C1ql2. Of course, this does not exclude that additional mechanisms regulate C1ql2 localization at the synapse, as both the mutant C1ql2 in Bcl11b cKO and the endogenous C1ql2 in Nrxn123 cKO show residual immunofluorescence at the SL.

We note here that we have not previously quantified the co-localization of C1ql2 with individual synapses. C1ql2 is a secreted molecule that localizes at the MFS synaptic cleft. However, not much is known about the number of MFS that are positive for C1ql2 nor about the mechanisms regulating C1ql2 targeting, transport, and secretion to the MFS. Whether C1ql2 interaction with Nrxn3 is necessary for the protection of C1ql2 from degradation, its surface presentation and transport or stabilization to the synapse is currently unclear. Upon revision of our manuscript, we realized that we might have overstated this particular finding and have now rephrased the specific parts within the results to appropriately describe the observation and have also included a sentence in the discussion referring to the lack of understanding of the mechanism behind this observation.

-Title of Figure S5 is "Nrxn KO perturbs C1ql2 localization and SV recruitment at the MFS", but there is no data on C1ql2 localization.

This issue has been fixed in the revised manusript.

-S5 should be labeled more clearly than just Cre+/-

This issue has been fixed in the revised manuscript.

References

Castillo, P.E., Malenka, R.C., Nicoll, R.A., 1997. Kainate receptors mediate a slow postsynaptic current in hippocampal CA3 neurons. Nature 388, 182–186. https://doi.org/10.1038/40645

De Bruyckere, E., Simon, R., Nestel, S., Heimrich, B., Kätzel, D., Egorov, A.V., Liu, P., Jenkins, N.A., Copeland, N.G., Schwegler, H., Draguhn, A., Britsch, S., 2018. Stability and Function of Hippocampal Mossy Fiber Synapses Depend on Bcl11b/Ctip2. Front. Mol. Neurosci. 11. https://doi.org/10.3389/fnmol.2018.00103

Kaeser, P.S., Regehr, W.G., 2017. The readily releasable pool of synaptic vesicles. Curr. Opin. Neurobiol. 43, 63–70. https://doi.org/10.1016/j.conb.2016.12.012

Lerma, J., 2003. Roles and rules of kainate receptors in synaptic transmission. Nat. Rev. Neurosci. 4, 481–495. https://doi.org/10.1038/nrn1118

Orlando, M., Dvorzhak, A., Bruentgens, F., Maglione, M., Rost, B.R., Sigrist, S.J., Breustedt, J., Schmitz, D., 2021. Recruitment of release sites underlies chemical presynaptic potentiation at hippocampal mossy fiber boutons. PLoS Biol. 19, e3001149. https://doi.org/10.1371/journal.pbio.3001149

Vandael, D., Borges-Merjane, C., Zhang, X., Jonas, P., 2020. Short-Term Plasticity at Hippocampal Mossy Fiber Synapses Is Induced by Natural Activity Patterns and Associated with Vesicle Pool Engram Formation. Neuron 107, 509-521.e7. https://doi.org/10.1016/j.neuron.2020.05.013

Author response:

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

Public Reviews:

Reviewer 1:

Weaknesses:

The authors do not discuss based on genomic information; the genomes of the cichlids from the three lakes have been decoded and are therefore available. However, indeed, the species in Lake Tanganyika and Lake Malawi/Victoria are genetically distant from each other, so a comparative genome analysis would not have yielded the results presented here. I recommend adding such a discussion to the Discussion.

We appreciate your comment. We added the discussion regarding the genomic aspect of parallel evolution.

Line 386-393: “From a genomic perspective, several studies have investigated the genetic basis of hypertrophied lip cichlids (Masonick et al., 2023; Nakamura et al., 2021). Importantly, some Wnt pathway-related genes (tcf4 and daam2) and ECM-related genes (postna, col12a1a, and col12a1b) have been found to be under positive selection in cichlids with hypertrophied lips of Lake Victoria (see Nakamura et al., 2021 Table S3). For future research, examining whether these genes are under selection in other lakes is crucial to understand the genetic mechanisms underlying the parallel evolution of hypertrophied lips.”

Minor comments:

Line 30, the Wnt --> the genes in Wnt

We appreciate your comment. According to the comment, we corrected the sentence.

Line 30: “the Wnt signaling pathway” -> “the genes in Wnt signaling pathway”

Line 42-44, "It is considered that the same direction of natural selection drives phenotypic changes among species since it is unlikely that these complex phenotypes have been acquired repeatedly just by neutral evolution". How about "Since it is unlikely that such a complex phenotype was acquired repeatedly by neutral evolution alone, the same direction of natural selection among species is likely to drive the parallel phenotypic change."?

We agree with your suggestion and correct the sentence of our manuscript.

Line 42-44: “It is considered that the same direction of natural selection drives phenotypic changes among species since it is unlikely that these complex phenotypes have been acquired repeatedly just by neutral evolution”

“Since it is unlikely that such a complex phenotype was acquired repeatedly by neutral evolution alone, the same direction of natural selection among species is likely to drive the parallel phenotypic change”

Line 60, polygenic --> likely to be polygenic

We appreciate your comment. Indeed, it is better to weaken the wording.

Line 60: “most traits are polygenic” -> “most traits are likely to be polygenic”

Line 91, the Wnt --> the genes in Wnt

We appreciate your correction. Last paragraph of introduction has been corrected according to the suggestion of Reviewer 2 (Q1).

Line 230, NovaSeq --> Illumina NovaSeq

We appreciate your correction.

Line 222: “NovaSeq 6000” -> “Illumina NovaSeq 6000”

Line 231 "mRNA Library Prep Kit". Please add a company name.

We appreciate your correction. We added company’s information.

Line 223: “a TruSeq stranded mRNA Library Prep Kit.” -> “a TruSeq stranded mRNA Library Prep Kit (Illumina)”

Line 267, as for the tip of hypertrophied lips, could you add and point out which part is the tip?

We dissected hypertrophied lips in two half anterior and half posterior. We added the sentence in the materials and methods section.

Line 156-158: “The lips of H. chilotes were analyzed separately for the base and tip.” -> “The lips of H. chilotes were dissected in two half anterior (tip) and half posterior (base), which are analyzed separately.”

Line 272, "133 proteins upregulated and 5 proteins downregulated" in hypertrophied lip or normal lip?

We appreciate your correction. We added the sentence as follows.

Line 264: “133 proteins upregulated and 5 proteins downregulated”

“133 proteins upregulated and 5 proteins downregulated in the hypertrophied lip”

Line 274, "hypertrophied lips" means tip of hypertrophied lips?

We appreciate your correction. We corrected the sentence as follows.

Line 266: “hypertrophied lips are abundant” -> “tip of hypertrophied lips is abundant”

Line 277, Did you perform multiple testing correction for statistical significance?

We appreciate your comment about multiple testing corrections. We did not apply multiple testing corrections in our “exploratory” analysis of proteomics not to miss biologically important candidates in a limited sample size (n=3). We calculated the multiple corrected p-value in the Benjamini Hochberg method (Author response image 1, right). The result suggested that almost the same proteoglycans and its related proteins as we focused on are highly accumulated in the hypertrophied lips in milder conditions (significance level of 0.1).

Author response image 1.

Thus, our main conclusions remain unchanged even with correction applied, however, the overall balance of the volcano plot is not visually appealing (Author response image 1, right).

It is important to note that we selected the Top 20 proteins based on fold change rather than statistical significance. In addition, our proteomic findings show consistency with our histological and transcriptome data, providing the biological validation from various aspects. While we understand the potential benefits of multiple testing correction, our current approach without multiple testing still offers valuable and fair data to propose hypothesis on the molecular mechanisms of lip hypertrophy in cichlids. Therefore, we want to use original figure without multiple testing. We greatly appreciate the understanding of the reviewer.

Line 349-351, "The results of the enrichment analysis suggested that the genes that were categorized into both canonical and non-canonical Wnt signaling pathways, were highly expressed in the hypertrophied lips of juvenile and adult cichlids."

The wnt category was enriched by analyzing the highly expressed genes, so isn't it natural that the wnt category is highly expressed?

Did you mean to say as in the following sentence?

"Enrichment of genes categorized in the canonical and noncanonical Wnt signaling pathways suggested that high expression of genes in the Wnt signaling pathway is likely to be involved in the hypertrophied lips of juvenile and adult fish."

Thank you for your comments. We corrected our manuscript as follows.

Line 341-344: “The results of the enrichment analysis suggested that the genes that were categorized into both canonical and non-canonical Wnt signaling pathways, were highly expressed in the hypertrophied lips of juvenile and adult cichlids.”

“As a result of enrichment analysis, DEGs were categorized in the canonical and noncanonical Wnt signaling pathways, suggesting that high expression of genes in the Wnt signaling pathway is likely to be involved in the hypertrophied lips of juvenile and adult fish.”

Line 403-404, "several other pathways may be involved in the development of hypertrophied lips". Do you have any evidence?

We appreciate your comment regarding possible evidence for the involvement of multiple pathways in hypertrophied lip development. Our statement was based on two main points:

(1) While we highlighted the Wnt pathway because this pathway is known to increase proteoglycan expression, we cannot exclude the possibility of the involvement of other pathways. For instance, our enrichment analysis in adult cichlids identified VEGF-related pathways, which could contribute to lip hypertrophy by increasing vascularization and nutrient supply to the lip tissue.

(2) Previous quantitative trait locus (QTL) analysis by Henning et al. (2017) concluded that lip hypertrophy is likely influenced by numerous loci with small additive effects. This indicates that lip hypertrophy is a complex phenotype consisted of multiple genetic factors, some which probably correspond to different molecular pathways.

Given these points, we draw a conclusion that emphasize the importance of Wnt pathway while also recognizing the potential cooperative interaction of multiple pathways in developing lip hypertrophy. Without confusing the two statements, we corrected our manuscript as follows.

Line 398-412: “We uncovered the apparent relationships between hypertrophied lips and the expression profiles of ECM proteins, in particularly proteoglycans. The trends for the overall expression of ECM-related genes were similar across hypertrophied lip species, but we rarely observed a specific gene that was commonly expressed at high or low levels in all three examples of hypertrophied lips across all East African Great Lakes. Furthermore, although we focused primarily on the relationship between the Wnt signaling pathway and lip hypertrophy, several other pathways may be involved in the development of hypertrophied lips. These findings imply that although enlargement of proteoglycan-rich loose connective tissue is common in hypertrophied lips, the developmental pathways to accomplish this are diverse in each lake.”

“We uncovered the apparent relationships between hypertrophied lips and the expression profiles of ECM proteins, in particularly proteoglycans. The trends for the overall expression of ECM-related genes were similar across hypertrophied lip species, but we rarely observed a specific gene that was commonly expressed at high or low levels in all three examples of hypertrophied lips across all East African Great Lakes. Furthermore, although we focused primarily on the relationship between the Wnt signaling pathway and lip hypertrophy, several other pathways may be involved in the development of hypertrophied lips. For example, our enrichment analysis in adult cichlids identified VEGF-related pathways, which could contribute to lip hypertrophy by increasing vascularization and nutrient supply to the lip tissue. In addition, previous quantitative trait locus (QTL) analysis by Henning et al. (2017) concluded that lip hypertrophy is likely influenced by numerous loci with small additive effects. These lines of data imply that although enlargement of proteoglycan-rich loose connective tissue is common in hypertrophied lips, the developmental pathways to accomplish this are diverse in each lake.”

Reviewer 2:

Minor comments:

Last paragraph of Introduction: Remove the results of this study.

We appreciate your suggestion. We remove the specialized results from the last paragraph.

“In this study, we comprehensively compared the hypertrophied lips of cichlids across all East African Great Lakes using histology, proteomics, and transcriptomics. Histological and proteomic analyses revealed a distinct microstructure of hypertrophied lips compared to normal lips, and primary candidate proteins were identified. Transcriptome analysis at different developmental stages showed that the genes in Wnt signaling pathway was highly expressed in cichlids with hypertrophied lips at both the juvenile and adult stages. It is noteworthy that the distinct expression profiles observed in the proteome and transcriptome analyses of hypertrophied lips were similar among cichlids from each of the East African Great Lakes. The present study, which integrates comprehensive analyses for cichlids from all East African Great Lakes, provides insight for a better understanding of the molecular basis of a typical example of parallel evolution.”

Line 87-91: “In this study, we comprehensively compared the hypertrophied and normal lips of cichlids across all East African Great Lakes at various biological levels using histology, proteomics, and transcriptomics. As a result, we showed that a novel key pathway commonly involved in the formation of hypertrophied lips, providing insight into a better understanding of the molecular basis of a typical example of parallel evolution.”

Line 156: Italicize the scientific names.

We appreciate your correction.

Line 148: “M. zebra and O. niloticus” -> “M. zebra and O. niloticus”

Line 261: Remove the period after "Victoria."

We appreciate your correction.

Line 253: “Lake Victoria. (Figure 1; Figure S2).” -> “Lake Victoria (Figure 1; Figure S2).”

Line 416: Remove the period after "tissue."

We appreciate your correction.

Line 420: “tissue. (A,B)” -> “tissue (A,B)”

Line 646: Probably "the anterior side to the left."

We apologize for our mistake. As you commented, the anterior side is left. We corrected our manuscript as follows.

Line 648: “the anterior side to the right” -> “the anterior side to the left”

Fig. S2: Based on Fig. 1, the VG stained area appears larger in the Hypertrophied lip species; however, it is the opposite in Fig. S2.

We appreciate your comments. This is because we calculated the ratio of the VG-stained area to the whole lip area. While the absolute VG-stained area is larger in hypertrophied lips, the proportion of the VG-stained area relative to the total lip area is smaller. This correction using entire area allows us to simply compare the degree of lip hypertrophy among species.

Author response:

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

Gating of Kv10 channels is unique because it involves coupling between non-domain swapped voltage sensing domains, a domain-swapped cytoplasmic ring assembly formed by the N- and C-termini, and the pore domain. Recent structural data suggests that activation of the voltage sensing domain relieves a steric hindrance to pore opening, but the contribution of the cytoplasmic domain to gating is still not well understood. This aspect is of particular importance because proteins like calmodulin interact with the cytoplasmic domain to regulate channel activity. The effects of calmodulin (CaM) in WT and mutant channels with disrupted cytoplasmic gating ring assemblies are contradictory, resulting in inhibition or activation, respectively. The underlying mechanism for these discrepancies is not understood. In the present manuscript, Reham Abdelaziz and collaborators use electrophysiology, biochemistry and mathematical modeling to describe how mutations and deletions that disrupt inter-subunit interactions at the cytoplasmic gating ring assembly affect Kv10.1 channel gating and modulation by CaM. In the revised manuscript, additional information is provided to allow readers to identify within the Kv10.1 channel structure the location of E600R, one of the key channel mutants analyzed in this study. However, the mechanistic role of the cytoplasmic domains that this study focuses on, as well as the location of the ΔPASCap deletion and other perturbations investigated in the study remain difficult to visualize without additional graphical information. This can make it challenging for readers to connect the findings presented in the study with a structural mechanism of channel function.

The authors focused mainly on two structural perturbations that disrupt interactions within the cytoplasmic domain, the E600R mutant and the ΔPASCap deletion. By expressing mutants in oocytes and recording currents using Two Electrode Voltage-Clamp (TEV), it is found that both ΔPASCap and E600R mutants have biphasic conductance-voltage (G-V) relations and exhibit activation and deactivation kinetics with multiple voltage-dependent components. Importantly, the mutant-specific component in the G-V relations is observed at negative voltages where WT channels remain closed. The authors argue that the biphasic behavior in the G-V relations is unlikely to result from two different populations of channels in the oocytes, because they found that the relative amplitude between the two components in the G-V relations was highly reproducible across individual oocytes that otherwise tend to show high variability in expression levels. Instead, the G-V relations for all mutant channels could be well described by an equation that considers two open states O1 and O2, and a transition between them; O1 appeared to be unaffected by any of the structural manipulations tested (i.e. E600R, ΔPASCap, and other deletions) whereas the parameters for O2 and the transition between the two open states were different between constructs. The O1 state is not observed in WT channels and is hypothesized to be associated with voltage sensor activation. O2 represents the open state that is normally observed in WT channels and is speculated to be associated with conformational changes within the cytoplasmic gating ring that follow voltage sensor activation, which could explain why the mutations and deletions disrupting cytoplasmic interactions affect primarily O2. 

Severing the covalent link between the voltage sensor and pore reduced O1 occupancy in one of the deletion constructs. Although this observation is consistent with the hypothesis that voltage-sensor activation drives entry into O1, this result is not conclusive. Structural as well as functional data has established that the coupling of the voltage sensor and pore does not entirely rely on the S4-S5 covalent linker between the sensor and the pore, and thus the severed construct could still retain coupling through other mechanisms, which is consistent with the prominent voltage dependence that is observed. If both states O1 and O2 require voltage sensor activation, it is unclear why the severed construct would affect state O1 primarily, as suggested in the manuscript, as opposed to decreasing occupancy of both open states. In line with this argument, the presence of Mg2+ in the extracellular solution affected both O1 and O2. This finding suggests that entry into both O1 and O2 requires voltage-sensor activation because Mg2+ ions are known to stabilize the voltage sensor in its most deactivated conformations. 

We agree with the reviewer that access to both states requires a conformational change in the voltage sensor. This was stated in our revised article: “In contrast, to enter O2, all subunits must complete both voltage sensor transitions and the collective gating ring transition.” We interpret the two gating steps as sequential; the effective rotation of the intracellular ring would happen only once the sensor is in its fully activated position.

We also agree that the S4-S5 segment cannot be the only interaction mechanism, as we demonstrated in our earlier work (Lörinczi et al., 2015; Tomczak et al., 2017).  

Activation towards and closure from O1 is slow, whereas channels close rapidly from O2. A rapid alternating pulse protocol was used to take advantage of the difference in activation and deactivation kinetics between the two open components in the mutants and thus drive an increasing number of channels towards state O1. Currents activated by the alternating protocol reached larger amplitudes than those elicited by a long depolarization to the same voltage. This finding is interpreted as an indication that O1 has a larger macroscopic conductance than O2. In the revised manuscript, the authors performed single-channel recordings to determine why O1 and O2 have different macroscopic conductance. The results show that at voltages where the state O1 predominates, channels exhibited longer open times and overall higher open probability, whereas at more depolarized voltages where occupancy of O2 increases, channels exhibited more flickery gating behavior and decreased open probability. These results are informative but not conclusive because additional details about how experiments were conducted, and group data analysis are missing. Importantly, results showing inhibition of single ΔPASCap channels by a Kv10-specific inhibitor are mentioned but not shown or quantitated - these data are essential to establish that the new O1 conductance indeed represents Kv10 channel activity.

We observed the activity of a channel compatible with Kv10.1 ΔPAS-Cap (long openings at low-moderate potentials, very short flickery activity at strong depolarizations) in 12 patches from oocytes obtained from different frog operations over a period of two and a half months once the experimental conditions could be established. As stated in the text, we did not proceed to generate amplitude histograms because we could not resolve clear single-channel events at strong depolarizations. Astemizole abolished the activity and (remarkably) strongly reduced the noise in traces at strong depolarizations, which we interpret as partially caused by flicker openings.

Author response image 1.

We include two example recordings of Astemizole application (100µM) on two different patches. Both recordings are performed at -60 mV (to decrease the likelihood that the channel visits O2) with 100 mM internal and 60 mM external K+. In both cases, the traces in Astemizole are presented in red.

It is shown that conditioning pulses to very negative voltages result in mutant channel currents that are larger and activate more slowly than those elicited at the same voltage but starting from less negative conditioning pulses. In voltage-activated curves, O1 occupancy is shown to be favored by increasingly negative conditioning voltages. This is interpreted as indicating that O1 is primarily accessed from deeply closed states in which voltage sensors are in their most deactivated position. Consistently, a mutation that destabilizes these deactivated states is shown to largely suppress the first component in voltage-activation curves for both ΔPASCap and E600R channels.

The authors then address the role of the hidden O1 state in channel regulation by calmodulation. Stimulating calcium entry into oocytes with ionomycin and thapsigarging, assumed to enhance CaM-dependent modulation, resulted in preferential potentiation of the first component in ΔPASCap and E600R channels. This potentiation was attenuated by including an additional mutation that disfavors deeply closed states. Together, these results are interpreted as an indication that calcium-CaM preferentially stabilizes deeply closed states from which O1 can be readily accessed in mutant channels, thus favoring current activation. In WT channels lacking a conducting O1 state, CaM stabilizes deeply closed states and is therefore inhibitory. It is found that the potentiation of ΔPASCap and E600R by CaM is more strongly attenuated by mutations in the channel that are assumed to disrupt interaction with the C-terminal lobe of CaM than mutations assumed to affect interaction with the N-terminal lobe. These results are intriguing but difficult to interpret in mechanistic terms. The strong effect that calcium-CaM had on the occupancy of the O1 state in the mutants raises the possibility that O1 can be only observed in channels that are constitutively associated with CaM. To address this, a biochemical pull-down assay was carried out to establish that only a small fraction of channels are associated with CaM under baseline conditions. These CaM experiments are potentially very interesting and could have wide physiological relevance. However, the approach utilized to activate CaM is indirect and could result in additional nonspecific effects on the oocytes that could affect the results.

Finally, a mathematical model is proposed consisting of two layers involving two activation steps for the voltage sensor, and one conformational change in the cytoplasmic gating ring - completion of both sets of conformational changes is required to access state O2, but accessing state O1 only requires completion of the first voltage-sensor activation step in the four subunits. The model qualitatively reproduces most major findings on the mutants. Although the model used is highly symmetric and appears simple, the mathematical form used for the rate constants in the model adds a layer of complexity to the model that makes mechanistic interpretations difficult. In addition, many transitions that from a mechanistic standpoint should not depend on voltage were assigned a voltage dependence in the model. These limitations diminish the overall usefulness of the model which is prominently presented in the manuscript. The most important mechanistic assumptions in the model are not addressed experimentally, such as the proposition that entry into O1 depends on the opening of the transmembrane pore gate, whereas entry into O2 involves gating ring transitions - it is unclear why O2 would require further gating ring transitions to conduct ions given that the gating ring can already support permeation by O1 without any additional conformational changes.

In essence, we agree with the reviewer; we already have addressed these points in our revised article:

Regarding the voltage dependence we write “the κ/λ transition could reasonably be expected to be voltage independent because we related it to ring reconfiguration, a process that should occur as a consequence of a prior VSD transition. We have made some attempts to treat this transition as voltage independent but state-specific with upper-layer bias for states on the right and lower-layer bias for states on the left. This is in principle possible, as can already be gleaned from the similar voltage ranges of the left-right transition (α/β) and the κL/λ transition. However, this approach leads to a much larger number of free, less well constrained kinetic parameters and drastically complicated the parameter search. ” As you can see, we also formulated a strategy to free the model of the potentially spurious voltage dependence and (in bold here) explained why we did not follow this route in this study. 

Regarding the need for gating ring transitions after O1, we wrote, “Thus, the underlying gating events can be separated into two steps: The first gating step involves only the voltage sensor without engaging the ring and leads to a pre-open state, which is non-conducting in the WT but conducting in our mutants. The second gating event operates at higher depolarizations, involves a change in the ring, and leads to an open state both in WT and in the mutants. ” 

We interpret your statements such that you expect the conducting state to remain available once O1 is reached. However, the experimental evidence speaks against that the pore availability remains regardless of the further gating steps beyond O1. The description of model construction is informative here: “... we could exclude many possible [sites at which O1 connects to closed states] because the attachment site must be sufficiently far away from the conventional open state [O2]. Otherwise, the transition from "O1 preferred" to "O2 preferred" via a few closed intermediate states is very gradual and never produces the biphasic GV curves [that we observed]. ” 

In other words, voltage-dependent gating steps beyond the state that offers access to O1 appear to close the pore, after it was open. That might occur because only then (for states in which at least one voltage sensor exceeded the intermediate position) the ring is fixed in a particular state until all sensors completed activation. In the WT, closing the pore in deactivated states might rely on an interaction that is absent in the mutant because, at least in HERG: “the interaction between the PAS domain and the C-terminus is more stable in closed than in open KV11.1 (HERG) channels, and a single chain antibody binding to the interface between PAS domain and CNBHD can access its epitope in open but not in closed channels, strongly supporting a change in conformation of the ring during gating ”

Reviewer #3 (Public Review):

In the present manuscript, Abdelaziz and colleagues interrogate the gating mechanisms of Kv10.1, an important voltage-gated K+ channel in cell cycle and cancer physiology. At the molecular level, Kv10.1 is regulated by voltage and Ca-CaM. Structures solved using CryoEM for Kv10.1 as well as other members of the KCNH family (Kv11 and Kv12) show channels that do not contain a structured S4-S5 linker imposing therefore a non-domain swapped architecture in the transmembrane region. However, the cytoplasmatic N- and C- terminal domains interact in a domain swapped manner forming a gating ring. The N-terminal domain (PAS domain) of one subunit is located close to the intracellular side of the voltage sensor domain and interacts with the C-terminal domain (CNBHD domain) of the neighbor subunit. Mutations in the intracellular domains has a profound effect in the channel gating. The complex network of interactions between the voltage-sensor and the intracellular domains makes the PAS domain a particularly interesting domain of the channel to study as responsible for the coupling between the voltage sensor domains and the intracellular gating ring.

The coupling between the voltage-sensor domain and the gating ring is not fully understood and the authors aim to shed light into the details of this mechanism. In order to do that, they use well established techniques such as site-directed mutagenesis, electrophysiology, biochemistry and mathematical modeling. In the present work, the authors propose a two open state model that arises from functional experiments after introducing a deletion on the PAS domain (ΔPAS Cap) or a point mutation (E600R) in the CNBHD domain. The authors measure a bi-phasic G-V curve with these mutations and assign each phase as two different open states, one of them not visible on the WT and only unveiled after introducing the mutations.

The hypothesis proposed by the authors could change the current paradigm in the current understanding for Kv10.1 and it is quite extraordinary; therefore, it requires extraordinary evidence to support it.

STRENGTHS: The authors use adequate techniques such as electrophysiology and sitedirected mutagenesis to address the gating changes introduced by the molecular manipulations. They also use appropriate mathematical modeling to build a Markov model and identify the mechanism behind the gating changes.

WEAKNESSES: The results presented by the authors do not fully support their conclusions since they could have alternative explanations. The authors base their primary hypothesis on the bi-phasic behavior of a calculated G-V curve that do not match the tail behavior, the experimental conditions used in the present manuscript introduce uncertainties, weakening their conclusions and complicating the interpretation of the results. Therefore, their experimental conditions need to be revisited. 

We respectfully disagree. We think that your suggestions for alternative explanations are addressed in the current version of the article. We will rebut them once more below, but we feel the need to point out that our arguments are already laid out in the revised article.

I have some concerns related to the following points:

(1) Biphasic gating behavior

The authors use the TEVC technique in oocytes extracted surgically from Xenopus Leavis frogs. The method is well established and is adequate to address ion channel behavior. The experiments are performed in chloride-based solutions which present a handicap when measuring outward rectifying currents at very depolarizing potentials due to the presence of calcium activated chloride channel expressed endogenously in the oocytes; these channels will open and rectify chloride intracellularly adding to the outward rectifying traces during the test pulse. The authors calculate their G-V curves from the test pulse steady-state current instead of using the tail currents. The conductance measurements are normally taken from the 'tail current' because tails are measured at a fix voltage hence maintaining the driving force constant. 

We respectfully disagree. In contrast to other channels, like HERG, a common practice for Kv10 is not to use tail currents. It is long known that in this channel, tail currents and test-pulse steady-state currents can appear to be at odds because the channels deactivate extremely rapidly, at the border of temporal resolution of the measurements and with intricate waveforms. This complicates the estimation of the instantaneous tail current. Therefore, the outward current is commonly used to estimate conductance (Terlau et al., 1996; Schönherr et al., 1999; Schönherr et al., 2002; Whicher and MacKinnon, 2019), while the latter authors also use the extreme of the tail for some mutants.

Due to their activation at very negative voltage, the reversal potential in our mutants can be measured directly; we are, therefore, more confident with this approach. Nevertheless, we have determined the initial tail current in some experiments. The behavior of these is very similar to the average that we present in Figure 1. The biphasic behavior is unequivocally present.

Author response image 2.

Calculating the conductance from the traces should not be a problem, however, in the present manuscript, the traces and the tail currents do not agree. 

The referee’s observation is perfectly in line with the long-standing experience of several labs working with KV10: tail current amplitudes in KV10 appear to be out of proportion for the WT open state (O2). Importantly, this is due to the rapid closure, which is not present in O1. As a consequence, the initial amplitude of tail currents from O1 are easier to estimate correctly, and they are much more obvious in the graphs. Taken together, these differences between O1 and O2 explain the misconception the reviewer describes next.

The tail traces shown in Fig1E do not show an increasing current amplitude in the voltage range from +50mV to +120mV, they seem to have reached a 'saturation state', suggesting that the traces from the test pulse contain an inward chloride current contamination. 

As stated in the text and indicated in Author response image 3, the tail currents In Figure 1E increase in amplitude between +50 and +120 mV, as can be seen in the examples below from different experiments (+50 is presented in black, +120 in red). As stated above, the increase is not as evident as in traces from other mutants because the predominance of O2 also implies a much faster deactivation.

Author response image 3. 

We are aware that Ca2+-activated Cl- currents can represent a problem when interpreting electrophysiological data in oocytes. In fact, we show in Supplement 1 to Figure 8 that this can be the case during the Ca2+-CaM experiments, where the increase in Ca2+ would certainly augment Cl- contribution to the outward current. This is why we performed these experiments in Cl--free solutions. As we show in Figure 8, the biphasic behavior was also present in those experiments. 

Importantly, Cl- free bath solutions would not correct contamination during the tail, since this would correspond to Cl- exiting the oocyte. Yet, if there would be contamination of the outward currents by Cl-, one would expect it to increase with larger depolarizations as the typical Ca2+activated Cl- current in oocytes does. As the reviewer states, this does not seem to be the case.

In addition, this second component identified by the authors as a second open state appears after +50mV and seems to never saturate. The normalization to the maximum current level during the test pulse, exaggerates this second component on the calculated G-V curve. 

We agree that this second component continues to increase; the reviewer brought this up in the first review, and we have already addressed this in our reply and in the discussion of the revised version: “This flicker block might also offer an explanation for a feature of the mutant channels, that is not explained in the current model version: the continued increase in current amplitude, hundreds of milliseconds into a strong depolarization (Supp. 4 to Fig. 9). If the relative stability of O2 and C2 continued to change throughout depolarization, such a current creep-up could be reproduced. However, this would require either the introduction of further layers of On ↔Cn states, or a non-Markovian modification of the model’s time evolution.” With non-Markovian, we mean a Langevin-type diffusive process. 

It's worth noticing that the ΔPASCap mutant experiments on Fig 5 in Mes based solutions do not show that second component on the G-V.

For the readers of this conversation, we would like to clarify that the reviewer likely refers to experiments shown in Fig. 5 of the initial submission but shown in Fig. 6 of the revised version (“Hyperpolarization promotes access to a large conductance, slowly activating open state.” Fig. 5 deals with single channels). We agree that these data look different, but this is because the voltage protocols are completely different (compare Fig. 6A (fixed test pulse, varied prepulse) and Fig. 2A (varied test pulse, fixed pre-pulse). Therefore, no biphasic behavior is expected. 

Because these results are the foundation for their two open state hypotheses, I will strongly suggest the authors to repeat all their Chloride-based experiments in Mes-based solutions to eliminate the undesired chloride contribution to the mutants current and clarify the contribution of the mutations to the Kv10.1 gating.

In summary, we respectfully disagree with all concerns raised in point (1). Our detailed arguments rebutting them are given above, but there is a more high-level concern about this entire exchange: the referee casts doubt on observations that are not new. Several labs have reported for a group of mutant KCNH channels: non-monotonic voltage dependence of activation (see, e.g., Fig. 6D in Zhao et al., 2017), multi-phasic tail currents (see e.g. Fig. 4A in Whicher and MacKinnon, 2019, in CHO cells where Cl- contamination is not a concern), and activation by high [Ca2+]i (Lörinczi et al., 2016). Our study replicates those observations and hypothesizes that the existence of an additional conducting state can alone explain all previously unexplained observations. We highlight the potency of this hypothesis with a Markov model that qualitatively reproduces all phenomena. We not only factually disagree with the individual points raised, but we also think that they don't touch on the core of our contribution

(2) Two step gating mechanism.

The authors interpret the results obtained with the ΔPASCap and the E600R as two step gating mechanisms containing two open states (O1 and O2) and assign them to the voltage sensor movement and gating ring rotation respectively. It is not clear, however how the authors assign the two open states.

The results show how the first component is conserved amongst mutations; however, the second one is not. The authors attribute the second component, hence the second open state to the movement of the gating ring. This scenario seems unlikely since there is a clear voltagedependence of the second component that will suggest an implication of a voltage-sensing current.

We do not suggest that the gating ring motion is not voltage dependent. We would like to point out that voltage dependence can be conveyed by voltage sensor coupling to the ring; this is the widely accepted theory of how the ring can be involved. Should the reviewer mean it in a narrow sense, that the model should be constructed such that all voltage-dependent steps occur before and independently of ring reconfiguration and that only then an additional step that reflects the (voltage-independent) reconfiguration solely, we would like to point the reviewer to the article, where we write: “the κ/λ transition could reasonably be expected to be voltage independent because we related it to ring reconfiguration, a process that should occur as a consequence of a prior VSD transition. We have made some attempts to treat this transition as voltage independent but state-specific with upper-layer bias for states on the right and lower-layer bias for states on the left. This is in principle possible, as can already be gleaned from the similar voltage ranges of the left-right transition (α/β) and the κL/λ transition. However, this approach leads to a much larger number of free, less well constrained kinetic parameters and drastically complicated the parameter search. ” As you can see, we also formulated a strategy to free the model from the potentially spurious voltage dependence and (in bold here) explained why we did not follow this route in this study. 

The split channel experiment is interesting but needs more explanation. I assume the authors expressed the 2 parts of the split channel (1-341 and 342-end), however Tomczak et al showed in 2017 how the split presents a constitutively activated function with inward currents that are not visible here, this point needs clarification.

As stated in the panel heading, the figure legend, and the main text, we did not use 1-341 and 342-end as done in Tomczak et al. Instead, “we compared the behavior of ∆2-10 and ∆210.L341Split,”. Evidently, the additional deletion (2-10) causes a shift in activation that explains the difference you point out. However, as we do not compare L341Split and ∆210.L341Split but ∆2-10 and ∆2-10.L341Split, our conclusion remains that “As predicted, compared to ∆2-10, ∆2-10.L341Split showed a significant reduction in the first component of the biphasic GV (Fig. 2C, D).” Remarkably, the behavior of the ∆3-9 L341Split described in Whicher and MacKinnon, 2019 (Figure 5) matches that of our ∆2-10 L341Split, which we think reinforces our case.

Moreover, the authors assume that the mutations introduced uncover a new open state, however the traces presented for the mutations suggest that other explanations are possible. Other gating mechanisms like inactivation from the closed state, can be introduced by the mutations. The traces presented for ΔPASCap but specially E600R present clear 'hooked tails', a direct indicator of a populations of inactive channels during the test pulse that recover from inactivation upon repolarization (Tristani-Firouzi M, Sanguinetti MC. J Physiol. 1998). 

There is a possibility that we are debating nomenclature here. In response to the suggestion that all our observations could be explained by inactivation, we attempted a disambiguation of terms in the reply and the article. As the argument is brought up again without reference to our clarification attempts, we will try to be more explicit here:

If, starting from deeply deactivated states, an open state is reached first, and then, following further activation steps, closed states are reached, this might be termed “inactivation”. In such a reading, our model features many inactivated states. The shortest version of such a model is C-O-I. It is for instance used by Raman and Bean (2001; DOI: 10.1016/S00063495(01)76052-3) to explain NaV gating in Purkinje neurons. If “inactivation” is meant in the sense that a gating transition exists, which is orthogonal to an activation/deactivation axis, and that after this orthogonal transition, an open state cannot be reached anymore, then all of the upper floor in our model is inactivated with respect to the open state O1. Finally, the state C2 is an inactivated state to O2. In this view, “inactivation” explains the observed phenomena. 

However, we must disagree if the referee means that a parsimonious explanation exists in which a single conducting state is the only source for all observed currents.   

There is a high-level reason: we found a single assumption that explains three different phenomena, while the inactivation hypothesis with one conducting state cannot explain one of them (the increase of the first component under raised CaM). But there is also a low-level reason: the tails in Tristani-Firouzi and Sanguinetti 1998 are fundamentally different from what we report herein in that they lack a third component. Thus, those tails are consistent with recovery from inactivation through a single open state, while a three-component tail is not. In the framework of a Markov model, the time constants of transitions from and to a given state (say O2), cannot change unless the voltage changes. During the tail current, the voltage does not change, yet we observe: 

i) a rapid decrease with a time constant of at most a few milliseconds (Fig 9 S2, 1-> 2),  ii) a slow increase in current, peaking after approximately 25 milliseconds and iii) a relaxation to zero current with a time constant of >50 ms. 

According to the reviewer’s suggestion, these processes on three timescales should all be explained by depopulating and repopulating the same open state while all rates are constant. There might well be a complicated multi-level state diagram with a single open state with different variants, like (open and open inactivated) that could produce triphasic tails with these properties if the system had not reached a steady state distribution at the end of the test pulse. It cannot, however, achieve it from an equilibrated system, and certainly, it cannot at the same time produce “biphasic activation” and “activation by CaM”. 

The results presented by the authors can be alternatively explained with a change in the equilibrium between the close to inactivated/recovery from inactivation to the open state. 

Again, we disagree. The model construction explains in detail that the transition from the first to the second phase is not gradual. Shifting equilibria cannot reproduce this. We have extensively tested that idea and can exclude this possibility.

Finally, the authors state that they do not detect "cumulative inactivation after repeated depolarization" but that is considering inactivation only from the open state and ignoring the possibility of the existence of close state inactivation or, that like in hERG, that the channel inactivates faster that what it activates (Smith PL, Yellen G. J Gen Physiol. 2002). 

We respectfully disagree. We explicitly model an open state that inactivates faster (O2->C2) than it activates. Once more, this is stated in the revised article, which we point to for details. Again, this alternative mechanism does not have the potential to explain all three effects. As discussed above about the chloride contamination concerns, this inactivation hypothesis was mentioned in the first review round and, therefore, addressed in our reply and the revised article. We also explained that “inactivation” has no specific meaning in Markov models. In the absence of O1, all transitions towards the lower layer are effectively “inactivation from closed states”, because they make access to the only remaining open state less likely”. But this is semantics. What is relevant is that no network of states around a single open state can reproduce the three effets in a more parsimonious way than the assumption of the second open state does.

(3) Single channel conductance.

The single channels experiments are a great way to assess the different conductance of single channel openings, unfortunately the authors cannot measure accurately different conductances for the two proposed open states. The Markov Model built by the authors, disagrees with their interpretation of the experimental results assigning the exact same conductance to the two modeled open states. To interpret the mutant data, it is needed to add data with the WT for comparison and in presence of specific blockers. 

We respectfully disagree. As previously shown, the conductance of the flickering wild-type open state is very difficult to resolve. Our recordings do not show that the two states have different single-channel conductances, and therefore the model assumes identical singlechannel conductance. 

The important point is that the single-channel recordings clearly show two different gating modes associated with the voltage ranges in which we predict the two open states. One has a smaller macroscopic current due to rapid flickering (aka “inactivation”). These recordings are another proof of the existence of two open states because the two gating modes occur.  Wild-type data can be found in Bauer and Schwarz, (2001, doi:10.1007/s00232-001-0031-3) or Pardo et al., (1998, doi:10.1083/jcb.143.3.767) for comparison.

We appreciate the effort editors and reviewers invested in assessing the revised manuscript. Yet, we think that the demanded revision of experimental conditions and quantification methods contradicts the commonly accepted practice for KV10 channels. Some of the reviewer comments are skeptical about the biphasic behavior, which is an established and replicated finding for many mutants and by many researchers. The alternative explanations for these disbelieved findings are either “semantics” or cannot quantitatively explain the measurements. Therefore, only the demand for more explanations and unprecedented resolution in singlechannel recordings remains. We share these sentiments.

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

(1) The authors must show that the second open state is not just an artifact of endogenous activity but represents the activity of the same EAG channels. I suggest that the authors repeat these experiments in Mes-based solutions. 

(2) Along the same lines, it is necessary to show that these currents can be blocked using known EAG channel blockers such as astemizole. Ultimately, it will be important to demonstrate using single-channel analysis that these do represent two distinct open states separated by a closed state. 

We have addressed these concerns using several approaches. The most substantial change is the addition of single-channel recordings on ΔPASCap. In those experiments, we could provide evidence of the two types of events in the same patch, and the presence of an outward current at -60 mV, 50 mV below the equilibrium potential for chloride. The channels were never detected in uninjected oocytes, and Astemizole silenced the activity in patches containing multiple channels. These observations, together with the maintenance of the biphasic behavior that we interpret as evidence of the presence of O1 in methanesulfonate-based solutions, strongly suggest that both O1 and O2 obey the expression of KV10.1 mutants.

(3) Currents should be measured by increasing the pulse lengths as needed in order to obtain the true steady-state G-V curves. 

We agree that the endpoint of activation is ill-defined in the cases where a steady-state is not reached. This does indeed hamper quantitative statements about the relative amplitude of the two components. However, while the overall shape does change, its position (voltage dependence) would not be affected by this shortcoming. The data, therefore, supports the claim of the “existence of mutant-specific O1 and its equal voltage dependence across mutants.”

(4) A more clear and thorough description should be provided for how the observations with the mutant channels apply to the behavior of WT channels. How exactly does state O1 relate to WT behavior, and how exactly do the parameters of the mathematical model differ between WT and mutants? How can this be interpreted at a structural level? What could be the structural mechanism through which ΔPASCap and E600R enable conduction through O1? It seems contradictory that O1 would be associated exclusively with voltage-sensor activation and not gating ring transitions, and yet the mutations that enable cation access through O1 localize at the gating ring - this needs to be better clarified. 

We have undertaken a thorough rewriting of all sections to clarify the structural correlates that may explain the behavior of the mutants. In brief, we propose that when all four voltage sensors move towards the extracellular side, the intracellular ring maintains the permeation path closed until it rotates. If the ring is altered, this “lock” is incompetent, and permeation can be detected (page 34). By fixing the position of the ring, calmodulin would preclude permeation in the WT and promote the population of O1 in the mutants.

(5) Rather than the t80% risetime, exponential fits should be performed to assess the kinetics of activation. 

We agree that the assessment of kinetics by a t80% is not ideal. We originally refrained from exponential fits because they introduce other issues when used for processes that are not truly exponential (as is the case here). We had planned to perform exponential fits in this revised version, but because the activation process is not exponential, the time constants we could provide would not be accurate, and the result would remain qualitative as it is now. In the experiments where we did perform the fits (Fig. 3), the values obtained support the statement made. 

(6) It is argued based on the G-V relations in Figure 2A that none of the mutations or deletions introduced have a major effect on state O1 properties, but rather affect state O2. However, the occupancy of state O2 is undetermined because activation curves do not reach saturation. It would be interesting to explore the fitting parameters on Fig.2B further to test whether the data on Fig 2A can indeed only be described by fits in which the parameters for O1 remain unchanged between constructs. 

We agree that the absolute occupancy of O2 cannot be properly determined if a steady state is not reached. This is, however, a feature of the channel. During very long depolarizations in WT, the current visually appears to reach a plateau, but a closer look reveals that the current keeps increasing after very long depolarizations (up to 10 seconds; see, e.g., Fig. 1B in Garg et al., 2013, Mol Pharmacol 83, 805-813. DOI: 10.1124/mol.112.084384). Interestingly, although the model presented here does not account for this behavior, we propose changes in the model that could. “If the relative stability of O2 and C2 continued to change throughout the depolarization such a current creep-up could be reproduced. However, this would require either the introduction of further layers of On↔Cn states or a non-Markovian modification of the model’s evolution.” Page 34.

(7) The authors interpret the results obtained with the mutants DPASCAP and E600R -tested before by Lorinczi et al. 2016, to disrupt the interactions between the PASCap and cNBHD domains- as a two-step gating mechanism with two open states. All the results obtained with the E600R mutant and DPASCap could also be explained by inactivation/recovery from inactivation behavior and a change in the equilibrium between the closed states closed/inactivated states and open states. Moreover, the small tails between +90 to +120 mV suggest channels accumulate in an inactive state (Fig 1E). It is not convincing that the two open-state model is the mechanism underlying the mutant's behavior.  

We respectfully disagree with the notion that a single open state can provide a plausible explanation for "All the results obtained with the E600R mutant and DPASCap". We think that our new single channel results settle the question, but even without this direct evidence, a quantitative assessment of the triphasic tail currents all but excludes the possibility of a single open state. We agree that it is, in principle, possible to obtain some form of a multiphasic tail with a single open state using the scheme suggested in this comment: at the end of the test pulse, a large fraction of the channels must be accumulated in inactive states, and a few are in the open state. The hyperpolarization to -100mV then induces a rapid depopulation of the open state, followed by slower replenishments from the inactive state. Exactly this process occurs in our model, when C2 empties through O2 (Supp. 5 to Fig 9, E600R model variant). However, this alone is highly unlikely to quantitatively explain the measured tail currents, because of the drastically different time scales of the initial current decay (submillisecond to at most a few milliseconds lifetime) and the much slower transient increase in current (several tens of milliseconds) and the final decay with time constants of >100 ms (see for instance data in Fig. 1 E for E600R +50 to +120mV test pulse). To sustain the substantial magnitude of slowly decaying current by slow replenishment of an open state with a lifetime of 1 ms requires vast amounts of inactivated channels. A rough estimation based on the current integral of the initial decay and the current integral of the slowly decaying current suggests that at the end of the test pulse, the ratio inactivated/open channels would have to be 500 to 1500 for this mechanism to quantitatively explain the observed tail currents. To put this in perspective: This would suggest that without inactivation all the expressed channels in an oocyte would provide 6 mA current during the +100 mV test pulse. While theoretically possible, we consider this a less likely explanation than a second open state.

(8) Different models should be evaluated to establish whether the results in Figure 4 can also be explained by a model in which states O1 and O2 have the same conductance. It would be desirable if the conductance of both states were experimentally determined - noise analysis could be applied to estimate the conductance of both states. 

In the modified model, O1 and O2 have the same single-channel conductance. The small conductance combined with the fast flickering did not allow an accurate determination, but we can state that there is no evidence that the single-channel conductance of the states is different.

(9) Although not included, it looks like the model predicts some "conventional inactivation" This can be appreciated in Fig 8, and in the traces at -60mV. Interestingly, the traces obtained in the absence of Cl- also undergo slow inactivation, or 'conventional inactivation' as referred to by the authors. Please revise the following statement "Conventional inactivation was never detected in any mutants after repeated or prolonged depolarization. In the absence of inactivation, the pre-pulse dependent current increase at +40 mV could be related to changes in the relative occupancy of the open states". 

We have carefully edited the manuscript to address this concern. The use of the term inactivation admittedly represents a challenge. We agree that the state that results from the flickering block (C2) could be defined as “inactivated” because it is preceded by an open state. Yet, in that case, the intermediate states that the channel travels between O1 and O2 would also be sensu stricto “inactivated”, but only in the mutants. We have made this clear in page 17.

Recommendations for improving the writing and presentation.

(1) Methods section: Please state the reversal potential calculated for the solution used. It looks like the authors used an Instantaneous I-V curve method to calculate the reversal potential; if that's correct, please show the I-V and the traces together with the protocol used. 

We have provided the calculated reversal potentials for excised patches. We cannot predict the reversal potential in whole oocytes because we have no control over the intracellular solution. The reversal potential was determined in the mutants through the current at the end of the stimulus because the mutants produced measurable inward currents. The differences in reversal potential were not significant among mutants.

Pulse protocols have been added to the figures.

(2) Figure 1 suggestion: Combine the two panels in panel D and move the F panel up so the figure gets aligned in the lower end.

Thank you, this has been done.

(3) Please clarify the rationale for using the E600R-specific mutant. I assume it is based on the Lorinzci et al. 2016 effect and how this is similar to the DPASCap phenotype, or is it due to the impact of this mutation in the interactions between the N-term and the cNBHD? 

We have explained the rationale for the use of E600R explicitly on page 6.

(4) Fig S1A is not present in the current version of the manuscript. Include a cartoon as well as a structural figure clearly depicting the perturbations introduced by E600R, ΔPASCap, and the other deletions that are tested. Additional structural information supporting the discussion would also be helpful to establish clearer mechanistic links between the experimental observations described here and the observed conformational changes between states in Kv10 channel structures. 

We have corrected this omission, thank you for pointing it out.

(5) It would be informative to see the traces corresponding to the I-V shown in Fig 7 A and B at the same indicated time points (0, 60, 150, and 300s). Did the authors monitor the Ca2+ signal rise after the I&T treatment to see if it coincides with the peak in the 60s? 

In Figure 7 (now Figure 8) we used voltage ramps instead of discrete I-V protocols because of the long time required for recording the latter. This is stated on page 19. Ca2+ was monitored through Cl- current after ionomycin/thapsigargin. The duration of the Ca2+ increase was reproducible among oocytes and in good agreement with the changes observed in the biphasic behavior of the mutants (Supplement 1 to Figure 8).

(6) Fig 4. Please state in the legend what the different color traces correspond to in E600R and DPASCap. Is there a reason to change the interpulse on DPASCap to -20mV and not allow this mutant to close? Please state. How do the authors decide the 10 ms interval for the experiments in Fig 2? 

Thank you for pointing this out, we have added the description. We have explained why we use a different protocol for ΔPASCap and the reason for using 10 ms interval (we believe the referee means Figure 4) on page 12.  

(7) Fig. 5. Since the pre-pulse is supposed to be 5s, but the time scale doesn't correspond with a pre-pulse of 5 s before the test pulse to +40mV. Has the pre-pulse been trimmed for representation purposes? If so, please state. 

The pre-pulse was 5s, but as the reviewer correctly supposed, the trace is trimmed to keep the +40 mV stimulus visible. This has now been clearly stated in the legend.

(8) The mutant L322H is located within the S4 helix according to the Kv10.1 structure (PDB 5K7L), not in the 'S3-S4 linker'; please correct. 

This has been done, thank you.

The introduction of this mutant should also shift the voltage dependence toward more hyperpolarizing potentials (around 30mV, according to Schoenherr et al. 1999). It looks like that shift is present within the first component of the G-V. Still, since the max amplitude from the second component could be contaminated by endogenous Cl- currents, this effect is minimized. Repeating these experiments in the no Cl- solutions will help clarify this point and see the effect of the DPASCap and E600R in the background of a mutation that accelerates the transitions between the closed states (see Major comment 1). Did the authors record L322H alone for control purposes? 

We have decided not to measure L322H alone or repeat the measurements in Cl--free solutions because we do not see a way to use the quantitative assessment of the voltage dependence of L322H and the L322H-variants of the eag domain mutants. Like in our answer to main point 3, we base our arguments not on the precise voltage dependence of the second component but on the shape of the G-V curves instead, specifically the consistent appearance of the first component and the local conductance minimum between the first and second components. After the introduction of L322H the first component is essentially absent.

We think that the measurements of the L322H mutants cannot be interpreted as a hyperpolarizing shift in the first component. The peak of the first conductance component occurs around -20 mV in ΔPASCap and E600R (Fig. 7 C, D). After a -30mV shift, in L322H+DPASCap and L322H+E600R, this first peak would still be detected within the voltage range in our experiments, but it is not. A contamination of the second component would have little impact on this observation, which is why we refrain from the suggested measurements.  

(9) The authors differentiate between an O1 vs. O2 state with different conductances, and maybe I missed it, but there's no quantitative distinction between the components; how are they different?

Please see the response to the main comments 1 and 2. This has been addressed in singlechannel recordings.

(10) Please state the voltage protocols, holding voltages, and the solutions (K+ concentration and Cl-presence/absence) used for the experiments presented in the legends on the figures. Hence, it's easier to interpret the experiments presented. 

Thank you, this has been done.

(11) The authors state on page 7 that "with further depolarizations, the conductance initially declined to rise again in response to strong depolarizations. This finding matches the changes in amplitude of the tail currents, which, therefore, probably reflect a true change in conductance" However, the tails in the strong voltage range (+50 to +120 mV) for the E600R mutant argue against this result. Please review.

The increase in the amplitude of the tail current is also present in E600R, but the relative increase is smaller. We have decided against rescaling these traces because the Figure is already rather complex. We indicated this fact with a smaller arrow and clarified it in the text (page 8).

(12) The authors mention that the threshold of activation for the WT is around -20mV; however, the foot of the G-V is more around -30 or -40mV. Please revise. 

Thank you. We have done this. 

(13) The authors state on page 9 that the 'second component occurs at progressively more depolarized potentials for increasingly larger N-terminal deletions" However E600R mutant that conserves the N-terminal intact has a shift as pronounced as the DPASCap and larger than the D2-10. How do the authors interpret this result? 

We have corrected this statement in page 10 : “…the second component occurs at progressively more depolarized potentials for increasingly larger N-terminal deletions and when the structure of the ring is altered through disruption of the interaction between N- and C-termini (E600R)”.

(14) The equation defined to fit the G-Vs, can also be used to describe the WT currents. If the O1 is conserved and present in the WT, this equation should also fit the WT data properly. The 1-W component shown could also be interpreted as an inactivating component that, in the WT, shifts the voltage-dependence of activation towards depolarizing potentials and is not visible. Still, the mutants do show it as if the transition from closed-inactivated states is controlled by interactions in the gating ring, and disturbing them does affect the transitions to the open state. 

Out of the two open states in the mutant, O2 is the one that shares properties with the WT (e.g. it is inaccessible during Ca2+-CaM binding) while O1 is the open state with the voltage dependence that is conserved across the mutants. We, therefore, believe that this question is based on a mix-up of the two open states. We appreciate the core of the question: does the pattern in the mutants’ G-V curves find a continuation in the WT channel? 

Firstly, the component that is conserved among mutants does not lead to current in the WT because the corresponding open state (O1) is not observed in WT. However, the gating event represented by this component should also occur in WT and –given its apparent insensitivity to eag domain mutations–  this gating step should occur in WT with the same voltage dependence as in all the mutants. This means that this first component sets a hard boundary for the most hyperpolarized G-V curve we can expect in the WT, based on our mutant measurements. Secondly, the second component shows a regular progression across mutants: The more intact the eag domain is, the more hyperpolarized the Vhalf values of transition term (1-W) and O2 activation. In Δ2-10, the transition term already almost coincides with O1 activation (estimated Vhalf values of -33.57 and -33.47 mV). A further shift of (1-W) in the WT is implausible because, if O1 activation is coupled to the earliest VSD displacement, the transition should not occur before O1 activation. Still, the second component might shift to more hyperpolarized values in the WT, depending on the impact of amino acids 2 to 10 on the second VSD transition.

In summary, in WT the G-V should not be more hyperpolarized than the first component of the mutants, and the (1-W)-component probably corresponds to the Δ2-10 (1-W)-component. In WT the second component should be no more depolarized than the second component of Δ2-10. The WT G-V (Fig.1B) meets all these predictions derived from the pattern in the mutant GVs: When we use Eq. 4 to fit the WT G-V with A1=0 (O1 is not present in WT) and the parameters of the transition term (1-W)  fixed to the values attained in Δ2-10, we obtain a fit for the O2 component with Vhalf\=+21mV. This value nicely falls into the succession of Vhalf values for Δeag, ΔPASCap, and Δ2-10 (+103mV,+80mV,+52mV) and, at the same time, it is not more hyperpolarized than the conserved first component (Vhalf -34mV). Our measurements therefore support that the O2 component in the mutants corresponds to the single open state in the WT. 

(15) Page 15, the authors state that 'The changes in amplitude and kinetics in response to rising intracellular Ca2+ support our hypothesis that Ca-CaM stabilized O1, possibly by driving the channels to deep closed states (Fig 5 and 6)' (pg 15). This statement seems contradictory; I can't quite follow the rationale since Ca2+ potentiates the current (Fig 7), and the addition of the L322H mutant in Fig 7 makes the shift of the first component to negative potentials visible.

Please check the rationale for this section. 

We have explained this more explicitly in the discussion (page 32). “Because access to O1 occurs from deep closed states, this could be explained by an increased occupancy of such deactivated states in response to CaM binding. This appears to be the case since CaM induces a biphasic behavior in the mutant channels that show reduced access to deep closed states; thus, L322H mutants behave like the parental variants in the presence of Ca2+-CaM. This implies a mechanistic explanation for the effect of Ca2+-CaM on WT since favoring entry into deep closed states would result in a decrease in current amplitude in the absence of (a permeable) O1”.

Also, Figs 5 and 6 seem miscited here. 

Thank you, we have corrected this.

(16) For Figure 5, it would be helpful if each of the current traces corresponding to a particular voltage had a different color. That way, it will be easier to see how the initial holding voltage modulates current. 

We have considered this suggestion, and we agree that it would make it easier to follow. Yet, since we have identified the mutants with different colors, it would be inconsistent if we used another color palette for this Figure. Supplement 3 to Figure 9 shows the differences in a clearer way.

(17) Add zero-current levels to all current traces.

We have done this.

(18) The mathematical model should be described better. Particularly, the states from which O1 can be accessed should be described more clearly, as well as whether the model considers any direct connectivity between states O1 and O2. The origin of the voltage-dependence for transitions that do not involve voltage-sensor movements should be discussed. Also, it separation of kappa into kappa-l and kappa-r should be described. 

We have extensively rewritten the description of the mathematical model to address these concerns.

(19) Page 4, "reveals a pre-open state in which the transmembrane regions of the channel are compatible with ion permeation, but is still a nonconducting state". Also, page 27, "renders a hydrophobic constriction wider than 8 Å, enough to allow K+ flow, but still corresponds to a non-conducting state". These sentences are confusing - how can the regions be compatible with ion permeation, and still not be conducting? Is cation conductance precluded by a change in the filter, or elsewhere? How is it established that it represents a non-conducting state? 

We have rephrased to clarify this apparent inconsistence. Page 4: “(…) in which the transmembrane regions of the channel are compatible with ion permeation (the permeation path is dilated, like in open states) but the intracellular gate is still in the same conformation as in closed states (Zhang et al., 2023).” Page 31: “The presence of an intact intracellular ring would preclude ionic flow in the WT, and its alteration would explain the permeability of this state in the mutants.”

Author response:

The following is the authors’ response to the original reviews

Reviewer #1 (Public review):

Weaknesses:  

(1) The heatmaps (for example, Figure 3A, B) are challenging to read and interpret due to their size. Is there a way to alter the visualization to improve interpretability? Perhaps coloring the heatmap by general anatomical region could help? We feel that these heatmaps are critical to the utility of the registration strategy, and hence, clear visualization is necessary. 

We thank the reviewers for this point on aesthetic improvement, and we agree that clearer visualization of our correlation heatmaps is important. To address this point, we have incorporated the capability of grouping “child” subregions in anatomical order by their more general “parent” region into the package function, plot_correlation_heatmaps(). Parent regions will be can now be plotted as smaller sub-facets in the heatmaps. We have also rearranged our figures to fit enlarged heatmaps in Figures 3-5, and Supplementary Figure 10 for easier visualization. 

(2) Additional context in the Introduction on the use of immediate early genes to label ensembles of neurons that are specifically activated during the various behavioral manipulations would enable the manuscript and methodology to be better appreciated by a broad audience. 

We thank the reviewers for this suggestion and have revised the first part of our Introduction to reflect the broader use and appeal of immediate early genes (IEGs) for studying neural changes underlying behavior.

(3) The authors mention that their segmentation strategies are optimized for the particular staining pattern exhibited by each reporter and demonstrate that the manually annotated cell counts match the automated analysis. They mention that alternative strategies are compatible, but don't show this data. 

We thank the reviewers for this comment. We also appreciate that integration with alternative strategies is a major point of interest to readers, given that others may be interested in compatibility with our analysis and software package, rather than completely revising their own pre-existing pipelines. 

Generally, we have validated the ability to import datasets generated from completely different workflows for segmentation and registration. We have since released documentation on our package website with step-by-step instructions on how to do so (https://mjin1812.github.io/SMARTTR/articles/Part5.ImportingExternalDatasets). We believe this tutorial is a major entry point to taking advantage of our analysis package, without adopting our entire workflow.

This specific point on segmentation refers to the import_segmentation_custom()function in the package. As there is currently not a standard cell segmentation export format adopted by the field, this function still requires some data wrangling into an import format saved as a .txt file. However, we chose not to visually demonstrate this capability in the paper for a few reasons.  

i) A figure showing the broad testing of many different segmentation algorithms, (e.g., Cellpose, Vaa3d, Trainable Weka Segmentation) would better demonstrate the efficacy of segmentation of these alternative approaches, which have already been well-documented. However, demonstrating importation compatibility is more of a demonstration of API interface, which is better shown in website documentation and tutorial notebooks.

ii) Additionally, showing importation with one well-established segmentation approach is still a demonstration of a single use case. There would be a major burden-of-proof in establishing importation compatibility with all potential alternative platforms, their specific export formats, which may be slightly different depending on post-processing choices, and the needs of the experimenters (e.g., exporting one versus many channels, having different naming conventions, having different export formats). For example, output from Cellpose can take the form of a NumPy file (_seg.npy file), a .png, or Native ImageJ ROI archive output, and users can have chosen up to four channels. Until the field adopts a standardized file format, one flexible enough to account for all the variables of experimental interest, we currently believe it is more efficient to advise external groups on how to transform their specific data to be compatible with our generic import function.  

(4) The authors provided highly detailed information for their segmentation strategy, but the same level of detail was not provided for the registration algorithms. Additional details would help users achieve optimal alignment.

We apologize for this lack of detail. The registration strategy depends upon the WholeBrain (Fürth et al., 2018) package for registration to the Allen Mouse Common Coordinate Framework. While this strategy has been published and documented elsewhere, we have substantially revised our methods section on the registration process to better incorporate details of this approach.

(5) The authors illustrate registration to the Allen atlas. Can they comment on whether the algorithm is compatible with other atlases or with alternative sectioning planes (horizontal/sagittal)? 

Since the current registration workflow integrates WholeBrain (Fürth et al., 2018), any limitations of WholeBrain apply to our approach, which means limited support for registering non-coronal sectioning planes and reliance on the Allen Mouse Atlas (Dong, 2008). However, network analysis and plotting functions are currently compatible with the Allen Mouse Brain Atlas and the Kim Unified Mouse Brain Atlas version (2019) (Chon et al., 2019). Therefore, current limitations in registration do not preclude the usefulness of the SMARTTR software in generating valuable insights from network analysis of externally imported datasets. 

There are a number of alternative workflows, such as the QUINT workflow (Yates et al., 2019), that support multiple different mouse atlases, and registration of arbitrarily sectioned angles. We have plans to support and a facilitate an entry point for this workflow in a future iteration of SMARTTR, but believe it is of benefit to the wider community to release and support SMARTTR in its current state.

(6) Supplemental Figures S10-13 do not have a legend panel to define the bar graphs. 

We apologize for this omission and have fixed our legends in our resubmission. Our supplement figure orders have changed and the corresponding figures are now Supplemental Figures S11-14.

(7) When images in a z-stack were collapsed, was this a max intensity projection or average? Assuming this question is in regards to our manual cell counting validation approach, the zstacks were collapsed as a maximum intensity projection.  

Reviewer #2 (Public review): 

Weaknesses: 

(1) While I was able to install the SMARTR package, after trying for the better part of one hour, I could not install the "mjin1812/wholebrain" R package as instructed in OSF. I also could not find a function to load an example dataset to easily test SMARTR. So, unfortunately, I was unable to test out any of the packages for myself. Along with the currently broken "tractatus/wholebrain" package, this is a good example of why I would strongly encourage the authors to publish SMARTR on either Bioconductor or CRAN in the future. The high standards set by Bioc/CRAN will ensure that SMARTR is able to be easily installed and used across major operating systems for the long term. 

We greatly thank the reviewer for pointing out this weakness; long-term maintenance of this package is certainly a mutual goal. Loading an .RDATA file is accomplished by either doubleclicking directly on the file in a directory window, after specifying this file type should be opened in RStudio or by using the load() function, (e.g., load("directory/example.RData")). We have now explicitly outlined these directions in the online documentation. 

Moreover, we have recently submitted our package to CRAN and are currently working on revisions following comments. This has required a package rebranding to “SMARTTR”, as there were naming conflicts with a previously archived repository on CRAN. Currently, SMARTTR is not dependent on the WholeBrain package, which remains optional for the registration portion of our workflow. Ultimately, this independence will allow us to maintain the analysis and visualization portion of the package independently.

In the meantime, we have fully revised our installation instructions (https://mjin1812.github.io/SMARTTR/articles/SMARTTR). SMARTTR is now downloadable from a CRAN-like repository as a bundled .tar.gz file, which should ease the burden of installation significantly. Installation has been verified on a number of different versions of R on different platforms. Again, we hope these changes are sufficient and improve the process of installation. 

(2) The package is quite large (several thousand lines include comments and space). While impressive, this does inherently make the package more difficult to maintain - and the authors currently have not included any unit tests. The authors should add unit tests to cover a large percentage of the package to ensure code stability. 

We have added unit testing to improve the reliability of our package. Unit tests now cover over 71% of our source code base and are available for evaluation on our github website (https://github.com/mjin1812/SMARTTR). We focused on coverage of the most front-facing functions. We appreciate this feedback, which has ultimately enhanced the longevity of our software.

(3) Why do the authors choose to perform image segmentation outside of the SMARTTR package using ImageJ macros? Leading segmentation algorithms such as CellPose and StarMap have well-documented APIs that would be easy to wrap in R. They would likely be faster as well. As noted in the discussion, making SMARTTR a one-stop shop for multi-ensemble analyses would be more appealing to a user. 

We appreciate this feedback. We believe parts of our response to Reviewer 1, Comment 3, are relevant to this point. Interfaces for CellPose and ClusterMap (which processes in situ transcriptomic approaches, like STARmap) are both in python, and currently there are ways to call python from within R (https://rstudio.github.io/reticulate/index.html). We will certainly explore incorporating these APIs from R. However, we would anticipate this capability is more similar to “translation” between programming languages, but would not currently preclude users from the issue of needing some familiarity with the capabilities of these python packages, and thus with python syntax.

(4) Given the small number of observations for correlation analyses (n=6 per group), Pearson correlations would be highly susceptible to outliers. The authors chose to deal with potential outliers by dropping any subject per region that was> 2 SDs from the group mean. Another way to get at this would be using Spearman correlation. How do these analyses change if you use Spearman correlation instead of Pearson? It would be a valuable addition for the author to include Spearman correlations as an option in SMARTTR. 

We thank reviewers for this suggestion and we have updated our code base to include the possibility for using Spearman’s correlation coefficient as opposed to Pearson’s correlation coefficient for heatmaps in the get_correlations() function. Users can now use the `method` parameter, set to either “pearson” or “spearman” and results will propagate throughout the rest of the analysis using these results.

Below, in Author response image 1 we show a visual comparison of the correlation heat maps for active eYFP⁺ ensembles in the CT and IS groups using both Pearson and Spearman correlations. We see a strongly qualitative similarity between the heat maps. Of course, since the statistical assumptions underlying the relationship between variables using Pearson correlation (linear) vs Spearman correlation (monotonic) are different, users should take this into account when interpreting results using different approaches.

Author response image 1.

Pearson and Spearmen regional correlations of eYFP+ ensembles activity in the CT and IS groups.

(5) I see the authors have incorporated the ability to adjust p-values in many of the analysis functions (and recommend the BH procedure) but did not use adjusted p-values for any of the analyses in the manuscript. Why is this? This is particularly relevant for the differential correlation analyses between groups (Figures 3P and 4P). Based on the un-adjusted pvalues, I assume few if any data points will still be significant after adjusting. While it's logical to highlight the regional correlations that strongly change between groups, the authors should caution which correlations are "significant" without adjusting for multiple comparisons. As this package now makes this analysis easily usable for all researchers, the authors should also provide better explanations for when and why to use adjusted p-values in the online documentation for new users. 

We appreciate the feedback note that our dataset is presented as a more demonstrative and exploratory resource for readers and, as such, we accept a high tolerance for false positives, while decreasing risk of missing possible interesting findings. As noted by Reviewer #2, it is still “logical to highlight the regional correlations that strongly change between groups.” We have clarified in our methods that we chose to present uncorrected p-values when speaking of significance. 

We have also removed any previous recommendations for preferred methods for multiple comparisons adjustment in our function documentations, as some previous documentation was outdated. Moreover, the standard multiple comparisons adjustment approaches assume complete independence between tests, whereas this assumption is violated in our differential correlational analysis (i.e., a region with one significantly altered connection is more likely than another to have another significantly altered connection).

Ultimately, the decision to correct for multiple comparisons with standard FDR, and choice of significance threshold, should still be informed by standard statistical theory and user-defined tolerance for inclusion of false-positives and missing of false-negatives. This will be influenced by factors, such as the nature and purpose of the study, and quality of the dataset.  

(6) The package was developed in R3.6.3. This is several years and one major version behind the current R version (4.4.3). Have the authors tested if this package runs on modern R versions? If not, this could be a significant hurdle for potential users. 

We thank reviewers for pointing out concerns regarding versioning. We have since updated our installation approach for SMARTTR, which is compatible with versions of R >= 3.6 and has been tested on Mac ARM-based (Apple silicon) architecture (R v4.4.2), and Windows 10 (R v3.6.3, v4.5.0 [devel]). 

The recommendation for users to install R 3.6.3 is primarily for those interested in using our full workflow, which requires installation of the WholeBrain package, which is currently a suggested package. We anticipate updating and supporting the visualization and network analysis capabilities, whilst maintaining previous versioning for the full workflow presented in this paper.  

(7) In the methods section: "Networks were constructed using igraph and tidygraph packages." - As this is a core functionality of the package, it would be informative to specify the exact package versions, functions, and parameters for network construction. 

We thank reviewers for pointing out the necessity for these details for code reproducibility. We have since clarified our language in the manuscript on the exact functions we use in our analysis and package versions, which we also fully document in our online tutorial. Additionally. We have printed our package development and analysis environment online at https://mjin1812.github.io/SMARTTR/articles/Part7.Development.

(8) On page 11, "Next, we examined the cross-correlations in IEG expression across brain regions, as strong co-activation or opposing activation can signify functional connectivity between two regions" - cross-correlation is a specific analysis in signal processing. To avoid confusion, the authors should simply change this to "correlations". 

We thank the reviewer for pointing out this potentially confusing phrasing. We have changed all instances of “cross-correlation” to “correlation”.

(9) Panels Q-V are missing in Figure 5 caption. 

We thank the reviewer for pointing out this oversight. We have now fixed this in our revision.

References

Chon, U., Vanselow, D. J., Cheng, K. C., & Kim, Y. (2019). Enhanced and unified anatomical labeling for a common mouse brain atlas. Nature Communications, 10(1), 5067. https://doi.org/10.1038/s41467-019-13057-w

Dong, H. W. (2008). The Allen reference atlas: A digital color brain atlas of the C57Bl/6J male mouse (pp. ix, 366). John Wiley & Sons Inc.

Fürth, D., Vaissière, T., Tzortzi, O., Xuan, Y., Märtin, A., Lazaridis, I., Spigolon, G., Fisone, G., Tomer, R., Deisseroth, K., Carlén, M., Miller, C. A., Rumbaugh, G., & Meletis, K. (2018). An interactive framework for whole-brain maps at cellular resolution. Nature Neuroscience, 21(1), 139–149. https://doi.org/10.1038/s41593-017-0027-7

Yates, S. C., Groeneboom, N. E., Coello, C., Lichtenthaler, S. F., Kuhn, P.-H., Demuth, H.-U., Hartlage-Rübsamen, M., Roßner, S., Leergaard, T., Kreshuk, A., Puchades, M. A., & Bjaalie, J. G. (2019). QUINT: Workflow for Quantification and Spatial Analysis of Features in Histological Images From Rodent Brain. Frontiers in Neuroinformatics, 13. https://www.frontiersin.org/articles/10.3389/fninf.2019.00075

Author response:

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

Reviewer 1:

This research used cell-based signaling assay and Gaussian-accelerated molecular dynamics (GaMD) to study peptide-mediated signaling activation of Polycystin-1 (PC1), which is responsible for the majority of autosomal dominant polycystic kidney disease (ADPKD) cases. Synthetic peptides of various lengths derived from the N-terminal portion of the PC1 C-terminal fragment (CTF) were applied to HEK293T cells transfected with stalkless mouse CTF expression construct. It was shown that peptides including the first 7, 9, and 17 residues of the N-terminal portion could activate signaling to the NFAT reporter. To further understand the underlying mechanism, docking and peptide-GaMD simulations of peptides composed of the first 9, 17, and 21 residues from the N-terminal portion of the human PC1 CTF were performed. These simulations revealed the correlation between peptide-CTF binding and PC1 CTF activation characterized by the close contact (salt bridge interaction) between residues R3848 and E4078. Finally, a Potts statistical model was inferred from diverged PC1 homologs to identify strong/conserved interacting pairs within PC1 CTF, some of which are highly relevant to the findings from the peptide GaMD simulations. The peptide binding pockets identified in the GaMD simulations may serve as novel targets for the design of therapeutic approaches for treating ADPKD.

We greatly appreciate the reviewer’s encouraging and positive comments. The reviewer’ specific comments are addressed pointwise below and changes to the text will be highlighted in yellow in the revised manuscript.

(1) The GaMD simulations all include exogenous peptides, thus lacking a control where no such peptide is present (and only stalkless CTF). An earlier study (PNAS 2022 Vol. 119 No. 19 e2113786119) covered this already, but it should be mentioned here that there was no observation of close/activation for the stalkless CTF.

We appreciate the reviewer’s concern about the lack of a control where no exogenous peptide is present. As suggested by the reviewer, we are adding more details about the study on the stalkless CTF as a control in the Introduction of the revised manuscript. 

(2) Although 5 independent trajectories were generated for each peptide, the authors did not provide sufficient details regarding the convergence of the simulation. This leaves some uncertainties in their results. Given that the binding poses changed relative to the starting docked poses for all three peptides, it is possible that some other binding pockets and/or poses were not explored.

We appreciate the reviewer’s comment regarding the convergence of the simulation results. This is clarified in the revised manuscript as: 

“We have calculated free energy profiles of individual simulations for each system, including the p9, p17, and p21, as shown below (Figs. S5, S6 and S8). For the p9 peptide, the “Bound” lowenergy state was consistently identified in the 2D free energy profile of each individual simulation (Fig. S5). For the p17 peptide, Pep-GaMD simulations were able to refine the peptide conformation from the "Unbound” to the "Intermediate” and “Bound” states in Sim1 and Sim5, while the peptide reached only the "Intermediate” state in the other three simulations (Fig. S6). For the p21 peptide, Pep-GaMD was able to refine the peptide docking conformation to the

"Bound” state in all the five individual simulations (Fig. S8).”

“It is important to note that the free energy profiles calculated from GaMD simulations of PC1 CTF were not fully converged since certain variations were observed among the individual simulations. Nevertheless, these calculations allowed us to identify representative low-energy binding conformations of the peptides.”

(3) The free energy profiles (Figures 2 to 4) based on the selected coordinates provide important information regarding binding and CTF conformational change. However, it is a coarsegrained representation and complementary analysis such as RDFs, and/or contact maps between the peptide and CTF residues might be helpful to understand the details of their interactions. These details are currently only available in the text. 

Following the reviewer's suggestion, we have now included a set of protein contact maps showing contacts between the peptides and the TOP domain for each peptide in the representative "Bound” state in revised Supplementary Information (Fig. S4). The contact maps serve to visualize the list of contacts mentioned in the main text. This will be clarified in the revised manuscript.

(4) The use of a stalkless CTF is necessary for studying the functions of the exogenous peptides. However, the biological relevance of the stalkless CTF to ADPKD was not clearly explained, if any.

We appreciate the reviewer’s comment. As correctly assessed by the reviewer, the stalkless CTF is not a biological form of PC1 observed in ADPKD, but rather was used as the simplest or least complex system in which the activities and binding of exogenous peptides could be studied. However, in ADPKD, there are numerous missense mutations reported within the GPCR autoproteolysis-inducing (GAIN) domain that have been shown to prevent or inhibit cleavage at the GPCR-coupled proteolysis site (GPS). Loss of PC1 GPS cleavage, which is known to cause ADPKD, would retain or sequester the stalk tethered agonist within the interior of the GAIN domain, which would presumably interfere with interactions between stalk tethered agonist residues and the remainder of the CTF. Furthermore, there are 10 single nucleotide polymorphisms reported within the stalk sequence (ADPKD Variant Database; https://pkdb.mayo.edu/welcome), most of which we have found to significantly reduce CTF-mediated activation of the NFAT reporter (Magenheimer BS, et al., Constitutive signaling by the C-terminal fragment of polycystin1 is mediated by a tethered peptide agonist; bioRxiv 2021.08.05.455255). In particular, the ADPKD-associated G3052R stalk mutation that was analyzed along with the stalkless CTF by GaMD simulations (Pawnikar et al, PNAS, 2022) has the same reduction in activity as the stalkless CTF in the cellular signaling reporter assays and the same loss of closed conformation interactions in GaMD analyses. As such, we believe the stalkless CTF has biological relevance from the aspect that it mimics the deficiency in signaling activation observed for PC1 CTF stalk mutants. This is clarified in the revised manuscript in the Introduction, page 5, “constructs encoding a stalkless PC1 CTF (a nonbiological mutant of PC1 with deletion of the first 21 N-terminal residues of CTF) and three ADPKD-associated…”) and near the beginning of the Discussion, page 16, where the biological relevance of studying the stalkless CTF is explained

(5) The authors might want to clarify if a stalkless CTF is commonly seen in ADPKD, or if it is just a construct used for this study.

The stalkless CTF is not a biological form of PC1, but rather a construct used for this study. This was clarified in the revised manuscript (see response above).

(6) (Pages 7-8) "...we generated expression constructs of mouse (m) PC1 consisting of the CD5 signal peptide sequence fused in frame with the stalk sequence of mCTF ...". What is the CD5 signal peptide sequence here? What is its use?

The CD5 signal peptide sequence is “MPMGSLQPLATLYLLGMLVASVLG” from the T cell surface glycoprotein, CD5. Since the N-terminus of PC1 CTF is derived from a posttranslational, autocatalytic, endoproteolytic cleavage event, this isoform is already membraneembedded and therefore lacks its endogenous signal peptide. The CD5 signal peptide coding sequence is added to the PC1 CTF expression constructs in order to ensure translation and insertion of the encoded protein at the endoplasmic reticulum. Additional details were added to the Experimental Procedures, page 2 of Supporting Information.

(7) (Page 8) "All peptides were appended with a C-terminal, 7-residue hydrophilic sequence (GGKKKKK) to increase solubility". How did the authors make sure that this sequence has no influence on the signaling? 

To determine the possible effect of the hydrophilic GGKKKKK sequence on signaling, we had a ‘solubility tag’ peptide (LGGKKKKK) synthesized and purified by GenScript. It was necessary to add an N-terminal Leu residue to the 7-residue hydrophilic tag sequence in order for the highly hydrophilic peptide to be recovered. Effect of treatment with the solubility tag peptide on activation of the NFAT reporter was assessed for both empty vector- and ∆stalkCTF-transfected cells in 3 separate signaling experiments (see figure below). Each experiment also included a negative control treatment (no peptide/culture medium only addition) and a positive control treatment (stalk peptide p17). The p17 peptide we had available was derived from the stalk sequence of human PC1 that differs from the mouse PC1 sequence at residues 15 and 17, which are two poorly conserved positions within the stalk sequence (see Reviewer 2, Response 3). In the first experiment with the solubility tag and human p17 peptides (B in figure below), we inadvertently used the empty expression vector and ∆stalkCTF expression construct from mouse PC1. After realizing our error, we then performed 2 additional signaling experiments (C and D in figure below) with the ‘correct’ human ∆stalkCTF expression construct and empty vector. In the revised manuscript, we have provided the results from each of the 3 experiments as Fig. S2 (below).

(8) (Page 9) "Using a computational model of the ΔStalk PC1 CTF developed previously". The authors might want to expand here a little to give a short review about the structure preparation.

We appreciate the reviewer’s suggestion regarding the addition of details for structure preparation for Stalkless CTF. We have added these details in section “Docking and Pep-GaMD simulations of peptide agonist binding to stalkless PC1 CTF” on Page 10 in the revised manuscript:  “The cryo-EM structure of human PC1-PC2 complex (PDB: 6A70) was used to build the computational model for WT PC1 CTF. As the protein had several missing regions including the Stalk and several loops, homology modeling of the missing regions was done using I-TASSER web server. Using the WT PC1 CTF model, computational model for ΔStalk was generated by deleting the first 21 residues (3049-3069) of the WT PC1 and using the structure for stalkless CTF, we successfully docked the p9, p17 and p21 stalk peptides with HPEPDOCK.  The peptides all bound to the TOP domain and the interface between the TOP domain and extracellular loop 1 (ECL1) of CTF.”

(9) How was "contact" defined when counting the number of contacts used in the 2D PMFs (Figures 2-4). Response: We appreciate the reviewer’s comment regarding the definition of the number of contacts used in the 2D PMFs. This has been clarified in the revised manuscript as: “The number of contacts is calculated between any atom pairs within 4 Å distance of the peptide and extracellular domains of PC1 protein.”

(10) How was the ranking of GaMD clusters done? It looks from Figure 3A that the "intermediate" state is more favorable compared to the "bound" state, but it was claimed in the text the "bound" state was ranked 1st. 

Thanks to the reviewer for this comment. It has been clarified in the revised

Supplementary Information: “Three independent Pep-GaMD simulations were combined to perform structural clustering using the hierarchical agglomerative clustering algorithm in CPPTRAJ. A 3 Å RMSD cutoff was used for each peptide system. PyReweighting was then applied to calculate the original free energy values of each peptide structural cluster with a cutoff of 500 frames. The structural clusters were finally ranked according to the reweighted free energy values.” And in the revised main text: “It is important to note that the free energy profiles calculated from GaMD simulations of PC1 CTF were not fully converged since certain variations were observed among the individual simulations. The free energy values of 2D PMF minima shown in Figure 3A could differ from those in the 1D PMF minima of peptide structural clusters, especially with the usage of distinct reaction coordinates. Nevertheless, these calculations allowed us to identify representative low-energy binding conformations of the peptides.”

(11) When mentioning residue pair distances, such as in the sentence "The distance between the TOP domain residue R3848 and PL residue E4078 was 3.8 Å (Fig. 4D)" on page 12, it should be clarified if these distances are average distance, or a statistical error can be given.

We appreciate the reviewer’s comment regarding the TOP Domain and PL distance between residues R3848-E4078. This has been clarified on page 14 in the revised manuscript as:

“The distance between the TOP domain residue R3848 and PL residue E4078 was 3.8 Å. The distance was extracted from the top-ranked structural cluster of the p21 bound to the ΔStalk CTF, corresponding to the “Closed/Active” low-energy conformational state. (Fig. 4E)”.

(12) More analysis of the GaMD can be performed. For example, the authors observed a single "bound" state for p21, but there must be some flexibility in the peptide and the protein itself. The authors might want to consider adding some plots illustrating the flexibility of the peptide residues (for example, a RMSD plot). Contact maps can also be added to visualize the results currently discussed in the text. 

We thank the reviewer for their constructive suggestions. To characterize flexibility of the peptide and protein in the revised manuscript, we have added plots of the TOP-PL interaction distance between residues R3848-E4078 in PC1, the radius of gyration (Rg) of p21 and root-mean square deviation (RMSD) of p21 relative to the starting HPEPDOCK conformation of the peptide in the new Fig. S7. The peptide-protein contact map has also been added in the new Fig. S4.

(13) (Page 7) In the sentence `...sampled the "Closed/Active" low-energy state relative to the large number of Stalk-TOP contacts`, I suggest using "related to" instead of "relative to"

We thank the reviewer for the comment, and we have replaced "relative to" to “related to” in the following sentence `...sampled the "Closed/Active" low-energy state relative to the large number of Stalk-TOP contacts`

(14) (Page 7) In the sentence `Our previous study utilized expression constructs of human PC1 CTF, however, in order to prepare for ...`, "PC1 CTF, however," -> "PC1 CTF. However,"

We thank the reviewer for the comment, and we have replaced "PC1 CTF, however," to "PC1 CTF. However," in the following sentence `Our previous study utilized expression constructs of human PC1 CTF, however, in order to prepare for ...`.

Reviewer 2:

The autosomal dominant polycystic kidney disease (ADPKD) is a major form of polycystic kidney disease (PKD). To provide better treatment and avoid side effects associated with currently available options, the authors investigated an interesting GPCR, polycystin-1 (PC1), as a potential therapeutic target. In vitro and in silico studies were combined to identify peptide agonists for PC1 and to elucidate their roles in PC1 signaling. Overall, regarding the significance of the findings, this work described valuable peptide agonists for PC1 and the combined in vitro and in silico approach can be useful to study a complex system like PC1. However, the strength of the evidence is incomplete, as more experiments are needed as controls to validate the computational observations. The work appears premature.

We greatly appreciate the reviewer’s encouraging and positive comments. The reviewer’ specific comments are addressed pointwise below and changes to the text will be highlighted in yellow in the revised manuscript.

(1) The therapeutic potential of PC1 peptide agonists is unclear in the introduction. For example, while the FDA-approved drug Jynarque was mentioned, the text was misleading as it sounded like Jynarque targeted PC1. In fact, it targets another GPCR, the vasopressin receptor 2 (V2). A clear comparison of targeting PC1 over V2 pathways and their therapeutic relevance can help the readers better understand the importance of this work. Importantly, a clear background on the relationship between PC1 agonism and treatments for ADPKD is necessary.

We understand the confusion that was caused by the brevity of our introductory paragraph and will clarify the differences in therapeutic targeting between Jynarque and our PC1 stalk-derived peptides in the revised manuscript. We will also expound on the rationale for targeting PC1 agonism as a therapeutic approach for ADPKD versus Jynarque. For example: It is known that ADPKD disease severity is dependent on the functional levels of PC1. Jynarque is a small molecule antagonist of the arginine vasopressin receptor 2, V2R, whose signaling, and production of cAMP has been shown to be increased in ADPKD. As this drug targets one of the downstream aberrant pathways, it is only capable of slowing disease progression and has numerous undesirable side effects. We reasoned that a therapeutic agent capable of stimulating and thus augmenting PC1 signaling function would be a safer, cyst initiation-proximal treatment capable of preventing cyst formation with few side effects.

(2) PC1 is a complex membrane protein, and most figures focus on the peptide-binding site. For general readers (or readers that did not read the previous PNAS publication), it is hard to imagine the overall structure and understand where the key interactions (e.g., R3848-E4078) are in the protein and how peptide binding affects locally and globally. I suggest enhancing the illustrations.

We thank the reviewer for the constructive comment on adding more illustrations for the PC1 protein to understand the overall structure and the location of the key interaction R3848E4078. We have included these suggestions and modified the main figures in the revised manuscript.  

(3) The authors used the mouse construct for the cellular assays and the peptide designs in preparation for future in vivo assays. This is helpful in understanding biology, but the relevance of drug discovery is weakened. Related to Point 1, the therapeutic potential of PC1 peptide agonist is largely missing.

The therapeutic potential of a PC1 peptide agonist is addressed in response #1 above. As mentioned in the manuscript and recognized by the reviewer, the cellular signaling assays were performed with the mouse PC1 CTF expression construct and with peptides based on the mouse PC1 stalk sequence for future, pre-clinical studies, while the peptide binding studies were performed with the human PC1 stalk sequence. We feel the relevance for drug discovery is not significantly weakened for a number of reasons: 1) as shown in Fig. 1A, the stalk sequence is highly conserved between mouse and human PC1, specifically there are only 2 residue differences present within peptides p17 and p21. One of the differences is a ‘semi-conservative’ Gln-Arg substitution at peptide residue 15, while the second difference is a conservative Ile-Val substitution at peptide residue 17; 2) we have found that an Arg to Cys mutation within the mouse PC1 CTF stalk has the same effect on signaling as the corresponding human Gln to Cys ADPKD-associated mutation which was analyzed in Pawnikar et al., 2022; and 3) both peptide residues 15 and 17 represent highly variable positions within the PC1 stalk as shown in the sequence logo (below) of the stalk sequence from 16 vertebrate species; and 4) while addressing the potential effect of the hydrophilic solubility tag on stalk peptide-mediated rescue of CTF∆stalk signaling (see Reviewer 1 comments, point #7), we utilized the ‘human’ version of p17 as a positive control and tested its activation with both mouse and human CTF∆stalk expression constructs and found that human p17 peptide was also capable of stimulating the mouse CTF∆stalk protein (Fig. S2).

Author response image 1.

(4) More control experiments are needed. For example, a 7-residue hydrophilic sequence (GGKKKKK) is attached to the peptide design to increase solubility. This 7-residue peptide should be tested for PC1 activation as a control. Second, there is no justification for why the peptide design must begin with residue T3041. Can other segments of the stalk also be agonists?

As mentioned above for Reviewer 1, the hydrophilic peptide has been synthesized and tested for activation of signaling by the stalkless CTF in the revised manuscript as Fig. S2. The design of peptides that begin with residue T3041 of mouse PC1 CTF is modeled on numerous similar studies for the family of adhesion GPCRs. Optimization of the binding and activity of the PC1 peptide agonist will be investigated in future studies and could include such parameters as whether the peptide must include the first residue and whether subsegments of the stalk are also agonists, however, we feel these questions are beyond the scope of this initial report.

(5) There are some major concerns about the simulations: The GaMD simulations showed different binding sites of p-21, p-17, and p-9, and the results report the simulated conformations as "active conformational states". However, these are only computational findings without structural biology or mutagenesis data to validate. Further, neither docking nor the simulation data can explain the peptide SAR. Finally, it will be interesting if the authors can use docking or GaMD and explain why some peptide designs (like P11-P15) are less active (as control simulations).

The reviewer brings up an important observation regarding differences in binding sites between peptides p9, p17 and p21. We will include discussion of this observation and our interpretations to the revised manuscript. While the present study is focused on identification of initial peptides that are able to activate the PC1 CTF, we shall include further mutation experiments and simulations, peptide SAR and optimization of the lead peptides in future studies. This has been clarified in the revised manuscript.

(6) Additional experiments for the controls and for validating the simulations. Additional simulations to explain the SAR.

We appreciate the reviewer’s comment for additional experiments for the controls and additional simulations to explain the SAR. For future studies, we shall include further mutation experiments and simulations, peptide SAR and optimization of the lead peptides.

(7) What is the selectivity of the peptides between PC1 and PC2?

We have not tested the selectivity of the peptides for PC1 versus PC2 primarily because transfection of PC2 does not activate the NFAT reporter. However, it is possible that co-transfection of PC2 with the PC1 CTF could alter stalk peptide binding. This will be important to consider in future studies.

Reviewer 3:

The authors demonstrate the activation of Polycystin-1 (PC1), a G-protein coupled receptor, using small peptides derived from its original agonist, the stalk TA protein. In the experimental part of the study, the authors performed cellular assays to check the peptide-induced reactivation of a mutant form of PC1 which does not contain the stalk agonist. The experimental data is supported by computational studies using state-of-the-art Gaussian accelerated Molecular Dynamics (GaMD) and bioinformatics analysis based on sequence covariance. The computer simulations revealed the mechanistic details of the binding of the said peptides with the mutant PC1 protein and discovered different bound, unbound, and intermediate conformations depending on the peptide size and sequence. The use of reliable and well-established molecular simulation algorithms and the physiological relevance of this protein autosomal dominant polycystic kidney disease (ADPKD) make this work particularly valuable.

We greatly appreciate the reviewer’s encouraging and positive comments. The reviewer’ specific comments are addressed pointwise below and changes to the text will be highlighted in yellow in the revised manuscript.

(1) No control has been used for the computational (GaMD) study as the authors only report the free energy surface for 3 highly agonistic peptides but for none of the other peptides that did not induce an agonistic effect. Therefore, in the current version, the reliability of the computational results is not foolproof.

We appreciate the reviewer’s concern about the lack of control with the other peptides that did not induce an agonistic effect. To address the reviewer’s concern, we have included more details on the study of the stalkless CTF and the solubility tag peptide (Fig. S2) as controls in the revised manuscript.

(2) All discussions about the residue level interactions focused only on geometric aspects (distance, angle, etc) but not the thermodynamic aspect (e.g. residue-wise interaction energy). Considering they perform a biased simulation; the lack of interaction energy analysis only provides a qualitative picture of the mechanism.

As mentioned by the reviewer, we have added MM/PBSA analysis results in the revised manuscript and SI.

Molecular Mechanics/Poisson-Boltzmann Surface Area (MM/PBSA) analysis was performed to calculate the binding free energies of peptides p9, p17 and p21 to PC1 CTF. The analysis was performed using the trajectory in which the peptide was bound to the receptor. In MM/PBSA, the binding free energy of the ligand (L) to the receptor (R) to form the complex (RL) is calculated as:

where GRL is the Gibbs free energy of the complex RL, GR is the Gibbs free energy of the molecule R in its unbound state and GL is the Gibbs free energy of the molecule L in its unbound state, respectively. 

𝛥𝐺𝑏𝑖𝑛𝑑 can be divided into contributions of different interactions as:

in which

where ΔEMM , ΔGsol , 𝞓H and −TΔS are the changes in the gas-phase molecular mechanics (MM) energy, solvation free energy, enthalpy and conformational entropy upon ligand binding, respectively. ΔEMM includes the changes in the internal energies ΔEint (bond, angle and dihedral energies), electrostatic energies ΔEelec , and the van der Waals energies ΔEvdW. ΔGsol is the sum of the electrostatic solvation energy ΔGPB/GB (polar contribution) and the nonpolar contribution ΔGSA between the solute and the continuum solvent. The polar contribution is calculated using either the Poisson Boltzmann (PB) or Generalized Born (GB) model, while the nonpolar energy is usually estimated using the solvent-accessible surface area (SASA) where 𝞬 is surface tension coefficient and b is the constant offset. The change in conformational entropy −TΔS is usually calculated by normal-mode analysis on a set of conformational snapshots taken from MD simulations. However, due to the large computational cost, changes in the conformational entropy are usually neglected as we were concerned more on relative binding free energies of the similar peptide ligands.

MM/PBSA analysis was performed using the gmx_MMPBSA software with the following command line:

gmx_MMPBSA -O -i mmpbsa.in -cs com.tpr -ci index.ndx -cg 1 13 -ct com_traj.xtc -cp topol.top -o FINAL_RESULTS_MMPBSA.dat -eo FINAL_RESULTS_MMPBSA.csv Input file for running MM/PBSA analysis:

&general

sys_name="Prot-Pep-CHARMM",

startframe=1, endframe=200, # In gmx_MMPBSA v1.5.0 we have added a new PB radii set named charmm_radii. 

# This radii set should be used only with systems prepared with CHARMM force fields. 

# Uncomment the line below to use charmm_radii set

# PBRadii=7,

/

&pb

# radiopt=0 is recommended which means using radii from the prmtop file for both the PB calculation and for the NP

# calculation

istrng=0.15, fillratio=4.0, radiopt=0

The relative rank of the overall peptide binding free energies (Table S1) was consistent with the experimental signaling data, i.e., p21>p9>p17, for which p21 showed the largest binding free energy value of binding (-40.29±6.94 kcal/mol).

(3) It is not mentioned clearly whether the reader should interpret the free energy landscapes quantitatively or qualitatively. Considering no error analysis or convergence plots are reported for the GaMD free energy surfaces, it may be assumed the results are qualitative. The readers should consider this caveat and not try to quantitatively reproduce these free energy landscapes with other comparable techniques.

We appreciate the reviewer’s comment whether the free energy landscapes should be interpreted quantitatively or qualitatively. The presented free energy landscapes could be considered semi-quantitative since the simulations are not fully converged. This will be clarified in the revised manuscript as: “It is important to note that the free energy profiles calculated from GaMD simulations of PC1 CTF were not fully converged since certain variations were observed among the individual simulations. Nevertheless, these calculations allowed us to identify representative low-energy binding conformations of the peptides.”

(4) Energy decomposition analysis similar to the following paper (https://pubs.acs.org/doi/10.1021/bi201856m) should be provided to understand the residue level enthalpic contribution in the peptide-protein interaction.

As mentioned by the reviewer, we have performed residue-wise interaction energy analysis and included the analysis results in the revised manuscript and SI.

Residue-wise interaction energy analysis was performed on peptides p9, p17 and p21 using the trajectory in which the peptide was bound to the PC1 CTF using the gmx_MMPBSA software with the following command line:

gmx_MMPBSA -O -i mmpbsa.in -cs com.tpr -ct com_traj.xtc -ci index.ndx -cg 3 4 -cp topol.top -o FINAL_RESULTS_MMPBSA.dat -eo FINAL_RESULTS_MMPBSA.csv -do FINAL_DECOMP_MMPBSA.dat -deo FINAL_DECOMP_MMPBSA.csv

Input file for running residue-wise energy decomposition analysis:

&general

sys_name="Decomposition", startframe=1, endframe=200,

# forcefields="leaprc.protein.ff14SB"

/

&gb

igb=5, saltcon=0.150,

/

# make sure to include at least one residue from both the receptor #and peptide in the print_res mask of the &decomp section.

# this requirement is automatically fulfilled when using the within keyword.

# http://archive.ambermd.org/201308/0075.html

&decomp

idecomp=2, dec_verbose=3, print_res="A/854-862 A/1-853”,

/

Residue-wise energy decomposition analysis allowed us to identify key residues that contributed the most to the peptide binding energies. These included residues T1 and V9 in p9 (Table S2), residues T1, R15 and V17 in p17 (Table S3), and residues P10, P11, P19 and P21 in p21 and residue W3726 in the PC1 CTF (Table S4). The energetic contributions of these residues apparently correlated to the sequence coevolution predicted from the Potts model.

(5) To showcase the reliability of the computational approach, the authors should perform the MD simulation studies with one peptide that did not show any significant agonistic effect in the experiment. This will work as a control for the computational protocol and will demonstrate the utility of the pep-GaMD simulation in this work.

We appreciate the reviewer’s concern about the lack of control with the other peptides that did not induce an agonistic effect. It is difficult for us to add more MD simulations on the other peptides, due to student leave after PhD graduation. But to address the reviewer’s concern, we have included more details on the study of the stalkless CTF as a control in the revised manuscript.

(6) To assess the accuracy of the computational results the authors should mention (either in the main text or SI) whether the reported free energy surfaces were the average of the five simulations or computed from one simulation. In the latter case, free energy surfaces computed from the other four simulations should be provided in the SI. In addition, how many binding unbinding events have been observed in each simulation should be mentioned.

We appreciate the reviewer’s comment regarding convergence of the simulation free energy surfaces. In response to Reviewer 1, we have calculated free energy profiles of individual simulations for each system, including the p9, p17, and p21 (Figs. S5, S6 and S8). 

“We have calculated free energy profiles of individual simulations for each system, including the p9, p17, and p21 (Figs. S5, S6 and S8). For the p9 peptide, the “Bound” low-energy state was consistently identified in the 2D free energy profile of each individual simulation (Fig. S5). For the p17 peptide, Pep-GaMD simulations were able to refine the peptide conformation from the "Unbound” to the "Intermediate” and “Bound” states in Sim1 and Sim5, while the peptide reached only the "Intermediate” state in the other three simulations (Fig. S6). For the p21 peptide, PepGaMD was able to refine the peptide docking conformation to the "Bound” state in all the five individual simulations (Fig. S8).”

“It is important to note that the free energy profiles calculated from GaMD simulations of PC1 CTF were not fully converged since certain variations were observed among the individual simulations. Nevertheless, these calculations allowed us to identify representative low-energy binding conformations of the peptides.”

Author response:

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

eLife Assessment

This is a useful report of a spatially-extended model to study the complex interactions between immune cells, fibroblasts, and cancer cells, providing insights into how fibroblast activation can influence tumor progression. The model opens up new possibilities for studying fibroblast-driven effects in diverse settings, which is crucial for understanding potential tumor microenvironment manipulations that could enhance immunotherapy efficacy. While the results presented are solid and follow logically from the model’s assumptions, some of these assumptions may require further validation, as they appear to oversimplify certain aspects in light of complex experimental findings, system geometry, and general principles of active matter research.

We thank the editor for recognizing the usefulness of our work. This work does not aim to precisely describe the complexity of the tumor microenvironment in lung cancer, but rather to classify and rigorously calibrate a minimum number of parameters to the clinical data we collect and generate, and reproduce the global structures of the microenvironment. We identify different scenarios, and show how they depend on the local interactions within this framework. Although we started in the first version with coalescence in the main text and anisotropic geometry in the supporting information, we realized that we needed to provide more directions to better show how our model can be extended. Thus, in Section III-4 we added an analysis of a microenvironment with blood vessels, and showed how to introduce anisotropic friction as a function of fiber orientation, as well as active stress, paving the way for further studies, that would make our model more complex. However, in a first step, it is crucial to start with a limited number of parameters that can be rigorously determined, and this is how this first work was conceived.

Public Reviews:

Reviewer #1 (Public review):

The authors present an important work where they model some of the complex interactions between immune cells, fibroblasts and cancer cells. The model takes into account the increased ECM production of cancer-associated fibroblasts. These fibres trap the cancer but also protect it from immune system cells. In this way, these fibroblasts’ actions both promote and hinder cancer growth. By exploring different scenarios, the authors can model different cancer fates depending on the parameters regulating cancer cells, immune system cells and fibroblasts. In this way, the model explores non-trivial scenarios. An important weakness of this study is that, though it is inspired by NSCLC tumors, it is restricted to modelling circular tumor lesions and does not explore the formation of ramified tumors, as in NSCLC. In this way, is only a general model and it is not clear how it can be adapted to simulate more realistic tumor morphologies.

We thank the reviewer for highligting the importance of our work. We acknowledge that although we provided anisotropic geometries and the study of the coalescence in the first version, more effort was needed to provide tools to extend our formalism to non-ideal cases. This is now added as Section III-4, where we analyze the impact of blood vessels, and the anisotropic friction due to the nematic order for the fibers; this nematic order can also be used to introduce active nematic stress.

Reviewer #2 (Public review):

Summary:

The authors develop a computational model (and a simplified version thereof) to treat an extremely important issue regarding tumor growth. Specifically, it has been argued that fibroblasts have the ability to support tumor growth by creating physical conditions in the tumor microenvironment that prevent the relevant immune cells from entering into contact with, and ultimately killing, the cancer cells. This inhibition is referred to as immune exclusion. The computational approach follows standard procedures in the formulation of models for mixtures of different material species, adapted to the problem at hand by making a variety of assumptions as to the activity of different types of fibroblasts, namely ”normal” versus ”cancer-associated”. The model itself is relatively complex, but the authors do a convincing job of analyzing possible behaviors and attempting to relate these to experimental observations.

Strengths:

As mentioned, the authors do an excellent job of analyzing the behavior of their model both in its full form (which includes spatial variation of the concentrations of the different cellular species) and in its simplified mean field form. The model itself is formulated based on established physical principles, although the extent to which some of these principles apply to active biological systems is not clear (see Weaknesses). The results of the model do offer some significant insights into the critical factors which determine how fibroblasts might affect tumor growth; these insights could lead to new experimental ways of unraveling these complex sets of issues and enhancing immunotherapy.

We thank the referee for this summary and for recognizing the strengths of our paper.

Weaknesses:

Models of the form being studied here rely on a large number of assumptions regarding cellular behavior. Some of these seemed questionable, based on what we have learned about active systems. The problem of T cell infiltration as well as the patterning of the extracellular matrix (ECM) by fibroblasts necessarily involve understanding cell motion and cell interactions due e.g. to cell signaling. Adopting an approach based purely on physical systems driven by free energies alone does not consider the special role that active processes can play, both in motility itself and in the type of self-organization that can occur due to these cell-cell interactions. This to me is the primary weakness of this paper.

We thank the referee for this important comment, that allows us to clarify this important point. Although biological materials are out of equilibrium, their behavior often resembles that dictated by thermodynamics. Hence the usefulness of constructing a free energy, in terms of these variables. In a first approach to decipher the complex interactions and describe the different and sometimes non-trivial outcomes in this system that involves many components, we must start by minimizing the number of parameters, and identifying those complex processes, that control the evolution of the system. The free energy that we build on this biological system contains therefore out-of-equilibrium processes that can be approximated by a ”close to equilibrium” description. Our approach is a classical one in statistical physics of active systems, namely in the effort to construct an equivalent free-energy for out-of-equilibrium systems. This allows to gain a clearer insight into those complex processes.

We have added a sentence in the main text, section III.1, to clarify this point:

“Building a free-energy density for a biological material is justified, because, although biological materials are out of equilibrium, their behavior often resembles that dictated by thermodynamics. It is therefore useful to write a free energy in terms of state variables.”

Nevertheless, we recognize that we should have provided more tools for using our formalism by making it active. This is why we introduced the nematic order in the fibers in Section III-4. This nematic order can be used to introduce active stress, and we have cited previous works by some of us see [?, ?, ?] as references for building active processes out of it.

We must also note that cell signaling has been introduced a minima in our system for providing the cue for the arrival of T-cells and NAFs from the boundaries. However, we found that although we had evoked the other role of the chemicals in the transformation from NAFs to CAFs in the text, details were not well explained. We have therefore corrected and added some explanations in the introduction of section III, and III.1, III.2.

A separate weakness concerns the assumption that fibroblasts affect T cell behavior primarily by just making a more dense ECM. There are a number of papers in the cancer literature (see, for some examples, Carstens, J., Correa de Sampaio, P., Yang, D. et al. Spatial computation of intratumoral T cells correlates with survival of patients with pancreatic cancer. Nat Commun 8, 15095 (2017);Sun, Xiujie, Bogang Wu, Huai-Chin Chiang, Hui Deng, Xiaowen Zhang, Wei Xiong, Junquan Liu et al. ” Tumour DDR1 promotes collagen fibre alignment to instigate immune exclusion.” Nature 599, no. 7886 (2021): 673-678) that seem to indicate that density alone is not a sufficient indicator of T cell behavior. Instead, the organization of the ECM (for example, its anisotropy) could be playing a much more essential role than is given credit for here. This possibility is hinted at in the Discussion section but deserves much more emphasis.

The referee is right in his comment, and we thank him for raising this issue. We have therefore introduced the anisotropic orientation of the fibers, which induces an anisotropic friction in a new section III-4. In addition, the references pointed out were included in this section. However, although the anisotropy strongly influences the fate of the tumor when the fibers are oriented perpendicular to the surface of the cancer nest, it is less effective when the fibroblasts are oriented in the direction of surface of the cancer nest. In the latter case, which is often the case before cancer cells reshape the tumor microenvironment, the matrix density should correlate with the friction.

Finally, the mixed version of the model is, from a general perspective, not very different from many other published models treating the ecology of the tumor microenvironment (for a survey, see Arabameri A, Asemani D, Hadjati J (2018), A structural methodology for modeling immune-tumor interactions including pro-and anti-tumor factors for clinical applications. Math Biosci 304:48-61). There are even papers in this literature that specifically investigate effects due to allowing cancer cells to instigate changes in other cells from being tumor-inhibiting to tumor-promoting. This feature occurs not only for fibroblasts but also for example for macrophages which can change their polarization from M1 to M2. There needed to be some more detailed comparison with this existing literature.

The referee is right that the first part of our approach, namely the dynamical system may be common in this kind of system, and it needs to be mentioned. So we added the following sentence in the discussion: ”This is in line with several similar mathematical models, that study through this lens the inhibition/activation of the immune system by cancer cells either by means of compartmental nonlinear models similar to our dynamical system, for instance regarding macrophage recruitment and cytokine signaling {arabameri2018structural} {li2019computational}, or mixture models {fotso2024mixture}. We combine the two approaches in order to rigorosly derive the parameters of the model and gain insights from both.”

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

The authors should address the following points:

Major issues

(1) The shape of tumors simulated differs immensely from the observed tumors in Fig. 2. Here, the tumor is constituted by irregular domains, not dissimilar from domains in phase separating mixtures. The domains simulated are circular. Since the authors are using the space dependent model to model the increase in tumor cells with time in the different scenarios (immune-desert, immune-excluded, immune inflamed), it should explain how non-spherical tumor structures can be observed in these scenarios. The authors introduce tumor coalescence in page 28, however, it is not expected that the structures observed in Fig 2 are the result from different tumors merging and coalescing, because that would result from an unlikely large number of initial mutation events in the same region of the tissue. The authors should explain what mechanisms present in the model can lead to non-spherical forms.

We agree with the reviewer that real tumors are rarely round contrary to what our numerics suggests. In fact, only the last figure of our paper in the supporting information was more appropriate for such a discussion. We are now adding discussions and new figures to better illustrate our spatial model, see Figure 6 and section III-4. The in situ geometry of tumors depends on the shape of the host organ, the diffusive (chemical) or advected species such as T cells and fibroblasts, and on the nutrients. Thus, in our case, only cancer cells are produced locally, but during growth the tumor is strongly constrained by the microenvironment, and thus the geometry of the domain we model in the numerics and its boundary conditions. This is also true for the chemicals responsible for growth, cellular advection and phenotypic transformation. Their concentration depends on a convection-diffusion equation and boundary conditions. For a tumor in situ, such as in the lung, the available space is a constraint that will dominate the final geometry of the tumor nests. We do not think that coalescence is controlled by mutational events, but most likely by the search for space necessary for growth. Compared to the first version, we add new figures (Figure 6) that show that the geometry of the organ, as well as the localization of blood vessels, are a cause of the irregularity of the tumor shapes. We also introduce orientational order, which as suggested in section III-4, can induce anisotropic friction and stresses, as well as anisotropic growth. We cite (Ackermann, Joseph, and Martine Ben Amar. ”Onsager’s variational principle in proliferating biological tissues, in the presence of activity and anisotropy.” The European Physical Journal Plus 138.12 (2023): 1103.) where we described active stresses and coupling related to anisotropic growth.

(2) According to the authors, the model presented in equations (1) and onwards simulates the evolution of the fraction of tumor cells in the tissue. However the fraction of tumor cells, for example, depends itself on the variation of other cell types. For example, if fibroblasts were to proliferate with rate alpha, even without tumor cells proliferating, the fraction of tumor cells in the mixture should decrease as alpha times the tumor cells fraction. These terms are missing. The equations do not describe the evolution of the cells’ fractions but of the amount of cells of each type, normalised by the total carrying capacity of non-normal cells in the tissue. The text should be rewritten accordingly.

We agree with the referee: our definition of cell density was not precise enough and may appear misleading. In the paragraph II1, we more explictly introduce the word mass fraction which is the correct physical quantity to introduce into the spatial model.

”All these cells have the same mass density and the sum of their mass fraction satisfies the relationship S = C + T + F_NA + F_A = 1-N, where N is a healthy non active component as healthy cells, for example.”

It is less intuitive than ”number of cells per unit volume” but necessary for the following (III)

(3) The authors start by calculating fixed points of different versions of the dynamical system without spatial dependence. They should explain what is the relevance of these fixed points: in a real situation, where the concentration of tumor fibroblasts and T-cells depend on position, in which conditions are these fixed points relevant?

The referee is right and we will clarify this point: the dynamic analysis is a help for understanding and predicting the scenario occurring in the system. After all the steps of paragraph 2.2, we are faced with 11 independent parameters only for the dynamical system and without the parameters generated by the space modeling itself. Our estimation concerns only lung cancer. These parameters do not appear in the literature. The parameters introduced in Sec. III which are more related to physical interactions such as friction, cell-cell adhesion, etc. can be found in the literature or can be estimated and thus measured in in vitro experiments (see Ackermann and Ben Amar, EPJP 2023, P. Benaroch, J. Nikolic et al. 2024, biorxiv). So what are the fixed points for: they help to get the right numbers for spatial analysis. To recover special features of cancer evolution, we need a model, but also correct estimates of the data in a code that is quite technical and heavy, with each simulation taking a certain amount of time. For users who only need rough predictions, the analysis in section 2 is sufficient.

It is also important to note that the global result depends only on the source terms, and on the boundary conditions. This can be illustrated with a simple example: Consider the governing equation for the density of a component with velocity v and source term:

Integrating the equation over a fixed volume V of surface S gives:

. This integrated equation can then be approximated by the dynamical system that we write. Thus, while the dynamical system does not give any information about the local structure of the system, it may be indicative of its global outcome.

(4)   In page 15, the authors identify that α_NA is proportional to δ𝝐⁴. However, in equation (7), they replace α_NA by δ𝝐⁴ without the proportionality constant. This should be corrected.

Thank you for your remark. This typo is now corrected.

(5) The tumor cell movement should be much slower than the T-cells. Here, the authors assign a similar friction coefficient for the cancer cells and T-cells, for example. However, in lung cancer tumor cells are epithelial, and adhere to each other in the tissue. Their movement is very restricted by the basement membranes and by cell-cell adhesion. Immune cells and T-cells on the other hand move rapidly throughout the stroma. It is a gross simplification to not consider the low epitelial tissue mobility in the context of lung cancer.

It is possible to assume different friction coe cients for each phase pair. This has been done in a previous publication, Ackermann et al., Physics report 2021. It is also possible to play with the cell-cell adhesion in the energy density and on the diffusion coe cient introduced in the Flory-Higgins free energy. Cell-cell adhesion is taken into account in the energy, and this makes the tumor a more dense phase, while T-cells can move towards cancer cells to which they are attracted. In the last part of the paper, we show the role of an anisotropic friction due to a nematic order for activated fibroblasts and all the other cells

(6) What is the biological mechanism by which the T-cells form a colony with a surface tension? In the phase-field model, the authors have a surface tension assigned to the cancer cells, T-cells and fibroblasts. Can the authors justify biologically why do they consider these surface tensions?

The fact that T-cells form a colony is due to the accumulation of T-cells at the outer boundary of the tumor, as they are attracted to it but cannot penetrate due to the strong cell-cell adhesion of the tumor cells in the nest. Adding a gradient square is standard in continuous models to limit the sharp variations. In a continuous approach, the gradient square contribution limits the sharp variations in cell density which are not physical.

Minor issues

(a) Page 6 (end), characterisation of the fibre barrier produced by CAFs missing: what is the fibre density, how it can hinder the spread of cancer and T-cell motility? Is it so dense that it prevents ameboid movement? Can cells move through it using matrix degradation proteins?

The fiber density corresponds to the fibrous organic extracellular matrix secreted by cancer-associated fibroblasts. In desmotic (highly fibrous tumors such as PDAC or NSCLC), this extracellular matrix deposited around the tumor forms a physical barrier around the tumor nest, preventing both cell migration and capillary and immune cells penetration. In these cases, the fibrous belt actually prevents ameboid movement and cells must deform significantly to migrate. The role of this barrier was particularly demonstrated in the reference (Grout, John A., et al. ”Spatial positioning and matrix programs of cancer-associated fibroblasts promote T-cell exclusion in human lung tumors.” Cancer Discovery 12.11 (2022): 2606-2625.). In later stages of cancer, the tumor may adapt and develop strategies to metastasize, such as matrix degradation. This matrix can be oriented, organized or disordered. To build a minimal model, we first considered an isotropic friction and also an anisotropic friction of the nematic belt, due to the activated fibroblasts. In the case of T-cells, as mentioned in section I.1, it is true that the biological literature also considers a phenotypic transformation of the T cells by the activated fibroblasts: this concerns both their proliferative capacities, antigen recognition and also their cytotoxic function. To better document the different mechanisms, we add the following publication: Cancer associated fibroblasts-an impediment to effective anti-cancer T cell immunity, by Koppensteiner, Lilian and Mathieson, Layla and O’Connor, Richard A and Akram, Ahsan R, Frontiers in immunology (2022).

However, our goal is to build a minimal model and to characterize and quantify the physical process in which CAFs are involved, namely the role of a physical barrier, that has been documented, as documented above.

(b) Page 19 (Fig 3), in the figure legend it is written ”resting fibroblasts”, should be ”non-activated fibroblasts”.

The referee is right: it will be better to write non-activated fibroblasts. This is now changed in the main text.

(c) Page 21 (equation), what is dΩ? It is dr?

We thank the referee for raising this point. The text was indeed ambiguous as sometimes dΩ was replaced by dr. To be clearer, all the elements of volume are now noted dV , and the element of surface of the system are noted dS.

In the article the units are in italic and should be in roman.

Thank you for raising this point. It has been corrected.

(d) Page 25 (beginning section III.3), the authors mention that the simulation is 2D, however, the simulation has radial symmetry. A 1D simulation in radial coordinates could simulate a 3D spherical system. Is the simulation of this section equivalent to a 1D radial simulation (in 2D)?

The referee is right that in radial symmetry, a 1d equation may be written. We therefore present numerics with irregular shapes of the tumor nest in order to make the system fully 2d.

(e) Page 26 (Fig 4). Legends inside the plots of plates A, B, C and D are not clear. Colorbar range of plates A and D is different. Would facilitate if the ranges were the same.

The referee is right: the surface plots presented in figure 4 would be easier to compare with the same colorbar range for the legends. In fact, as the referee noted, figures in A, B and C have the same legends, while figure in D has a different one. This is due to the fact that D represents the case of the immune-inflamed tumor where the cancer mass fraction is quite vanishing, resulting in values that are of 3 orders of magnitude lower than those present in A, B and C. Therefore, they would disappear if the colorbar range were equal to the others.We insist more on the change of scale in the legend of Figure 4, in the new version.

(f) Page 29 (Fig 5), would facilitate if the order of immune-desert, immune-excluded, immune-inflamed was maintained throughout the document. In this figure the immune-inflamed case appears first.

We agree with the reviewer that following the same order in which the different cases are presented throughout the manuscript would be helpful in comparing the different figures. Therefore, we have modified Figure 5.

(g) Page 31, the authors indicate that pharmacodynamics and pharmacokinetics are highly dependent on tumour spatial structure. Can they provide examples and citations?

In the discussion, we have added references concerning pharmacodynamics.

(h) Page 33 (Fig Sup2), would facilitate if the order of immune-desert, immune-excluded, immune-inflamed was maintained throughout the document. ±±

We thank the reviewer for pointing this out, the order of the different scenarios in Fig Sup 2 has now been changed.

Reviewer #2 (Recommendations for the authors):

Major points

(1) Following on from the discussion in the public review, I feel that there are a number of critical issues that need to be addressed regarding modeling assumptions. I would like to understand why the authors believe it is possible to use a free energy-driven model of the microenvironment when many of the processes relevant for their study have an undeniably ”active media” flavor.

The referee is right that processes in biology are active processes. However, it is a classical approach to model physical interactions between biological components with a free-energy, especially cell adhesion, as they often lead to quasi-stationary equilibrium-like patterns. The free-energy approach has also the advantage to derive straight-forwardly complex phenomena involving many components. Activity can indeed be introduced in such a framework, if we know that the fibroblasts transform into myo-fibroblasts, see for example our previous publication Ackermann and Ben Amar, EPJP 2023. However, in the interest of simplification and reduction of the number of free parameters, we have not not considered further complication of the model here, as a minimal model allows to distinguish the main processes that occur. Nevertheless, introducing more precisely activity, in the nematic approach already achieved for the friction, is a natural continuation of our work: See the new Section III-4, where we introduce the nematic order, and we indicate that active nematic stresses can be written from it.

Next, I don’t understand the assumption that T cells do not proliferate once they detect neoantigens on the cancer cells; activation of T cells usually causes them to become more proliferative.

We thank the referee for this question. The T-cell fraction has two origins: proliferation of T-cells in situ in the stroma or inside tumor nest or external arrival from the sources that we privilege. We recognize that a full analysis of the tumor-microenvironment would require to consider proliferation near the tumor, as many more other processes which is do able but requires the knowledge of more biological date. In addition, besides, the proliferation of T-cells will be equivalent to increase the killing abilities of T-cells and these two effect overlapp in our approach.

In order to clarify this point, we modify the following sentence in Section II.2:

“Although proliferation of cytotoxic T-cells has been observed, we do not consider explicitly proliferation in our study as we focus on their ability to infiltrate the tumor.”

Rather, we consider that T-cells proliferate outside the domain boundaries, so that this proliferation is included in the boundary source contributions.

Finally, the issue of whether the density of fibers is sufficient to understand the role of fibroblasts is not at all settled. There should be a full discussion of this issue including mentioning of the Nature paper (cited in the public review) that argues that orientation (and not density) is the key to the role of fibers, as well as the earlier cited work of Kalluri and collaborators on the role of ECM density in pancreatic cancer.

We thank the referee for this remark. As we wrote above in the response to the public review, we introduced significant additions that aim to tackle this question in the article.

(2) The authors present a picture of a tumor cell with fibroblasts apparently arrayed circumferentially around the tumor boundary and therefore blocking infiltration. This type of tumor structure has been seen before, for example in ”On the mechanism of long-range orientational order of fibroblasts.” Proceedings of the National Academy of Sciences 114, no. 34 (2017): 8974-8979, which should be cited. More importantly, in that paper the argument is made that positive feedback between fibroblasts and ECM geometry can cause structures like this to form. If this is indeed what is occurring, this would indicate the crucial importance of a mechanism beyond what is contained in the current model. This issue should therefore be discussed within this paper. This issue is of course connected to the previous point regarding the role of ECM structure beyond density.

We completely agree that the interplay between the fibroblast layer and the tumor shapes the tumor boundary. One of the authors has worked recently on this precise topic (Aging and freezing of active nematic dynamics of cancer-associated fibroblasts by fibronectin matrix remodeling, C Jacques, J Ackermann, S Bell, C Hallopeau, CP Gonzalez, ... bioRxiv, 2023.11. 22.568216, Ordering, spontaneous flows and aging in active fluids depositing tracks S Bell, J Ackermann, A Maitra, R Voituriez arXiv preprint arXiv:2409.05195). Since the fibroblast layer is an active material, it contributes to an anisotropic stress that can be introduced into the model. Our first strategy was to present the simplest modeling in order to focus on the most important interactions as cell-cell adhesion and cell-tissue adhesion. However, we recognize that those questions should be discussed in the text, and we discuss it in the new section III-4

Minor points

There are also a number of more minor points to consider:

(1) Since the parameter is taken to be O(1), why exactly does it matter how the other parameters scale with it?

It is very important to compare the order of magnitude of the other parameters once the selected parameter of order O(1) is really the driving parameter of the coupling. It gives a first picture of the main interactions that has to consider.

(2) I didn’t understand the relevance of referring specifically to IL 6 among many other possibly relevant signals, as is currently done on page 7.

This corresponds to studies aiming to correlate lung cancer risks and the concentration of interleukin, mostly IL6 and IL8 (McKeown, D. J., et al. ”The relationship between circulating concentrations of C-reactive protein, inflammatory cytokines and cytokine receptors in patients with non-small-cell lung cancer.” British journal of cancer 91.12 (2004): 1993-1995.,Brenner, Darren R., et al. ”Inflammatory cytokines and lung cancer risk in 3 prospective studies.” American journal of epidemiology 185.2 (2017): 86-95. ) but in the absence of very detailed biological information, the modeling and its results are not modified if other chemicals intervene..We slightly modeified the following phrase in section I.1:

“In particular, in the family of inflammatory proteins, also called cytokines, Interlukin-6 (IL6) and (IL8) seem, among others to stimulate the infiltration of CD8⁺.

(3) The authors need to mention the possibility of T-cell chemotaxis to the tumor being ”self-amplified” in the T cell system, as put forth in Galeano Nin˜o, Jorge Luis, Sophie V. Pageon, Szun S. Tay, Feyza Colakoglu, Daryan Kempe, Jack Hywood, Jessica K. Mazalo et al. ”Cytotoxic T cells swarm by homotypic chemokine signalling.” eLife 9 (2020): e56554. This might again reveal a needed extension of the current modelling strategy.

We thank the referee for his/her comment on the self-amplification of T-cell population in the stroma and we mention the indicated reference in our paper. This auto-chemoatactic process which induces a dynamic of more e cient recruitment towards the tumor, may be important for immunotherapy. To have more e cient T-cell arriving at the site of the tumor, will lead a better issue for the patient, if the swarming organization is maintained in a desmoplastic nematic stroma.

(4) It is not obvious to me that in sub figures 3F and 3H the tumor is enroute to being totally eradicated, as is stated in the text. The blue lines seemed to asymptote at non-zero population values.

Looking at sub-figures 3F and 3H, we stated in the main text that the tumor is eradicated as the representative population approaches a 0 value fraction, or at least decays around the 0 (0.01/0.05 to be more precise). This is even more evident when compared with the other cases where the tumor mass fraction reaches values of a higher order (up to 0.6), thus leading us to dinstinguish between these different scenarios.

(5) The description of the interaction of cells with fibers as being increased friction might be misleading, as the real effect could be actual trapping in the network (as opposed to just slowing down the motion).

We thank the referee for this question as it allow us to make an important distinction. Indeed, what the referee describes seems to correspond to a discrete event, namely a cell trapped in a network. However, coarse-graining the dynamics to the continuous modeling seems to us as leading to an effective friction between the two phases. Moreover, we also now introduced an anisotropic friction which can represent a trapping. The velocities are not only directed around the tumor but can also be oriented towards the tumor, so that eventually the friction along the radius mimics a trapping (see Fig.4 on top). We have introduced this anisotropic friction via a nematic model, see the appendix.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Valk and Engert et al. examined the potential relations between three different mental training modules, hippocampal structure and functional connectivity, and cortisol levels over a 9-month period. They found that among the three types of mental training: Presence (attention and introspective awareness), Affect (socio-emotional - compassion and prosocial motivation), and Perspective (socio-cognitive - metacognition and perspective taking) modules; Affect training most consistently related to changes in hippocampal structure and function - specifically, CA1-3 subfields of the hippocampus. Moreover, decreases in diurnal cortisol correlated to bilateral increases in volume, and decreases in diurnal and chronic cortisol left CA1-3 functional connectivity. Chronic cortisol levels also related to right CA4/DG volume and left subiculum function. The authors demonstrate that mindfulness training programs impact hippocampus and are a potential avenue for stress interventions, a potential avenue to improve health. The data contribute to the literature on plasticity of hippocampal subfields during adulthood, the impact of mental training interventions on the brain, and the link between CA1-3 and both short- and long-term stress changes. Additional clarification and extension of the methods is needed to strengthen the authors' conclusions.

We thank the Reviewer for their positive evaluation and summary of our findings and work. We made additional changes as suggested by the Reviewer and hope this clarified any open points.

(1) The authors thoughtfully approached the study of hippocampal subfields, utilizing a method designed for T1w images that outperformed Freesurfer 5.3 and that produced comparable results to an earlier version of ASHS. However, given the use of normalized T1-weighted images to delineate hippocampal subfield volume, some caution may be warranted (Wisse et al. 2020). While the authors note the assessment of quality control processes, the difficulty in ensuring valid measurement is an ongoing conversation in the literature. This also extends to the impact of functional co-registration using segmentations. I appreciate the inclusion of Table 5 in documenting reasons for missing data across subjects. Providing additional details on the distribution of quality ratings across subfields would help contextualize the results and ensure there is equal quality of segmentations across subfields.

We thank the Reviewer for bringing up this point. In the current work, we assessed the overall segmentation of all six subfields per individual. Thus, unfortunately, we have no data of quality of segmentation of individual subfields beyond our holistic assessment. Indeed, registration of hippocampal subfields remains a challenge and we have further highlighted this limitation in the Discussion of the current work.

“It is of note that the current work relies on a segmentation approach of hippocampal subfields including projection to MNI template space, an implicit correction for total brain volume through the use of a stereotaxic reference frame. Some caution for this method may be warranted, as complex hippocampal anatomy can in some cases lead to over- as well as underestimation of subfield volumes, as well as subfield boundaries may not always be clearly demarcated (1). Future work, studying the hippocampal surface at higher granularity, for example though unfolding the hippocampal sheet (2-5), may further help with both alignment and identification of not only subfield-specific change but also alterations as a function of the hippocampal long axis, a key dimension of hippocampal structural and functional variation that was not assessed in the current work (6, 7).”

(2) Given the consistent pattern of finding results with CA1-3, in contrast to other subfields, it would help to know if the effects of the different training modules on subfields differed from each other statistically (i.e., not just that one is significant, and one is not) to provide an additional context of the strength of results focused on Affect training and CA1-3 (for example, those shown in Figure 3).

Our work investigated i) whether the effects of the individual Training Modules differed from each other statistically. We found that the Affect Training Module showed increases in CA1-3 volume, and that these increases remained when testing effects relative to changes in this subfield following Perspective training and in retest controls. Moreover, in CA1-3 we found changes in functional connectivity when comparing the Affect to Perspective training Module. These changes were only present in this contrast, but not significant in each of the Training Modules per se. To test for specificity, we additionally evaluated whether subfield-specific changes were present above and beyond changes in the other ipsilateral hippocampal subfields. Relative to other subfields, right CA1-3 showed increases in the Affect vs Perspective contrast (left: t-value: 2.298, p=0.022, Q>0.1; right: t-value: 3.045, p=0.0025, Q=0.015). No other subfield showed significant changes. We now include this statement in the revised Results and Supplementary Tables.

“Moreover, associations between CA1-3 and Affect, relative to Perspective, seemed to go largely above and beyond changes in the other subfields (left: t-value: 2.298, p=0.022, Q>0.1; right: t-value: 3.045, p=0.0025, Q=0.015, see further Supplementary File 1h).”

Author response table 1.

Subfield-specific changes following the Training Modules, controlling for the other two ipsilateral subfields

Reviewer #1 (Recommendations For The Authors):

(1) In Figure 1, using different colors for subfields versus the modules (yellow, red, green) would help as it could lead the reader to try to draw connections between the two when it is namely a depiction of the delineations.

As suggested, we updated Figure 1 accordingly and present the subfields in different shades of purple for clarity. Please find the updated figure below.

Author response image 1.

(2) In the Results, it was at times hard to follow when Affect off Perspective where the focus of the results. Perhaps the authors could restructure or add additional context for clarity.

We are happy to clarify. For the first analysis on Module-specific changes in hippocampal subfield volume, we compared effects across Training Modules. Here, main contrasts were ran between subjects: Presence vs active control and within subjects: Affect versus Perspective. In additional secondary contrasts, we studied training effects vs retest control. After observing consistent increases in bilateral CA1-3 following Affect, in the following analysis, we evaluated 1) intrinsic functional networks in main and supplementary contrasts and 2) diurnal cortisol measures within the Training modules only and all three Training Modules combined, and also adopted 3) a multivariate approach (PLS) (see comments Reviewer 2). We now also report effects of cortisol change on structural and functional subfield change in Presence and Perspective, for additional completeness and clarity.

“To study whether there was any training module-specific change in hippocampal subfield volumes following mental training, we compared training effects between all three Training Modules (Presence, Affect, and Perspective). Main contrasts were: Presence vs Active control (between subjects) and Affect vs Perspective (within subjects). Supplementary comparisons were made vs retest controls and within training groups.”

“Overall, for all hippocampal subfields, findings associated with volume increases in CA1-3 fol-lowing the Affect training were most consistent across timepoints and contrasts (Supplementary File 1a-f).”

“Subsequently, we studied whether hippocampal CA1-3 would show corresponding changes in intrinsic function following the Affect mental training.”

“In particular, the moderately consistent CA1-3 volume increases following Affect training were complemented with differential functional connectivity alterations of this subfield when comparing Affect to Perspective training”

“Last, we probed whether group-level changes in hippocampal subfield CA1-3 volume would correlate with individual-level changes in diurnal cortisol indices (Presence: n= 86; Affect: n=92; Perspective: n=81), given that the hippocampal formation is a nexus of the HPA-axis (8). We took a two-step approach. First, we studied associations between cortisol and subfield change, particularly focusing on the Affect module and CA1-3 volume based on increases in CA1-3 volume identified in our group-level analysis.”

“We observed that increases in bilateral CA1-3 following Affect showed a negative association with change in total diurnal cortisol output […]”

“We did not observe alterations in CA1-3 volume in relation to change in cortisol markers in Presence or Perspective. Yet, for Presence, we observed association between slope and LCA4/DG change (t=-2.89, p=0.005, q=0.03), (Supplementary File 1uv).”

“In case of intrinsic function, we also did not observe alterations in CA1-3 in relation to change in cortisol markers in Presence or Perspective, nor in other subfields (Supplementary File 1wx).”

Author response table 2.

Correlating change in subfield volume and diurnal cortisol indices in Presence. Main focus was on CA1-3 based on volumetric observations and are highlighted in bold.

Author response table 3.

Correlating change in subfield volume and diurnal cortisol indices in Perspective. Main focus was on CA1-3 based on volumetric observations and are highlighted in bold.

Author response table 4.

Association between stress-markers and within functional network sub-regions in Affect and Perspective.

Author response table 5.

Correlating change in subfield function and diurnal cortisol indices in Presence. Main focus was on CA1-3 based on volumetric observations and are highlighted in bold. For these multiple comparisons (FDRq, corrected for two subfields) values are reported if uncorrected p values are below p<.05.

Author response table 6.

Correlating change in subfield function and diurnal cortisol indices in Perspective. Main focus was on CA1-3 based on volumetric observations and are highlighted in bold. For these multiple comparisons (FDRq, corrected for two subfields) values are reported if uncorrected p values are below p<.05.

(3) In the Methods, the authors note that corrections for multiple comparisons were used where needed, throughout the manuscript there is some switching between corrected and uncorrected p-values. At times, this made it difficult to follow in terms of when these corrections were needed.

For clarity, we added explicit multiple comparisons information a) in main and supplementary results, and b) wherever extra information was needed. Also, we only included main contrasts in Table 1-3 to avoid confusion and moved the information on changes in SUB and CA4/DG to the Supplementary tables.

(4) Typically, when correcting for intracranial volume the purpose is the ensure that sexual dimorphism in the size of the brain is accounted for. I would recommend the authors assess whether sex differences are accounted for by the MNI normalization approach taken. In the reading of the original Methods paper for the patch-based algorithm used, ICV was used to transform to MNI152 space. It would help to have additional information on how the normalization was done in the current study in order to draw comparisons to other findings in the literature.

We are happy to further clarify. In the current work, we used the same approach as in the original paper. Volumes were linearly registered to the MNI template using FSL flirt. We now provided this additional information in the revised methods.

“Hippocampal volumes were estimated based on T1w data that were linearly registered to MNI152 using FSL flirt (http://www.fmrib.ox.ac.uk/fsl/), such that intracranial volume was implicitly controlled for.”

We agree with the Reviewer that sex differences may still be present, and investigated this. At baseline, sex differences were found in all subfields in the left hemisphere, and right CA4/DG (FDRq<0.05). Regressing out ICV resolved remaining sex differences. We then evaluated whether main results of volumetric subfield change were impacted by ICV differences. Differences between Affect and Perspective remained stable. We have now added this additional analysis in the Supplementary Materials.

“Although stereotaxic normalization to MNI space would in theory account for global sex differences in intra-cranial volume, we still observed sex differences in various subfield volumes at baseline. Yet, accounting for ICV did not impact our main results suggesting changes in CA1-3 following Affect were robust to sex differences in overall brain volume (Supplementary File1j).”

Author response table 7.

Sex differences (female versus male) in hippocampal subfield volumes.

Reviewer #2 (Public Review):

In this study, Valk, Engert et al. investigated effects of stress-reducing behavioral intervention on hippocampal structure and function across different conditions of mental training and in relation to diurnal and chronic cortisol levels. The authors provide convincing multimodal evidence of a link between hippocampal integrity and stress regulation, showing changes in both volume and intrinsic functional connectivity, as measured by resting-state fMRI, in hippocampal subfield CA1-3 after socio-affective training as compared to training in a socio-cognitive module. In particular, increased CA1-3 volume following socio-affective training overlapped with increased functional connectivity to medial prefrontal cortex, and reductions in cortisol. The conclusions of this paper are well supported by the data, although some aspects of the data analysis would benefit from being clarified and extended.

A main strength of the study is the rigorous design of the behavioral intervention, including test-retest cohorts, an active control group, and a previously established training paradigm, contributing to an overall high quality of included data. Similarly, systematic quality checking of hippocampal subfield segmentations contributes to a reliable foundation for structural and functional investigations.

We thank the Reviewer for the thoughtful summary and appreciation of our work, as well as requests for further clarification and analyses. We addressed each of them in a point by point fashion below.

Another strength of the study is the multimodal data, including both structural and functional markers of hippocampal integrity as well as both diurnal and chronic estimates of cortisol levels.

(1) However, the included analyses are not optimally suited for elucidating multivariate interrelationships between these measures. Instead, effects of training on structure and function, and their links to cortisol, are largely characterized separately from each other. This results in the overall interpretation of results, and conclusions, being dependent on a large number of separate associations. Adopting multivariate approaches would better target the question of whether there is cortisol-related structural and functional plasticity in the hippocampus after mental training aimed at reducing stress.

We thank the Reviewer for this suggestion. Indeed, our project combined different univariate analyses to uncover the association between hippocampal subfield structure, function, and cortisol markers. While systematic, a downside of this approach is indeed that interpretation of our results depend on a large number of analyses. To further explore the question whether there is cortisol-related structural and functional plasticity in the hippocampus, we followed the Reviewer’s suggestion and additionally adopted a multivariate partial least squares (PLS) model. We ran two complementary models. One focusing on the bilateral CA1-3, as this region showed increases in volume following Affect training and differential change between Affect and Perspective training in our resting state analyses and one model including all subfields. Both models included all stress markers. We found that both models could significantly relate stress markers to brain measures, and that in particular Affect showed strong associations with significant the latent markers. Both analyses showed inverse effects of structure and function in relation to stress markers and both slope and AUC changes showed strongest loadings. We now include these analyses the revised manuscript.

Abstract

“Of note, using a multivariate approach we found that other subfields, showing no group-level changes, also contributed to alterations in cortisol levels, suggesting circuit-level alterations within the hippocampal formation.”

Methods

“Partial least squares analysis

To assess potential relationships between cortisol change and hippocampal subfield volume and functional change, we performed a partial least squares analysis (PLS) (9, 10). PLS is a multivariate associative model that to optimizes the covariance between two matrices, by generating latent components (LCs), which are optimal linear combinations of the original matrices (9, 10). In our study, we utilized PLS to analyze the relationships between change in volume and intrinsic function of hippocampal subfields and diurnal cortisol measures. Here we included all Training Modules and regressed out effects of age, sex, and random effects of subject on the brain measures before conducting the PLS analysis. The PLS process involves data normalization within training groups, cross-covariance, and singular value decomposition. Subsequently, subfield and behavioral scores are computed, and permutation testing (1000 iterations) is conducted to evaluate the significance of each latent factor solution (FDR corrected). We report then the correlation of the individual hippocampal and cortisol markers with the latent factors. To estimate confidence intervals for these correlations, we applied a bootstrapping procedure that generated 100 samples with replacement from subjects’ RSFC and behavioral data.”

Results

“Last, to further explore the question whether there is concordant cortisol-related structural and functional plasticity in the hippocampus we adopted a multivariate partial least square approach, with 1000 permutations to account for stability (9, 10) and bootstrapping (100 times) with replacement. We ran two complementary models including all Training Modules whilst regressing out age, sex and random effects of subject. First, we focused on the bilateral CA1-3, as this region showed increases in volume following Affect training and differential change between Affect and Perspective training in our resting state analyses. In the second model included structural and functional data of all subfields. Both models included all stress markers. We found that both models could identify significant associations between cortisol stress markers and hippocampal plasticity (FDRq<0.05), and that in particular Affect showed strongest associations with the latent markers for CA1-3 (Table 5). Both analyses showed inverse effects of subfield structure and function in relation to stress markers and both slope and AUC changes showed strongest associations with the latent factor.”

Author response table 8.

Multivariate PLS analyses linking cortisol markers to hippocampal subfield volume and function.

Discussion

“Last, performing multivariate analysis, we again observed associations between CA1-3 volume and function plasticity and stress change, strongest in Affect. Yet combining all subfields in a single model indicated that other subfields also link to stress alterations, indicating that ultimately circuit-level alterations within the hippocampal formation relate to latent changes in diurnal stress markers across Training Modules.”

“This interpretation is also supported by our multivariate observations.”

“In line with our observations in univariate analysis, we found multivariate associations between hippocampal subfield volume, intrinsic function and cortisol markers. Again, the contribution of volume and intrinsic function was inverse. This may possibly relate to the averaging procedure of the functional networks. Combined, outcomes of our univariate and multivariate analyses point to an association between change in hippocampal subfields and stress markers, and that these changes, at the level of the individual, ultimately reflect complex interactions within and across hippocampal subfields and may capture different aspects of diurnal stress. Future work may more comprehensively study the plasticity of the hippocampal structure, and link this to intrinsic functional change and cortisol to gain full insights in the specificity and system-level interplay across subfields, for example using more detailed hippocampal models (3). Incorporating further multivariate, computational, models is needed to further unpack and investigate the complex and nuanced association between hippocampal structure and function, in particular in relation to subfield plasticity and short and long-term stress markers.”

“…based on univariate analysis. Our multivariate analysis further nuanced this observation, but again pointed to an overall association between hippocampal subfield changes and cortisol changes, but this time more at a systems level.”

“Lastly, our multivariate analyses also point to a circuit level understanding of latent diurnal stress scores.”

Author response image 2.

Multivariate associations between changes in structure and function of hippocampal subfield volume and markers of stress change in Affect. A) Multivariate associations between bilateral CA1-3 volume and intrinsic function and stress markers. Left: Scatter of loadings, colored by Training Module; Right upper: individual correlations of stress markers; Right lower: individual correlation of subfields; B). Multivariate associations between all subfields’ volume and intrinsic function and stress markers. Left: Scatter of loadings, colored by Training Module; Right upper: individual correlations of stress markers; Right lower: individual correlation of subfields.

(2) The authors emphasize a link between hippocampal subfield CA1-3 and stress regulation, and indeed, multiple lines of evidence converge to highlight a most consistent role of CA1-3. There are, however, some aspects of the results that limit the robustness of this conclusion. First, formal comparisons between subfields are incomplete, making it difficult to judge whether the CA1-3, to a greater degree than other subfields, display effects of training.

We thank the Reviewer for this comment. To further test for specificity, we additionally evaluated subfield-specific changes relative to other subfields for our main contrasts (Presence versus Active Control and Affect versus Perspective). Relative to other subfields, right CA1-3 showed increases in the Affect vs Perspective contrast (left: t-value: 2.298, p=0.022, Q>0.1; right: t-value: 3.045, p=0.0025, Q=0.015); no other subfield showed significant changes. We now include this statement in Results and Supplementary Tables.

“Moreover, associations between CA1-3 and Affect, relative to Perspective, seemed to go largely above and beyond changes in the other subfields (left: t-value: 2.298, p=0.022, Q>0.1; right: t-value: 3.045, p=0.0025, Q=0.015, see further Supplementary File 1h).”

Author response table 9.

Subfield-specific changes following the Training Modules, controlling for the other two ipsilateral subfields

(3) Relatedly, it would be of interest to assess whether changes in CA1-3 make a significant contribution to explaining the link between hippocampal integrity and cortisol, as compared to structure and functional connectivity of the whole hippocampus.

We thank the Reviewer for this comment. Please see the PLS analysis performed above (R2Q1). Indeed, not only CA1-3 but also other subfields seem to show a relationship with cortisol, in line with circuit level accounts on stress regulation and hippocampal circuit alterations (8, 11-15).

(4) Second, both structural and functional effects (although functional to a greater degree), were most pronounced in the specific comparison of "Affect" and "Perspective" training conditions, possibly limiting the study's ability to inform general principles of hippocampal stress-regulation.

We agree with the Reviewer that the association between stress and hippocampal plasticity, on the one hand, and mental training and hippocampal plasticity, on the other hand, make it not very straightforward to inform general principles on hippocampal stress regulation. However, as underscored in the discussion, in previous work we could also link mental training to stress reductions(16-18). We hope that the additional analyses and explanations further explain the multilevel insights of the current work, on the one hand using group-level analysis to investigate and illustrate the association between mental training and hippocampal subfield volume and intrinsic function, and on the other hand using individual level analysis to unpack the association between cortisol change and hippocampal subfield change.

Reviewer #2 (Recommendations For The Authors):

(1) In the Results, the description of how the hippocampal subfields' functional networks were defined would benefit from some clarification. It is also somewhat unclear what is meant by (on page 10): "Evaluating functional connectivity changes, we found that connectivity of the right CA1-3 functional network showed differential changes when comparing Affect training to Perspective training (2.420, p=0.016, FDRq=0.032, Cohens D =0.289), but not versus retest control (Table 1 and Supplementary Table 8-14)." Were there significant changes in CA1-3 FC following both training conditions (but these differed from each other)? A description of what this difference reflected would increase the reader's understanding.

We are happy to clarify. We included information of change of individual modules in the Supplementary materials, Supplementary Table 1 and 2, 9 and 10. Changes for functional connectivity were largely due to the differences in Modules, but did not show strong effects in one Module alone. We now include information on Affect and Perspective un-contrasted change in the main results text:

“… which could be attributed to decreases in right CA1-3 mean FC following Perspective (t=-2.012, p=0.045, M:-0.024, std: 0.081, CI [-0.041 -0.006]), but not Affect (t=1.691, p=0.092, M: 0.010, std: 0.098, CI [-0.01 0.031]); changes were not present when comparing Affect training versus retest control (Table 1 and Supplementary File 1k-q).”

(2) As described in the Public Review, the lack of multivariate assessments may risk selling the data short. Including analyses of concomitant functional and structural changes, in relation to cortisol, seems like an approach better adapted to characterize meaningful interrelationships between these measures.

We thank the Reviewer for suggesting multivariate assessments. To understand the interrelation between behavioral intervention, hippocampal plasticity, and cortisol changes, the current work first evaluates a simpler operationalization of the relationship between hippocampal subfield structure and volume, and cortisol as a function of mental training. Thus, given the complex nature of the study, we initially opted for a model where we assess structural and functional changes independently, with structural changes as the basis of our investigations. Now we have also included a multivariate approach (PLS) to further test the association between hippocampal subfields and cortisol markers, please see our additions to the manuscript above. We now highlighted multivariate associations in the Discussion as well, and suggest this as an important next step for more detailed, future investigations.

“Incorporating further multivariate, computational, models is needed to further unpack and investigate the complex and nuanced association between hippocampal structure and function, in particular in relation to subfield plasticity and short and long-term stress markers.”

(3) A minor comment regards the Figures. Some main effects should be visualized in a clearer manner. For instance, the scatterplots in Figure 1, panel D. Also, some of the current headings within the figures could be made more intuitive to the reader.

We thank the Reviewer for this comment. To improve clarity, we updated figure headings. For Figure 1D, the challenge is that the data are quite scattered and we aimed to visualize our observations in a naturalistic way. Therefore, we added additional y-axis information to further clarify the figures. Creating more overlap or differentiation would make other elements of the figure less clear, hence we remained with the current set-up detailing the intra- and inter-individual alterations of the current model.

(1) Wisse LEM, Chetelat G, Daugherty AM, de Flores R, la Joie R, Mueller SG, et al. (2021): Hippocampal subfield volumetry from structural isotropic 1 mm(3) MRI scans: A note of caution. Hum Brain Mapp. 42:539-550.

(2) DeKraker J, Kohler S, Khan AR (2021): Surface-based hippocampal subfield segmentation. Trends Neurosci. 44:856-863.

(3) DeKraker J, Haast RAM, Yousif MD, Karat B, Lau JC, Kohler S, et al. (2022): Automated hippocampal unfolding for morphometry and subfield segmentation with HippUnfold. Elife. 11.

(4) Vos de Wael R, Lariviere S, Caldairou B, Hong SJ, Margulies DS, Jefferies E, et al. (2018): Anatomical and microstructural determinants of hippocampal subfield functional connectome embedding. Proc Natl Acad Sci U S A. 115:10154-10159.

(5) Bernhardt BC, Bernasconi A, Liu M, Hong SJ, Caldairou B, Goubran M, et al. (2016): The spectrum of structural and functional imaging abnormalities in temporal lobe epilepsy. Ann Neurol. 80:142-153.

(6) Vogel JW, La Joie R, Grothe MJ, Diaz-Papkovich A, Doyle A, Vachon-Presseau E, et al. (2020): A molecular gradient along the longitudinal axis of the human hippocampus informs large-scale behavioral systems. Nat Commun. 11:960.

(7) Genon S, Bernhardt BC, La Joie R, Amunts K, Eickhoff SB (2021): The many dimensions of human hippocampal organization and (dys)function. Trends Neurosci. 44:977-989.

(8) McEwen BS (1999): Stress and hippocampal plasticity. Annu Rev Neurosci. 22:105-122.

(9) Kebets V, Holmes AJ, Orban C, Tang S, Li J, Sun N, et al. (2019): Somatosensory-Motor Dysconnectivity Spans Multiple Transdiagnostic Dimensions of Psychopathology. Biol Psychiatry. 86:779-791.

(10) McIntosh AR, Lobaugh NJ (2004): Partial least squares analysis of neuroimaging data: applications and advances. Neuroimage. 23 Suppl 1:S250-263.

(11) Paquola C, Benkarim O, DeKraker J, Lariviere S, Frassle S, Royer J, et al. (2020): Convergence of cortical types and functional motifs in the human mesiotemporal lobe. Elife. 9.

(12) DeKraker J, Ferko KM, Lau JC, Kohler S, Khan AR (2018): Unfolding the hippocampus: An intrinsic coordinate system for subfield segmentations and quantitative mapping. Neuroimage. 167:408-418.

(13) McEwen BS, Nasca C, Gray JD (2016): Stress Effects on Neuronal Structure: Hippocampus, Amygdala, and Prefrontal Cortex. Neuropsychopharmacology. 41:3-23.

(14) Sapolsky RM (2000): Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Arch Gen Psychiatry. 57:925-935.

(15) Jacobson L, Sapolsky R (1991): The role of the hippocampus in feedback regulation of the hypothalamic-pituitary-adrenocortical axis. Endocr Rev. 12:118-134.

(16) Engert V, Hoehne K, Singer T (2023): Specific reduction in the cortisol awakening response after socio-affective mental training. Mindfulness.

(17) Puhlmann LMC, Vrticka P, Linz R, Stalder T, Kirschbaum C, Engert V, et al. (2021): Contemplative Mental Training Reduces Hair Glucocorticoid Levels in a Randomized Clinical Trial. Psychosom Med. 83:894-905.

(18) Engert V, Kok BE, Papassotiriou I, Chrousos GP, Singer T (2017): Specific reduction in cortisol stress reactivity after social but not attention-based mental training. Sci Adv. 3:e1700495.

Author response:

The following is the authors’ response to the original reviews

Public Reviews: 

Reviewer #1 (Public review): 

Summary: 

Govindan and Conrad use a genome-wide CRISPR screen to identify genes regulating retention of intron 4 in OGT, leveraging an intron retention reporter system previously described (PMID: 35895270). Their OGT intron 4 reporter reliably responds to O-GlcNAc levels, mirroring the endogenous splicing event. Through a genome-wide CRISPR knockout library, they uncover a range of splicing-related genes, including multiple core spliceosome components, acting as negative regulators of OGT intron 4 retention. They choose to follow up on SFSWAP, a largely understudied splicing regulator shown to undergo rapid phosphorylation in response to O-GlcNAc level changes (PMID: 32329777). RNA-sequencing reveals that SFSWAP depletion not only promotes OGT intron 4 splicing but also broadly induces exon inclusion and intron splicing, affecting decoy exon usage. While this study offers interesting insights into intron retention and O-GlcNAc signaling regulation, the RNA sequencing experiments lack the essential controls needed to provide full confidence to the authors' conclusions. 

Strengths: 

(1) This study presents an elegant genetic screening approach to identify regulators of intron retention, uncovering core spliceosome genes as unexpected positive regulators of intron retention. 

(2) The work proposes a novel functional role for SFSWAP in splicing regulation, suggesting that it acts as a negative regulator of splicing and cassette exon inclusion, which contrasts with expected SR-related protein functions. 

(3) The authors suggest an intriguing model where SFSWAP, along with other spliceosome proteins, promotes intron retention by associating with decoy exons. 

We thank the reviewer for recognizing and detailing the strengths of our manuscript. 

Weaknesses: 

(1) The conclusions on SFSWAP impact on alternative splicing are based on cells treated with two pooled siRNAs for five days. This extended incubation time without independent siRNA treatments raises concerns about off-target effects and indirect effects from secondary gene expression changes, potentially limiting confidence in direct SFSWAP-dependent splicing regulation. Rescue experiments and shorter siRNA-treatment incubation times could address these issues. 

We repeated our SFSWAP knockdown analysis and analyzed both OGT e4-e5 junction splicing and SFSWAP transcript levels by RT-qPCR (now included in Sup. Fig. S4) from day 2 to day 5 post siRNA treatment. We observed that the time point at which OGT intron 4 removal increases (day 2) coincides with the time at which SFSWAP transcript levels start decrease, consistent with a direct effect of SFSWAP knockdown on OGT intron 4 splicing. Moreover, the effect of SFSWAP knockdown on OGT intron 4 splicing peaks between day 4-5, supporting our use of these longer time points to cast a wide net for SFSWAP targets.

(2) The mechanistic role of SFSWAP in splicing would benefit from further exploration. Key questions remain, such as whether SFSWAP directly binds RNA, specifically the introns and exons (including the decoy exons) it appears to regulate. Furthermore, given that SFSWAP phosphorylation is influenced by changes in O-GlcNAc signaling, it would be interesting to investigate this relationship further. While generating specific phosphomutants may not yield definitive insights due to redundancy and also beyond the scope of the study, the authors could examine whether distinct SFSWAP domains, such as the SR and SURP domains, which likely overlap with phosphorylation sites, are necessary for regulating OGT intron 4 splicing. 

We absolutely agree with the reviewer that the current work stops short of a detailed mechanistic study, and we have made every attempt to be circumspect in our interpretations to reflect that limitation. In addition, we are very interested in delving more deeply into the mechanistic aspects of this regulation. In fact, we have initiated many of the experiments suggested by the reviewer (and more), but in each case, rigorous interpretable results will require a minimum another year’s time. 

For example, we have used crosslinking and biotin labeling techniques (using previously available reagents from Eclipsebio) to test whether SFSWAP binds RNA. The results were negative, but the lack of strong SFSWAP antibodies required that we use a transiently expressed myc-tagged SFSWAP. Therefore, this negative result could be an artifact of the exogenous expression and/or tagging. Given the difficulties of “proving the negative”, considerably more work will be required to substantiate this finding. As another example, we intend to develop a complementation assay as suggested. For an essential gene, the ideal complementation system employs a degron system, and we have spent months attempting to generate a homozygous AID-tagged SFSWAP. Unfortunately, we so far have only found heterozygotes. Of course, this could be because the tag interferes with function, the insert was not efficiently incorporated by homologous repair, or that we simply haven’t yet screened a sufficient number of clones. We’re confident that these technical issues that can be addressed, but they will take a significant amount of time to resolve. While we would ideally define a mechanism, we think that the data reported here outlining functions for SFSWAP in splicing represent a body of work sufficient for publication. 

(3) Data presentation could be improved (specific suggestions are included in the recommendations section). Furthermore, Excel tables with gene expression and splicing analysis results should be provided as supplementary datasheets. Finally, a more detailed explanation of statistical analyses is necessary in certain sections. 

We have addressed all specific suggestions as detailed in the recommendations below.

Reviewer #2 (Public review): 

Summary: 

The paper describes an effort to identify the factors responsible for intron retention and alternate exon splicing in a complex system known to be regulated by the O-GlcNAc cycling system. The CRISPR/Cas9 system was used to identify potential factors. The bioinformatic analysis is sophisticated and compelling. The conclusions are of general interest and advance the field significantly. 

Strengths: 

(1) Exhaustive analysis of potential splicing factors in an unbiased screen. 

(2) Extensive genome wide bioinformatic analysis. 

(3) Thoughtful discussion and literature survey. 

We thank the reviewer for recognizing and detailing the strengths of our manuscript. 

Weaknesses: 

(1) No firm evidence linking SFSWAP to an O-GlcNAc specific mechanism. 

We couldn’t agree more with this critique. Indeed, our intention at the outset for the screen was to find an O-GlcNAc sensor linking OGT splicing with O-GlcNAc levels. As often occurs with high-throughput screens, we didn’t find exactly what we were looking for, but the screen nonetheless pointed us to interesting biology. Prompted by our screen, we describe new insights into the function of SFSWAP a relatively uncharacterized essential gene. Currently, we are testing other candidates from our screen, and we are performing additional studies to identify potential O-GlcNAc sensors.  

(2) Resulting model leaves many unanswered questions. 

We agree (see Reviewer 1, point 2 response).  

Reviewer #3 (Public review): 

Summary: 

The major novel finding in this study is that SFSWAP, a splicing factor containing an RS domain but no canonical RNA binding domain, functions as a negative regulator of splicing. More specifically, it promotes retention of specific introns in a wide variety of transcripts including transcripts from the OGT gene previously studied by the Conrad lab. The balance between OGT intron retention and OGT complete splicing is an important regulator of O-GlcNAc expression levels in cells. 

Strengths: 

An elegant CRISPR knockout screen employed a GFP reporter, in which GFP is efficiently expressed only when the OGT retained intron is removed (so that the transcript will be exported from the nucleus to allow for translation of GFP). Factors whose CRISPR knockdown causes decreased intron retention therefore increase GFP, and can be identified by sequencing RNA of GFP-sorted cells. SFSWAP was thus convincingly identified as a negative regulator of OGT retained intron splicing. More focused studies of OGT intron retention indicate that it may function by regulating a decoy exon previously identified in the intron, and that this may extend to other transcripts with decoy exons. 

We thank the reviewer for recognizing the strengths of our manuscript. 

Weaknesses: 

The mechanism by which SFSWAP represses retained introns is unclear, although some data suggests it can operate (in OGT) at the level of a recently reported decoy exon within that intron.

Interesting/appropriate speculation about possible mechanisms are provided and will likely be the subject of future studies. 

We completely agree that this is a limitation of the current study (see above). Now that we have a better understanding of SFSWAP functions, we will continue to explore SFSWAP mechanisms as suggested. 

Overall the study is well done and carefully described but some figures and some experiments should be described in more detail. 

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors): 

(1) Clarify and add missing statistical details across the figures. For example, Figure S2 lacks statistical comparisons, and in Figures 4A and 4C the tests applied should be specified in the legend. 

We have added appropriate statistical analysis wherever missing and edited figure legends to specify the tests used.

(2) The authors are strongly encouraged to provide detailed tables of gene expression and alternative splicing analyses from RNA-Seq experiments (e.g., edgeR, rMATS, Whippet, and MAJIQ), as this would enhance transparency and facilitate data interpretation. 

We have added tables for gene expression and alternate splicing analysis as suggested (Suppl. tables 3-

6).

(3) Although the legend sometimes indicates differently (e.g., Figure 3b, 5a, 5c, etc), the volcano plots showing the splicing changes do not contain a cutoff for marginally differential percent spliced in or intron retention values. 

The legends have been edited to reflect the correct statistical and/or PSI cutoffs.

(4) For consistency, use a consistent volcano plot format across all relevant figures (Figures 3b, 5a-c, S3, S4, S7, and S8), including cutoffs for differential splicing and the total count of up- and down-regulated events. 

Due to different statistical frameworks and calculations employed by different alternate splicing pipelines, we could not use the same cutoffs for different pipelines.  However, we have now indicated the number of up- and down-regulated events for consistency among the volcano plots.

(5) What is the overlap of differentially regulated events between the different analytical methodologies applied? 

We analyzed the degree of overlap between the three pipelines used in the paper using a Venn diagram (added to Suppl. Fig. S7). However, as widely reported in literature (e.g., Olofsson et al., 2023; Biochem Biophys Res Commun. 2023; doi: 10.1016/j.bbrc.2023.02.053.), the degree of overlap between pipelines is quite low.

(6) To further substantiate your conclusions, additional validations of RNA-Seq splicing data, ideally visualized on an agarose gel, would be valuable, especially for exons and introns regulated by SFSWAP, and particularly for OGT decoy exons in Figure 4c. 

We have not included these experiments as we focused on other critiques for this resubmission. Because the RNA-seq, RT-PCR and RT-qPCR data all align, we are confident that the products we are seeing are correctly identified and orthogonally validated (Figs 2d, 4a, 4b, and 4c).  

(7) It would be more informative if the CRISPR screen data were presented in a format where both the adjusted p-value and LFC values of the hits are presented. Perhaps a volcano plot? 

We have now included these graphs in revised Supplementary Figure S2. 

(8) In Figure 2d, a cartoon showing primer binding sites for each panel could aid interpretation, particularly in explaining the unexpected simultaneous increase in OGT mRNA and intron retention upon SFSWAP knockdown. 

We have added a cartoon showing primer binding sites similar to that shown in Fig. 4a.

(9) Page 9, line 1, states that SFSWAP autoregulates its expression by controlling intron retention. Including a Sashimi plot would provide visual support for this claim. 

The data suggesting that SFSWAP autoregulates its own transcript abundance were reported in Zachar et al. (1994), not from our own studies. Validation of those data with our RNA-seq data is confounded by the fact that we are using siRNAs to knockdown the SFSWAP RNA at the transcript level (Fig. S15). 

(10) In the legend of Figure S2 the authors state that negative results are inconclusive because RNA knockdowns are not verified by western blotting or qRT-PCR. This is correct, but the reviewer would also argue that the positive results are also inconclusive as they are not supported by a rescue experiment to confirm that the effect is not due to off-target effects. 

This is a fair point with respect to the siRNA experiments on their own. However, the CRISPR screen was performed with sgRNAs, and MAGeCK RRA scores are high only for those genes that have multiple sgRNAs that up-regulate the gene. Examination of the SFSWAP sgRNAs individually shows that three of four SFSWAP sgRNAs had false discovery rates ≤10^-42 for GFP upregulation. Thus, the siRNAs provide an additional orthogonal approach. It seems unlikely that the siRNAs, and three independent sgRNAs will have the same off-target results. Thus, these combined observations support the conclusion that SFSWAP loss leads to decreased OGT intron retention.  

(11) For clarity in Figure 3a, consider using differential % spliced in or intron retention bar plots with directionality (positive and negative axis) and labeling siSFSWAP as the primary condition. 

(12) Consider presenting Figure 5D as a box plot with a Wilcoxon test for statistical comparison. 

For both points 11 and 12, we have tried the graphs as the reviewer suggested. While these were good suggestions, in both cases we felt that the original plots ended up presenting a clearer presentation of the data (see Author response image 1).

Author response image 1.

(13) Please expand the Methods section to detail the Whippet and MAJIQ analyses. 

We have expanded the methods section to include additional details of the alternate splicing analysis.

(14) Include coordinates for the four possible OGT decoy exon combinations analyzed in the Methods section. 

We have added the coordinates of all four decoy forms in the methods section.  

(15) A section on SFSWAP mass spectrometry is listed in Methods but is missing from the manuscript. 

This section has now been removed.

Reviewer #2 (Recommendations for the authors): 

This is an excellent contribution. The paper describes an effort to identify the factors responsible for intron retention and alternate exon splicing in a complex system known to be regulated by the O-GlcNAc cycling system. The CRISPR/Cas9 system was used to identify potential factors. The bioinformatic analysis is sophisticated and compelling. The conclusions are of general interest and advance the field significantly. 

Some specific recommendations. 

(1) The plots in Figure 3 describing SI and ES events are confusing to this reader. Perhaps the violin plot is not the best way to visualize these events. The same holds true for the histograms in the lower panel of Figure 3. Not sure what to make of these plots. 

For Figure 3b, we include both scatter and violin plots to represent the same data in two distinct ways. For Figure 3d, we agree that these are not the simplest plots to understand, and we have spent significant time trying to come up with a better way of displaying these trends in GC content as they relate to SE and RI events. Unfortunately, we were unable to identify a clearer way to present these data. 

(2) The model (Figure 6) is very useful but confusing. The legend and the Figure itself are somewhat inconsistent. The bottom line of the figure is apparent but I fear that the authors are trying to convey a more complete model than is apparent from this figure. Please revise. 

We have simplified the figure from the previous submission. As mentioned above, we admit that mechanistic details remain unknown. However, we have tried to generate a model that reflects our data, adds some speculative elements to be tested in the future, but remains as simple as possible. We are not quite sure what the reviewer was referring to as “somewhat inconsistent”, but we have attempted to clarify the model in the revised Discussion and Figure legend.  

(3) It is unclear how normalization of the RNA seq experiments was performed (eg. Figure S5 and 6).  

The normalization differences in Fig. S5 and S6 (now Fig S8 and S9) were due to scaling differences during the use of rmats2sashimiplot software. We have now replaced Fig. S5 to reflect correctly scaled images.

I am enthusiastic about the manuscript and feel that with some clarification it will be an important contribution. 

Thank you for these positive comments about our study!

Reviewer #3 (Recommendations for the authors): 

(1) In Figure 1f, it is clear that siRNA-mediated knockdown of OGT greatly increases spliced RNA as the cells attempt to compensate by more efficient intron removal (three left lanes). However, there is no discussion of the various treatments with TG or OSMI. Might quantitation of these lanes not also show the desired effects of TG and OSMI on spliced transcript levels? 

The strong effect of OGT knockdown masks the (comparatively modest) effects of subsequent inhibitor treatments on the reporter RNA. We have edited the results section to clarify this.

(2) In Figure 2c, why is the size difference between spliced RNA and intron-retained RNA so different in the GFP-probed gel (right) compared with the OGT-probed gel (left)? Even recognizing that the GFP probe is directed against reporter transcripts, and the OGT probe (I think) is directed against endogenous OGT transcripts, shouldn't the difference between spliced and unspliced bands be the same, i.e., +/- the intron 4 sequence. Also, why does the GFP probe detect the unspliced transcript so poorly? 

The fully spliced endogenous OGT mRNA is ~5.5 kb while the fully spliced reporter is only ~1.6kb, so the difference in size (the apparent shift relative to the mRNA) is quite different. Moreover, the two panels in Fig 2c are not precisely scaled to one another, so direct comparisons cannot be made. 

The intron retained isoform does not accumulate to high levels in this reporter, a phenotype that we also observed with our GFP reporter designed to probe the regulation of the MAT2A retained intron (Scarborough et al., 2021). We are not certain about the reason for these observations, but suspect that the reporter RNA’s retained intron isoforms are less stable in the nucleus than their endogenous counterparts. Alternatively, the lack of splicing may affect 3´ processing of the transcripts so that they do not accumulate to the high levels observed for the wild-type genes. 

(3) Please provide more information about the RNA-seq experiments. How many replicates were performed under each of the various conditions? The methods section says three replicates were performed for the UPF1/TG experiments; was this also true for the SFSWAP experiments?  

All RNA-seq experiments were performed in biological triplicates. We have edited the methods section to clarify this.

(4) Relatedly, the several IGV screenshots shown in Figure 3C presumably represent the triplicate RNA seq experiments. In part D, how many experiments does the data represent? Is it a compilation of three experiments? 

Fig. 3d is derived from alternate splicing analysis performed on three biological replicates. We have added the number of replicates (n=3) on the figure to clarify this. We have also noted that the three IGV tracks represent biological replicates in the Figure legend for 3c.  

(5) Please provide more details regarding the qRT-PCR experiments. 

We have provided the positions of primer sets used for RT-qPCR analysis and cartoon depictions of target sites below the data wherever appropriate.

(6) In the discussion of decoy exon function (in the Discussion section), several relevant observations are cited to support a model in which decoy exons promote assembly of splicing factors. One might also cite the finding that eCLIP profiling has found enriched binding of U2AF1 and U2AF2 at the 5' splice site region of decoy exons (reference 16). 

Excellent point. This has now been added to the Discussion. 

Minor corrections / clarifications: 

(1) In the Figure 2A legend, CRISPR is misspelled. 

Corrected.

(2) In the discussion, the phrase "indirectly inhibits splicing of exons 4 and 5, but promoting stable unproductive assembly of the spliceosome", the word "but" should probably be "by". 

Corrected.

Author Response

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

Reviewer #1 (Public Review):

Su et al propose the existence of two mechanisms repressing SBF activity during entry into meiosis in budding yeast. First, a decrease in Swi4 protein levels by a LUTI-dependent mechanism where Ime1 would act closing a negative feedback loop. Second, the sustained presence of Whi5 would contribute to maintaining SBF inhibited under sporulation conditions. The article is clearly written and the experimental approaches used are adequate to the aims of this work. The results obtained are in line with the conclusions reached by the authors but, in my view, they could also be explained by the existing literature and, hence, would not represent a major advance in the field of meiosis regulation.

We respectfully disagree with the reviewer about their comment that this work can be explained by the existing literature. First, while SWI4LUTI has been previously identified in meiotic cells along with ~ 380 LUTIs, the biological purpose of these alternative mRNA isoforms and their effect on cellular physiology still remain largely unknown. Our manuscript clarifies this gap in understanding for SWI4LUTI. Loss of SWI4LUTI contributes to dysregulation of meiotic entry and does so by failing to properly repress the known inhibitors of meiotic entry, the CLNs. Furthermore, even though Cln1 and Cln2 have been previously shown to antagonize meiosis, the mechanisms that restrict their activity was unclear prior to our study.

We recognize work done by others demonstrating Whi5-dependent repression of SBF during mitotic G1/S transition (De Bruin et al., 2004; Costanzo et al., 2004). We further examined Whi5’s involvement during meiotic entry and found that it acts in conjunction with the LUTI-based mechanism to restrict SBF activity. Combined loss of both mechanisms results in the increased expression of G1 cyclins, decreased expression of early meiotic genes, and a delay in meiotic entry (Figure 6). Neither mechanism was previously known to regulate meiotic entry. Our study not only adds to our broader understanding of gene regulation during meiosis but also raises additional questions regarding how LUTIs regulate gene expression and function.

Regarding the first mechanism, Fig 1 shows that Swi4 decreases very little after 1-2h in sporulation medium, whereas G1-cyclin expression is strongly repressed very rapidly under these conditions (panel D and work by others). This fact dampens the functional relevance of Swi4 downregulation as a causal agent of G1 cyclin repression.

Reviewer 1 expresses concern for the observation that by 2 h in sporulation media there is a 32% decrease in Swi4-3V5 protein abundance compared to 0 h in SPO. This is consistent with the range of protein level decrease typically accomplished by LUTI-based gene regulation (Chen et al., 2017; Chia et al., 2017; Tresenrider et al., 2021), and while it is a modest reduction, it is consistent across replicates. Furthermore, we don’t make the argument that reduction in Swi4 levels alone is the sole regulator of G1 cyclin levels. In fact, we report that in addition to Swi4 downregulation, Whi5 also functions to restrict SBF activity during meiotic entry, thereby ensuring G1 cyclin repression.

In addition, the LUTI-deficient SWI4 mutant does not cause any noticeable relief in CLN2 repression, arguing against the relevance of this mechanism in the repression of G1-cyclin transcription during entry into meiosis. The authors propose a second mechanism where Whi5 would maintain SBF inactive under sporulation conditions. The role of Whi5 as a negative regulator of the SBF regulon is well known. On the other hand, the double WHI5-AA SWI4-dLUTI mutant does not upregulate CLN2, the G1 cyclin with the strongest negative effect on sporulation, raising serious doubts on the functional relevance of this backup mechanism during entry into meiosis.

Due to replicate variance, CLN2 did not make the cut by our mRNA-seq data analysis as a significant hit. To address reviewer 1’s final point we opted for the “gold standard” of reverse transcription coupled with qPCR to measure CLN2 transcript levels in the double mutant ∆LUTI; WHI5-AA and the wild-type control. This revealed that CLN2 levels were significantly increased in the double mutant compared to wild type at 2 h in SPO (Author Response Image 1, *, p = 0.0288, two-tailed t-test).

Author response image 1.

Wild type (UB22199) and ∆LUTI;WHI5-AA (UB25428) cells were collected to perform RT-qPCR for CLN2 transcript abundance. Transcript abundance was quantified using primer sets specific for each respective gene from three technical replicates for each biological replicate. Quantification was performed in reference to PFY1 and then normalized to wild-type control. FC = fold change. Experiments were performed twice using biological replicates, mean value plotted with range. Differences in wild type versus ∆LUTI; WHI5-AA transcript levels compared with a two-tailed t-test (*, p = 0.0288)

Reviewer #2 (Public Review):

Summary:

The manuscript highlights a mechanistic insight into meiotic initiation in budding yeast. In this study, the authors addressed a genetic link between mitotic cell cycle regulator SBF (the Swi4-Swi6 complex) and a meiosis inducing regulator Ime1 in the context of meiotic initiation. The authors' comprehensive analyses with cytology, imaging, RNA-seq using mutant strains lead the authors to conclude that Swi4 levels regulates Ime1-Ume6 interaction to activate expression of early meiosis genes for meiotic initiation. The major findings in this paper are that (1) the higher level of Swi4, a subunit of SBF transcription factor for mitotic cell cycle regulation, is the limiting factor for mitosis-to-meiosis transition; (2) G1 cyclins (Cln1, Cln2), that are expressed under SBF, inhibit Ime1-Ume6 interaction under overexpression of SWI4, which consequently leads to downregulation of early meiosis genes; (3) expression of SWI4 is regulated by LUTI-based transcription in the SWI4 locus that impedes expression of canonical SWI4 transcripts; (4) expression of SWI4 LUTI is likely negatively regulated by Ime1; (5) Action of Swi4 is negatively regulated by Whi5 (homologous to Rb)-mediated inhibition of SBF, which is required for meiotic initiation. Thus, the authors proposed that meiotic initiation is regulated under the balance of mitotic cell cycle regulator SBF and meiosis-specific transcription factor Ime1.

Strengths:

The most significant implication in their paper is that meiotic initiation is regulated under the balance of mitotic cell cycle regulator and meiosis-specific transcription factor. This finding will provide a mechanistic insight in initiation of meiosis not only into the budding yeast also into mammals. The manuscript is overall well written, logically presented and raises several insights into meiotic initiation in budding yeast. Therefore, the manuscript should be open for the field. I would like to raise the following concerns, though they are not mandatory to address. However, it would strengthen their claims if the authors could technically address and revise the manuscript by putting more comprehensive discussion.

Weaknesses:

The authors showed that increased expression of the SBF targets, and reciprocal decrease in expression of meiotic genes upon SWI4 overexpression at 2 h in SPO (Figure 2F). However, IME1 was not found as a DEG in Supplemental Table 1. Meanwhile, IME1 transcript level was decreased at 2 h SPO condition in pATG8-CLN2 cells in Fig S4C.

Now this reviewer still wonders with confusion whether expression of IME1 transcripts per se is directly or in directly suppressed under SBF-activated gene expression program at 2 h SPO in pATG8-SWI4 and pATG8-CLN2 cells. This reviewer wonders how Fig S4C data reconciles with the model summarized in Fig 6F.

One interpretation could be that persistent overexpression of G1 cyclin caused active mitotic cell cycle, and consequently delayed exit from mitotic cell cycle, which may have given rise to an apparent reduction of cell population that was expressing IME1. For readers to better understand, it would be better to explain comprehensively this issue in the main text.

We believe there was an oversight here. In supplemental table 1, IME1 expression is reported as significantly decreased. The volcano plot shown below also highlights this change (Author response image 2).

Author response image 2.

Volcano plot of DE-Seq2 analysis for ∆LUTI;WHI5-AA versus wild type. Dashed line indicates padj (p value) = 0.05. Analysis was performed using mRNA-seq from two biological replicates. Wild type (UB22199) and ∆LUTI;WHI5-AA (UB25428) cells were collected at 2 h in SPO. SBF targets (pink) (Iyer et al., 2001) and early meiotic genes (blue) defined by (Brar et al., 2012). Darker pink or darker blue, labeled dots are well studied targets in either gene set list.

The % of cells with nuclear Ime1 was much reduced in pATG8-CLN2 cells (Fig 2B) than in pATG8-SWI4 cells (Fig 4C). Is the Ime1 protein level comparable or different between pATG8-CLN2 strain and pATG8-SWI4 strain? Since it is difficult to compare the quantifications of Ime1 levels in Fig S1D and Fig S4B, it would be better to comparably show the Ime1 protein levels in pATG8-CLN2 and pATG8-SWI4 strains.

Further, it is uncertain how pATG8-CLN2 cells mimics the phenotype of pATG8-SWI4 cells in terms of meiotic entry. It would be nice if the authors could show RNA-seq of pATG8-CLN2/WT and/or quantification of the % of cells that enter meiosis in pATG8-CLN2.

Analyzing bulk Ime1 protein levels across a population of cells (Author response image 3) reveals that overexpression of CLN2 causes a more severe decrease in Ime1 levels than overexpression of SWI4. This is consistent with our observation that pATG8-CLN2 has a more severe impact on meiotic entry than pATG8-SWI4. The higher CLN2 levels (Author response image 4) likely accounts for the observed difference in severity of phenotype between the two mutants.

Author response image 3.

Samples from strain wild type (UB22199), pATG8-SWI4 (UB2226), pATG8-CLN2 (UB25959) and were collected between 0-4 hours (h) in sporulation medium (SPO) and immunoblots were performed using α-GFP. Hxk2 was used a loading control.

Author response image 4.

Wild type (UB22199), pATG8-SWI4 (UB2226), pATG8-CLN2 (UB25959) cells were collected to perform RT-qPCR for CLN2 transcript abundance. Quantification was performed in reference to PFY1 and then normalized to wild-type control. FC = fold change.

The authors stated that reduced Ime1-Ume6 interaction is a primary cause of meiotic entry defect by CLN2 overexpression (Line 320-322, Fig 4J-L). This data is convincing. However, the authors also showed that GFP-Ime1 protein level was decreased compared to WT in pATG8-CLN2 cells by WB (Fig S4A).

Compared to wild type, pATG8-CLN2 cells have lower levels of Ime1. Consequently, reviewer 2 suggests that this reduction may be responsible for the observed meiotic defect. However, we tested this possibility and found it not to be the primary cause of the meiotic defect in pATG8-CLN2 cells. As shown in Figure S4A, when IME1 was overexpressed from the pCUP1 promoter, Ime1 protein levels were similar between wild-type and pATG8-CLN2 cells. Despite this similarity, we still observed a decrease in nuclear Ime1 (Figure 4F) and no rescue in sporulation (Figure 4A). Therefore, the reduction in Ime1 protein levels alone cannot explain the meiotic defect caused by CLN2 overexpression.

Further, GFP-Ime1 signals were overall undetectable through nuclei and cytosol in pATG8-CLN2 cells (Fig 4B), and accordingly cells with nuclear Ime1 were reduced (Fig 4C). Although the authors raised a possibility that the meiotic entry defect in the pATG8-CLN2 mutant arises from downregulation of IME1 expression (Line 282-283), causal relationship between meiotic entry defect and CLN2 overexpression is still not clear.

As reviewer 2 comments, we initially considered the possibility that meiotic entry defect induced by CLN2 overexpression could be attributed to decreased IME1 expression. However, in the following paragraph in the manuscript, we demonstrate equalizing IME1 transcript levels using the pCUP1-IME1 allele does not rescue the meiotic defect caused by CLN2 overexpression. Consequently, we conclude that the decrease in IME1 transcript levels alone cannot explain the meiotic defect caused by increased CLN2 levels.

Is the Ime1 protein level reduced in the pATG8-CLN2;UME6-⍺GFP strain compared to WT? It would be better to comparably show the Ime1 protein levels in the pATG8-CLN2 strain and the pATG8-CLN2;UME6-⍺GFP strain by WB. Also, it would be nice if the authors could show quantification of the % of cells that enter meiosis in the pATG8-CLN2;UME6-⍺GFP strain to see how and whether artificial tethering of Ime1 to Ume6 rescued normal meiosis program rather than simply showing % sporulation in Fig4A.

We do not agree with the suggestion to compare the pATG8-CLN2;UME6-⍺GFP with wild type as the kinetics of meiosis is rather different. The more appropriate comparison is UME6-⍺GFP and pATG8-CLN2;UME6-⍺GFP which shows GFP-Ime1 bulk protein levels are slightly lower (Author response image 5). However, when we use a more sensitive measurement of meiotic entry through the nuclear accumulation of Ime1 in single cells, as illustrated in Figure 4L, it becomes evident that the Ume6-Ime1 tether is capable of restoring nuclear Ime1 levels, even in the presence of CLN2 overexpression. Given that these cells exhibited wild type levels of nuclear Ime1 and underwent sporulation after 24 hours, we make the fair assumption that they have successfully initiated the meiotic program.

Author response image 5.

Wild type (UB22199), pATG8-SWI4 (UB35106), UME6-⍺GFP (UB35300), and UME6-⍺GFP; pATG8-CLN2 (UB35177) cells collected between 0-3 hours (h) in sporulation medium (SPO) and immunoblots were performed using α-GFP. Hxk2 was used a loading control

The authors showed Ume6 binding at the SWI4LUTI promoter (Figure 5K). However, since Ume6 forms a repressive form with Rpd3 and Sin3a and binds to target genes independently of Ime1, Ume6 binding at the SWI4LUTI promoter bind does not necessarily represent Ime1-Ume6 binding there. Instead, it would be better to show Ime1 ChIP-seq at the SWI4LUTI promoter.

We agree with reviewer 2 that Ime1 ChIP would be the ideal measurement. Unfortunately, this has proved to be technically challenging. To address this limitation, we utilized a published Ume6 ChIP-seq dataset along with a published UME6-T99N RNA-seq dataset. Cells carrying the UME6-T99N allele are unable to induce the expression of early meiotic transcripts due to lack of Ime1 binding to Ume6 (Bowdish et al., 1995). Accordingly, RNA-seq analysis should reveal whether or not the LUTIs identified by Ume6 ChIP are indeed regulated by Ime1-Ume6 during meiosis. For SWI4LUTI, this is exactly what we observe. Not only is there Ume6 binding at the SWI4LUTI promoter (Figure 5K), but there is also a significant decrease in SWI4LUTI expression in UME6-T99N cells under meiotic conditions (Figure S5). Based on these data, we conclude that the Ime1-Ume6 complex is responsible for regulating SWI4LUTI expression during meiosis.

The authors showed ∆LUTI mutant and WHI5-AA mutant did not significantly change the expression of SBF targets nor early meiotic genes relative to wildtype (Figure 6A, C). Accordingly, they concluded that LUTI- or Whi5-based repression of SBF alone was not sufficient to cause a delay in meiotic entry (Line451-452), and perturbation of both pathways led to a significant delay in meiotic entry (Figure 6E). This reviewer wonders whether Ime1 expression level and nuclear localization of Ime1 was normal in ∆LUTI mutant and WHI5-AA mutant.

Based on our observations in Figure 4, Ime1 protein and expression levels were not reliable indicators of meiotic entry. Consequently, we opted for a more downstream and functionally relevant measure of meiotic entry, which involved time-lapse fluorescence imaging of Rec8, an Ime1 target.

Reviewer #1 (Recommendations For The Authors):

The authors would like to mention previous work showing that G1-cyclin overexpression decreases the expression and nuclear accumulation of Ime1 (Colomina et al 1999 EMBO J 18:320). In this work, the interaction between Ime1 and Ume6 had been found to be resistant to G1-cyclin expression, arguing against a direct effect on the recruitment of Ime1 at meiotic promoters. Alternatively, differences in the experimental approaches used could be discussed to explain this apparent discrepancy.

To clarify, in the paper that reviewer 1 is referring to (Colomina et al., 1999), the authors determine that the interaction between Ime1 and Ume6 is regulated by the presence of a non-fermentable carbon source. Additional work by others reveals that Ime1 undergoes phosphorylation by the protein kinases Rim11 and Rim15, promoting its nuclear localization and enabling interaction with Ume6 (Vidan and Mitchell, 1997; Pnueli et al., 2004; Malathi et al., 1999, 1997). Furthermore, both Rim11 and Rim15 kinase activities are inhibited by the presence of glucose via the PKA pathway (Pedruzzi et al., 2003; Rubin-Bejerano et al., 2004; Vidan and Mitchell, 1997). Accordingly, the elimination of cyclins in the presence of a non-fermentable carbon source (glucose) in (Colomina et al., 1999) is unlikely to result in an interaction between Ime1 and Ume6, as Rim11 and Rim15 remain repressed. Removal of cyclins in acetate does not further increase Ime1-Ume6 interaction leading the authors to conclude that G1 cyclins do not block Ime1 function through its interaction with Ume6. This work however uses loss of function (removal of G1 cyclins) to study the G1 cyclins’ effect on Ime1-Ume6 interaction while using timepoints that are well beyond meiotic entry. Additionally, Ime1-Ume6 interaction is being tested using yeast-two hybrid analysis with just the proposed interaction domain of Ime1 (amino acids 270-360). Therefore, the interpretation that G1 cyclins are dispensable for regulating the interaction between Ime1 and Ume6 is unclear from this work alone.

There are many differences that can explain the discrepancy between our work and (Colomina et al., 1999). Our work uses increased expression of cyclins during meiotic entry. Additionally, in our study, we collected timepoints to measure meiotic entry (2 h in SPO) and sporulation (gamete formation) efficiency (24 h in SPO). Finally, we are using the endogenous, full length Ime1. These differences could very well explain the discrepancy with previous work. Lastly, in our discussion we acknowledge the lack of CDK consensus phosphorylation sites on Ime1. Therefore, it is most likely that G1 cyclins are not directly phosphorylating Ime1 and that other factors like Rim11 and Rim15 could be direct targets of the G1 cyclins, considering their involvement in the phosphorylation of Ime1-Ume6, as well as their role in regulating Ime1 localization and its interaction with Ume6. We have included these points in the revised manuscript (lines 547-551).

Reviewer #2 (Recommendations For The Authors):

This reviewer thinks that the findings in this paper are of general interest to meiosis field and help understanding the mechanism of meiotic initiation in mammals. The way of the current manuscript seems to be written for limited budding yeast scientists, and should not limited to the interest by the budding yeast scientists. Thus, it would be better to discuss more about what is known about the mechanism of initiation of meiosis not only in budding yeast but also in other species to share their finding to more broad scientists using other organisms.

We appreciate reviewer 2’s comment and have added more discussion about the parallels between yeast and mammalian systems in meiotic initiation (lines 613-624).

Reviewer #3 (Recommendations For The Authors):

The effect of overexpression of Swi4 is tested for MI and MII (Fig1F): this is a very indirect readout of meiotic entry. The authors could present Rec8 localization (Fig2I) at this stage. However, this is still a superficial description of the meiotic phenotype: is the phenotype only a delay or is the meiotic prophase altered. It is specifically important to analyse this in more detail to answer whether the overexpression of Swi4 leads to an identical phenotype to the one of CLN2. Also the comparison between overexpression of Swi4 and Cln2 is difficult to evaluate: what is the level of CLN2 when SwI4 is overexpressed compared to CLN2 overexpression. The percentage of nuclear Ime1 is 50% vs 5% when Swi4 or Cln2 are overexpressed. What is the interpretation? What are the levels of Ime1? (Y axis of quantifications not comparable, see also comment for Fig5F,H)

CLN2 is expressed at a much higher level in pATG8-CLN2 cells relative to pATG8-SWI4 (Author Response Image 4). Therefore, we don’t expect identical phenotypes, but rather a more severe deficiency in meiotic entry upon CLN2 overexpression. The key experiment that establishes causality between SWI4 and CLNs is reported in Figure 3, where deletion of either CLN1 or CLN2 rescues the meiotic entry delay exerted by SWI4 overexpression.

Fig3EF: What is the phenotype of Cln1 and Cln2 without overexpression of Swi4?

Meiotic entry is not faster in cln1∆ or cln2∆ cells compared to wild-type. We included these data in Supplemental Figure 3 and made the relevant changes in the manuscript (lines 257-261).

Fig4F: Need a control with CLN2 overexpression only.

A control with only CLN2 overexpression (pATG8-CLN2) is not appropriate since these meiotic time course experiments are synchronized using the pCUP1-IME1 allele. It would be a misleading comparison since the two meiosis would have different kinetics. Figure 4F reports that despite similar IME1 transcript levels and Ime1 protein levels, CLN2 overexpressing cells still have reduced nuclear Ime1. Since side-by-side comparison of pATG8-CLN2 and pCUP1-IME1 is not possible, we chose to measure sporulation efficiency at 24 h in Figure 4A. These data together suggest that elevated IME1 transcript and protein levels cannot rescue the defects associated with increased CLN2 expression.

Fig5E: in wild type, by Northern blot, Swi4canon level is increasing during meiosis, not decreasing?, whereas protein level is decreasing, what is the interpretation?

Northern data is less quantitative than smFISH, which show that SWI4canon transcript levels are significantly lower in meiosis compared to vegetative cells (Figure 5D). We also note that the Northern blot data were acquired from unsynchronized meiotic cells and could have additional limitations based on the population-based nature of the assay. Finally, additional analysis of a transcript leader sequencing (TL-seq) dataset from synchronized cells (Tresenrider et al., 2021) further confirms the decrease in SWI4canon transcript levels upon meiotic entry. (Author response image 6).

Author response image 6.

TL-seq data from (Tresenrider et al. 2021) visualized on IGV at the SWI4 locus. Two timepoints are plotted including premeiotic before IME1 induction (pink) and meiotic prophase or after IME1 induction (blue).

Fig5F, H. This quantification needs duplicates for validation.

Replicates are submitted for every blot in this paper to eLIFE.It can be found in the shared Dropbox folder to the editors (named Raw-blots-for-eLIFE).

Fig5F, H. Why are the wild type values so different?

The immunoblotting done between Figure 5F and Figure 5H are on separate blots and therefore should not be compared. Additionally, these values are not absolute measurements of wild type values of Swi4-3V5 and therefore we should not expect them to be the same. Any comparisons done of relative amounts of Swi4-3V5 are always done on the same blot and normalized to a loading control, hexokinase.

FigS5: What is the effect of the Ume6-T99N on Swi4 protein level and on meiotic entry? Is the backup mechanism proposed active?

We haven’t measured Swi4 protein levels in the UME6-T99N background but given that this mutation is known to disrupt the interaction between Ime1 and Ume6, we expect a similar trend to that reported in Figure 5I (pCUP1-IME1 uninduced).

What is the evidence that Swi4/6 is a E2F homolog? What is the homology at the protein level?

While there is no sequence homology between SBF and E2F there is remarkable similarity between metazoans and yeast in terms of the regulation of the G1/S transition (reviewed in Bertoli et al., 2013). E2F and SBF are both repressed before the G1/S transition by the inhibitors Rb and Whi5, respectfully (Costanzo et al., 2004; De Bruin et al., 2004; Hasan et al., 2014). During G1/S transition, a cyclin dependent kinase phosphorylates and inactivates these inhibitors. We have carefully edited our language in the manuscript to “functional homology” instead of just “homology”.

FigS3 is missing

Each supplemental figure was matched to its corresponding main figure. In the original submission, we didn’t have Figure S3. However, the revised manuscript now contains FigS3.

Bertoli, C., J.M. Skotheim, and R.A.M. De Bruin. 2013. Control of cell cycle transcription during G1 and S phases. Nat. Rev. Mol. Cell Biol. 14:518–528. doi:10.1038/nrm3629.

Bowdish, K.S., H.E. Yuan, and A.P. Mitchell. 1995. Positive control of yeast meiotic genes by the negative regulator UME6. Mol. Cell. Biol. 15:2955–2961. doi:10.1128/mcb.15.6.2955.

Brar, G.A., M. Yassour, N. Friedman, A. Regev, N.T. Ingolia, and J.S. Weissman. 2012. High-Resolution View of the Yeast Meiotic Program Revealed by Ribosome Profiling. Science (80-. ). 335:552–558. doi:10.1126/science.1215110.

De Bruin, R.A.M., W.H. McDonald, T.I. Kalashnikova, J. Yates, and C. Wittenberg. 2004. Cln3 activates G1-specific transcription via phosphorylation of the SBF bound repressor Whi5. Cell. 117:887–898. doi:10.1016/j.cell.2004.05.025.

Chen, J., A. Tresenrider, M. Chia, D.T. McSwiggen, G. Spedale, V. Jorgensen, H. Liao, F.J. Van Werven, and E. Ünal. 2017. Kinetochore inactivation by expression of a repressive mRNA. Elife. 6:1–31. doi:10.7554/eLife.27417.

Chia, M., A. Tresenrider, J. Chen, G. Spedale, V. Jorgensen, E. Ünal, and F.J. van Werven. 2017. Transcription of a 5’ extended mRNA isoform directs dynamic chromatin changes and interference of a downstream promoter. Elife. 6:1–23. doi:10.7554/eLife.27420.

Colomina, N., E. Garí, C. Gallego, E. Herrero, and M. Aldea. 1999. G1cyclins block the Ime1 pathway to make mitosis and meiosis incompatible in budding yeast. EMBO J. 18:320–329. doi:10.1093/emboj/18.2.320.

Costanzo, M., J.L. Nishikawa, X. Tang, J.S. Millman, O. Schub, K. Breitkreuz, D. Dewar, I. Rupes, B. Andrews, and M. Tyers. 2004. CDK activity antagonizes Whi5, an inhibitor of G1/S transcription in yeast. Cell. 117:899–913. doi:10.1016/j.cell.2004.05.024.

Hasan, M., S. Brocca, E. Sacco, M. Spinelli, P. Elena, L. Matteo, A. Lilia, and M. Vanoni. 2014. A comparative study of Whi5 and retinoblastoma proteins : from sequence and structure analysis to intracellular networks. 4:1–24. doi:10.3389/fphys.2013.00315.

Iyer, V.R., C.E. Horak, P.O. Brown, D. Botstein, V.R. Iyer, M. Snyder, and C.S. Scafe. 2001. Genomic binding sites of the yeast cell-cycle transcription factors SBF and MBF. Nature. 409:533–538. doi:10.1038/35054095.

Malathi, K., Y. Xiao, and A.P. Mitchell. 1997. Interaction of yeast repressor-activator protein Ume6p with glycogen synthase kinase 3 homolog Rim11p. Mol. Cell. Biol. 17:7230–7236. doi:10.1128/mcb.17.12.7230.

Malathi, K., Y. Xiao, and A.P. Mitchell. 1999. Catalytic roles of yeast GSK3β/shaggy homolog Rim11p in meiotic activation. Genetics. 153:1145–1152. doi:10.1093/genetics/153.3.1145.

Pedruzzi, I., F. Dubouloz, E. Cameroni, V. Wanke, J. Roosen, J. Winderickx, and C. De Virgilio. 2003. TOR and PKA Signaling Pathways Converge on the Protein Kinase Rim15 to Control Entry into G0. Mol. Cell. 12:1607–1613. doi:10.1016/S1097-2765(03)00485-4.

Pnueli, L., I. Edry, M. Cohen, and Y. Kassir. 2004. Glucose and Nitrogen Regulate the Switch from Histone Deacetylation to Acetylation for Expression of Early Meiosis-Specific Genes in Budding Yeast. Mol. Cell. Biol. 24:5197–5208. doi:10.1128/mcb.24.12.5197-5208.2004.

Rubin-Bejerano, I., S. Sagee, O. Friedman, L. Pnueli, and Y. Kassir. 2004. The In Vivo Activity of Ime1, the Key Transcriptional Activator of Meiosis-Specific Genes in Saccharomyces cerevisiae, Is Inhibited by the Cyclic AMP/Protein Kinase A Signal Pathway through the Glycogen Synthase Kinase 3- Homolog Rim11. Mol. Cell. Biol. 24:6967–6979. doi:10.1128/mcb.24.16.6967-6979.2004.

Tresenrider, A., K. Morse, V. Jorgensen, M. Chia, H. Liao, F.J. van Werven, and E. Ünal. 2021. Integrated genomic analysis reveals key features of long undecoded transcript isoform-based gene repression. Mol. Cell. 81:2231-2245.e11. doi:10.1016/j.molcel.2021.03.013.

Vidan, S., and A.P. Mitchell. 1997. Stimulation of yeast meiotic gene expression by the glucose-repressible protein kinase Rim15p. Mol. Cell. Biol. 17:2688–2697. doi:10.1128/mcb.17.5.2688.

Author response:

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

eLife assessment

This study provides an important cell atlas of the gill of the mussel Gigantidas platifrons using a single nucleus RNA-seq dataset, a resource for the community of scientists studying deep sea physiology and metabolism and intracellular host-symbiont relationships. The work, which offers solid insights into cellular responses to starvation stress and molecular mechanisms behind deep-sea chemosymbiosis, is of relevance to scientists interested in host-symbiont relationships across ecosystems.

Public Reviews:

Reviewer #1 (Public Review):

Wang et al have constructed a comprehensive single nucleus atlas for the gills of the deep sea Bathymodioline mussels, which possess intracellular symbionts that provide a key source of carbon and allow them to live in these extreme environments. They provide annotations of the different cell states within the gills, shedding light on how multiple cell types cooperate to give rise to the emergent functions of the composite tissues and the gills as a whole. They pay special attention to characterizing the bacteriocyte cell populations and identifying sets of genes that may play a role in their interaction with the symbiotes.

Wang et al sample mussels from 3 different environments: animals from their native methane-rich environment, animals transplanted to a methane-poor environment to induce starvation, and animals that have been starved in the methane-poor environment and then moved back to the methane-rich environment. They demonstrated that starvation had the biggest impact on bacteriocyte transcriptomes. They hypothesize that the upregulation of genes associated with lysosomal digestion leads to the digestion of the intracellular symbiont during starvation, while the non-starved and reacclimated groups more readily harvest the nutrients from symbiotes without destroying them.

Strengths:

This paper makes available a high-quality dataset that is of interest to many disciplines of biology. The unique qualities of this non-model organism and the collection of conditions sampled make it of special interest to those studying deep sea adaptation, the impact of environmental perturbation on Bathymodioline mussels populations, and intracellular symbiotes. The authors do an excellent job of making all their data and analysis available, making this not only an important dataset but a readily accessible and understandable one.

The authors also use a diverse array of tools to explore their data. For example, the quality of the data is augmented by the use of in situ hybridizations to validate cluster identity and KEGG analysis provides key insights into how the transcriptomes of bacteriocytes change.

The authors also do a great job of providing diagrams and schematics to help orient non-mussel experts, thereby widening the audience of the paper.

Thank the reviewer for the valuable feedback on our study. We are grateful that the reviewers found our work to be interesting and we appreciate their thorough evaluation of our research. Their constructive comments will be considered as we continue to develop and improve our study.

Weaknesses:

One of the main weaknesses of this paper is the lack of coherence between the images and the text, with some parts of the figures never being referenced in the body of the text. This makes it difficult for the reader to interpret how they fit in with the author's discussion and assess confidence in their analysis and interpretation of data. This is especially apparent in the cluster annotation section of the paper.

We appreciate the feedback and suggestions provided by the reviewer, and we have revised our manuscript to make it more accessible to general audiences.

Another concern is the linking of the transcriptomic shifts associated with starvation with changes in interactions with the symbiotes. Without examining and comparing the symbiote population between the different samples, it cannot be concluded that the transcriptomic shifts correlate with a shift to the 'milking' pathway and not other environmental factors. Without comparing the symbiote abundance between samples, it is difficult to disentangle changes in cell state that are due to their changing interactions with the symbiotes from other environmental factors.

We are grateful for the valuable feedback and suggestions provided by the reviewer. Our keen interest lies in understanding symbiont responses, particularly at the single-cell level. However, it's worth noting that existing commercial single-cell RNA-seq technologies rely on oligo dT priming for reverse transcription and barcoding, thus omitting bacterial gene expression information from our dataset. We hope that advancements in technology will soon enable us to perform an integrated analysis encompassing both host and symbiont gene expression.

Additionally, conclusions in this area are further complicated by using only snRNA-seq to study intracellular processes. This is limiting since cytoplasmic mRNA is excluded and only nuclear reads are sequenced after the organisms have had several days to acclimate to their environment and major transcriptomic shifts have occurred.

We appreciate the comments shared by the reviewer and agree that scRNA-seq provides more comprehensive transcriptional information by targeting the entire mRNA of the cell. However, we would like to highlight that snRNA-seq has some unique advantages over scRNA-seq. Notably, snRNA-seq allows for simple snap-freezing of collected samples, facilitating easier storage, particularly for samples obtained during field trips involving deep-sea animals and other ecologically significant non-model animal samples. Additionally, unlike scRNA-seq, snRNA-seq eliminates the need for tissue dissociation, which often involves prolonged enzymatic treatment of deep-sea animal tissue/cells under atmospheric pressure. This process can potentially lead to the loss of sensitive cells or alterations in gene expression. Moreover, snRNA-seq procedures disregard the size and shape of animal cells, rendering it a superior technology for constructing the cell atlas of animal tissues. Consequently, we assert that snRNA-seq offers flexibility and represents a suitable choice for the research objects of our current research.

Reviewer #2 (Public Review):

Wang, He et al. shed insight into the molecular mechanisms of deep-sea chemosymbiosis at the single-cell level. They do so by producing a comprehensive cell atlas of the gill of Gigantidas platifrons, a chemosymbiotic mussel that dominates the deep-sea ecosystem. They uncover novel cell types and find that the gene expression of bacteriocytes, the symbiont-hosting cells, supports two hypotheses of host-symbiont interactions: the "farming" pathway, where symbionts are directly digested, and the "milking" pathway, where nutrients released by the symbionts are used by the host. They perform an in situ transplantation experiment in the deep sea and reveal transitional changes in gene expression that support a model where starvation stress induces bacteriocytes to "farm" their symbionts, while recovery leads to the restoration of the "farming" and "milking" pathways.

A major strength of this study includes the successful application of advanced single-nucleus techniques to a non-model, deep-sea organism that remains challenging to sample. I also applaud the authors for performing an in situ transplantation experiment in a deep-sea environment. From gene expression profiles, the authors deftly provide a rich functional description of G. platifrons cell types that is well-contextualized within the unique biology of chemosymbiosis. These findings offer significant insight into the molecular mechanisms of deep-sea host-symbiont ecology, and will serve as a valuable resource for future studies into the striking biology of G. platifrons.

The authors' conclusions are generally well-supported by their results. However, I recognize that the difficulty of obtaining deep-sea specimens may have impacted experimental design. In this area, I would appreciate more in-depth discussion of these impacts when interpreting the data.

Thank the reviewer for their valuable feedback on our study. We're grateful that the reviewers found our work interesting, and we appreciate their thorough evaluation of our research. We'll consider their constructive comments as we continue to develop and improve our study.

Because cells from multiple individuals were combined before sequencing, the in situ transplantation experiment lacks clear biological replicates. This may potentially result in technical variation (ie. batch effects) confounding biological variation, directly impacting the interpretation of observed changes between the Fanmao, Reconstitution, and Starvation conditions. It is notable that Fanmao cells were much more sparsely sampled. It appears that fewer cells were sequenced, resulting in the Starvation and Reconstitution conditions having 2-3x more cells after doublet filtering. It is not clear whether this is due to a technical factor impacting sequencing or whether these numbers are the result of the unique biology of Fanmao cells. Furthermore, from Table S19 it appears that while 98% of Fanmao cells survived doublet filtering, only ~40% and ~70% survived for the Starvation and Reconstitution conditions respectively, suggesting some kind of distinction in quality or approach.

There is a pronounced divergence in the relative proportions of cells per cell type cluster in Fanmao compared to Reconstitution and Starvation (Fig. S11). This is potentially a very interesting finding, but it is difficult to know if these differences are the expected biological outcome of the experiment or the fact that Fanmao cells are much more sparsely sampled. The study also finds notable differences in gene expression between Fanmao and the other two conditions- a key finding is that bacteriocytes had the largest Fanmao-vs-starvation distance (Fig. 6B). But it is also notable that for every cell type, one or both comparisons against Fanmao produced greater distances than comparisons between Starvation and Reconstitution (Fig. 6B). Again, it is difficult to interpret whether Fanmao's distinctiveness from the other two conditions is underlain by fascinating biology or technical batch effects. Without biological replicates, it remains challenging to disentangle the two.

As highlighted by the reviewer, our experimental design involves pooling multiple biological samples within a single treatment state before sequencing. We acknowledge the concern regarding the absence of distinct biological replicates and the potential impact of batch effects on result interpretation. While we recognize the merit of conducting multiple sequencing runs for a single treatment to provide genuine biological replicates, we contend that batch effects may not exert a strong influence on the observed patterns.

In addition, we applied a bootstrap sampling algorithm to assess whether the gene expression patterns within a cluster are more similar than those between clusters. This algorithm involves selecting a portion of cells per cluster and examining whether this subset remains distinguishable from other clusters. Our assumption was that if different samples exhibited distinct expression patterns due to batch effect, the co-assignment probabilities of a cluster would be very low. This expectation was not met in our data, as illustrated in Fig. S2. The lack of significantly low co-assignment probabilities within clusters suggests that batch effects may not exert a strong influence on our results.

Indeed, we acknowledge a noticeable shift in the expression patterns of certain cell types, such as the bacteriocyte. However, this is not universally applicable across all cell types. For instance, the UMAP figure in Fig. 6A illustrates a substantial overlap among basal membrane cell 2 from Fanmao, Starvation, and Reconstitution treatments, and the centroid distances between the three treatments are subtle, as depicted in Fig. 6B. This consistent pattern is also observed in DEPC, smooth muscle cells, and the food groove ciliary cells.

The reviewer also noted variations in the number of cells per treatment. Specifically, Fanmao sequencing yielded fewer than 10 thousand cells, whereas the other two treatments produced 2-3 times more cells after quality control (QC). It is highly probable that the technician loaded different quantities of cells into the machine for single-nucleus sequencing—a not uncommon occurrence in this methodology. While loading more cells may increase the likelihood of doublets, it is crucial to emphasize that this should not significantly impact the expression patterns post-QC. It's worth noting that overloading samples has been employed as a strategic approach to capture rare cell types, as discussed in a previous study (reference: 10.1126/science.aay0267).

The reviewer highlighted the discrepancy in cell survival rates during the 'doublet filtering' process, with 98% of Fanmao cells surviving compared to approximately 40% and 70% for the Starvation and Reconstitution conditions, respectively. It's important to clarify that the reported percentages reflect the survival of cells through a multi-step QC process employing various filtering strategies.

Post-doublet removal, we filtered out cells with <100 or >2500 genes and <100 or >6000 unique molecular identifiers (UMIs). Additionally, genes with <10 UMIs in each data matrix were excluded. The observed differences in survival rates for Starvation and Reconstitution cells can be attributed to the total volume of data generated in Illumina sequencing. Specifically, we sequenced approximately 91 GB of data for Fanmao, ~196 GB for Starvation, and ~249 GB for Reconstitution. As a result, the qualified data obtained for Starvation and Reconstitution conditions was only about twice that of Fanmao due to the limited data volume.

The reviewer also observed a divergence in the relative proportions of cells per cell type cluster in Fanmao compared to Reconstitution and Starvation, as depicted in Fig. S1. This discrepancy may hold true biological significance, presenting a potentially intriguing finding. However, our discussion on this pattern was rather brief, as we acknowledge that the observed differences could be influenced by the sample preparation process for dissection and digestion. It is crucial to consider that cutting a slightly different area during dissection may result in variations in the proportion of cells obtained. While we recognize the potential impact of this factor, we do not think that the sparsity of sampling alone could significantly affect the relative proportions of cells per cell type.

In conclusion, we acknowledge the reviewer's suggestion that sequencing multiple individual samples per treatment condition would have been ideal, rather than pooling them together. However, the homogenous distribution observed in UMAP and the consistent results obtained from bootstrap sampling suggest that the impact of batch effects on our analyses is likely not substantial. Additionally, based on our understanding, the smaller number of cells in the Fanmao sample should not have any significant effect on the resulting different proportion of cells or the expression patterns per each cluster.

Reviewer #3 (Public Review):

Wang et al. explored the unique biology of the deep-sea mussel Gigantidas platifrons to understand the fundamental principles of animal-symbiont relationships. They used single-nucleus RNA sequencing and validation and visualization of many of the important cellular and molecular players that allow these organisms to survive in the deep sea. They demonstrate that a diversity of cell types that support the structure and function of the gill including bacteriocytes, specialized epithelial cells that host sulfur-oxidizing or methane-oxidizing symbionts as well as a suite of other cell types including supportive cells, ciliary, and smooth muscle cells. By performing experiments of transplanting mussels from one habitat which is rich in methane to methane-limited environments, the authors showed that starved mussels may consume endosymbionts versus in methane-rich environments upregulated genes involved in glutamate synthesis. These data add to the growing body of literature that organisms control their endosymbionts in response to environmental change.

The conclusions of the data are well supported. The authors adapted a technique that would have been technically impossible in their field environment by preserving the tissue and then performing nuclear isolation after the fact. The use of single-nucleus sequencing opens the possibility of new cellular and molecular biology that is not possible to study in the field. Additionally, the in-situ data (both WISH and FISH) are high-quality and easy to interpret. The use of cell-type-specific markers along with a symbiont-specific probe was effective. Finally, the SEM and TEM were used convincingly for specific purposes in the case of showing the cilia that may support water movement.

We appreciate the valuable feedback provided by the reviewer on our study. It is encouraging to know that our work was found to be interesting and that they conducted a thorough evaluation of our research. We will take their constructive comments into account as we strive to develop and enhance our study. Thank the reviewer for all the input.

The one particular area for clarification and improvement surrounds the concept of a proliferative progenitor population within the gill. The authors imply that three types of proliferative cells within gills have long been known, but their study may be the first to recover molecular markers for these putative populations. The markers the authors present for gill posterior end budding zone cells (PEBZCs) and dorsal end proliferation cells (DEPCs) are not intuitively associated with cell proliferation and some additional exploration of the data could be performed to strengthen the argument that these are indeed proliferative cells. The authors do utilize a trajectory analysis tool called Slingshot which they claim may suggest that PEBZCs could be the origin of all gill epithelial cells, however, one of the assumptions of this analysis is that differentiated cells are developed from the same precursor PEBZC population.

However, these conclusions do not detract from the overall significance of the work of identifying the relationship between symbionts and bacteriocytes and how these host bacteriocytes modulate their gene expression in response to environmental change. It will be interesting to see how similar or different these data are across animal phyla. For instance, the work of symbiosis in cnidarians may converge on similar principles or there may be independent ways in which organisms have been able to solve these problems.

We are grateful for the valuable comments and suggestions provided by the reviewer. All suggestions have been carefully considered, and the manuscript has been revised accordingly. We particularly value the reviewer's insights regarding the characterization of the G. platifrons gill proliferative cell populations. In a separate research endeavor, we have conducted experiments utilizing both cell division and cell proliferation markers on these proliferative cell populations. While these results are not incorporated into the current manuscript, we would be delighted to share our preliminary findings with the reviewer. Our preliminary results indicate that the proliferative cell populations exhibit positivity for cell proliferation markers and contain a significant number of mitotic cells..

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Further experiments are needed to link the changes in transcriptomes of Bathymodioline mussels in the different environmental conditions to changes in their interactions with symbiotes. For example, quantifying the abundance and comparing the morphology of symbiotes between the environmental conditions would lend much support for shifting between milking and farming strategies. Without analyzing the symbiotes and comparing them across populations, it is difficult to comment on the mechanisms of interactions between symbiotes and the hosts. Without this analysis, this data is better suited towards comments about the general effect of environmental perturbation and stress on gene expression in these mussels.

We appreciate the reviewer’s comments. We are also very curious about the symbiont responses, especially at the single-cell level. However, all the current commercial single-cell RNA-seq technologies are based on oligo dT priming for reverse transcription and barcoding. Therefore, the bacterial gene expression information is omitted from our dataset. Hopefully, with the development of technology, we could conduct an integrated analysis of both host and symbiont gene expression soon.

Additionally, clarification is needed on which types of symbiotes are being looked at. Are they MOX or SOX populations? Are they homogenous? What are the concentrations of sulfur at the sampled sites?

We thank you for your valuable comments and suggestions. Gigantidas platifrons harbors a MOX endosymbiont population characterized by a single 16S rRNA phylotype. We apologize for any confusion resulting from our previous wording. To clarify, we have revised lines 57-59 of our introduction

In the text and images, consider using standardized gene names and leaving out the genome coordinates. This would greatly help with readability. Also, be careful to properly follow gene naming and formatting conventions (ie italicizing gene names and symbols).

We appreciate the reviewer’s insightful comments. In model animals, gene nomenclature often stems from forward genetic approaches, such as the identification of loss-of-function mutants. These gene names, along with their protein products, typically correspond to unique genome coordinates. Conversely, in non-model invertebrates (e.g., Gigantidas platifrons of present study), gene prediction relies on a combination of bioinformatics methods, including de novo prediction, homolog-based prediction, and transcriptomics mapping. Subsequently, the genes are annotated by identifying their best homologs in well-characterized databases. Given that different genes may encode proteins with similar annotated functions, we chose to include both the gene ID (genome coordinates) and the gene name in our manuscript. This dual labeling approach ensures that our audience receives accurate and comprehensive information regarding gene identification and annotation.

Additionally, extending KEGG analysis to the atlas annotation section could help strengthen the confidence of annotations. For example, when identifying bacteriocyte populations, the functional categories of individual marker genes (lysosomal proteases, lysosomal traffic regulators, etc) are used to justify the annotation. Presenting KEGG support that these functional categories are upregulated in this population relative to others would help further support how you characterize this cluster by showing it's not just a few specific genes that are enriched in this cell group, but rather an overall functionality.

We appreciate the valuable suggestion provided by the reviewer. Indeed, incorporating KEGG analysis into the atlas annotation section could further enhance the confidence in our annotations. However, in our study, we encountered some limitations that impeded us from conducting a comprehensive KEGG enrichment analysis.

Firstly, the number of differentially expressed genes (DEGs) that we identified for certain cell populations was relatively small, making it challenging to meet the threshold required for meaningful KEGG enrichment analysis. For instance, among the 97 marker genes identified for the Bacteriocyte cluster, only two genes, Bpl_scaf_59648-4.5 (lysosomal alpha-glucosidase-like) and Bpl_scaf_52809-1.6 (lysosomal-trafficking regulator-like isoform X1), were identified as lysosomal genes. To generate reliable KEGG enrichments, a larger number of genes is typically required.

Secondly, single-nucleus sequencing, as employed in our study, tends to yield a relatively smaller number of genes per cell compared to bulk RNA sequencing. This limited gene yield can make it challenging to achieve sufficient gene representation for rigorous KEGG enrichment analysis.

Furthermore, many genes in the genome still lack comprehensive annotation, both in terms of KEGG and GO annotations. In our dataset, out of the 33,584 genes obtained through single-nuclei sequencing, 26,514 genes have NO KEGG annotation, and 25,087 genes have NO GO annotation. This lack of annotations further restricts the comprehensive application of KEGG analysis in our study.

The claim that VEPCs are symbiote free is not demonstrated. Additional double in situs are needed to show that markers of this cell type localize in regions free of symbiotes.

We appreciate your comments and suggestions. In Figure 5B, our results demonstrate that the bacteriocytes (green fluorescent signal) are distant from the VEPCs, which are located around the tip of the gill filaments (close to the food groove). We have revised our Figure 5B to make it clear.

Additionally, it does not seem like trajectory analysis is appropriate for these sampling conditions. Generally, to create trajectories confidently, more closely sampled time points are needed to sufficiently parse out the changes in expression. More justification is needed for the use of this type of analysis here and a discussion of the limitations should be mentioned, especially when discussing the hypotheses relating to PEBZCs, VEPCs, and DEPCs.

We greatly appreciate your thoughtful commentary. It is important to acknowledge that in the context of a developmental study, incorporating more closely spaced time points indeed holds great value. In our ongoing project investigating mouse development, for instance, we have implemented time points at 24-hour intervals. However, in the case of deep-sea adult animals, we hypothesized a slower transcriptional shift in such extreme environment, which led us to opt for a time interval of 3-7 days. Examining the differential expression profiles among the three treatments, we observed that most cell types exhibited minimal changes in their expression profiles. For the cell types strongly impacted by in situ transplantation, their expression profiles per cell type still exhibited highly overlap in the UMAP analysis (Figure 6a), thus enabling meaningful comparisons. Nevertheless, we recognize that our sampling strategy may not be flawless. Additionally, the challenging nature of conducting in situ transplantation in 1000-meter depths limited the number of sampling occasions available to us. We sincerely appreciate your input and understanding.

Finally, more detail should be added on the computational methods used in this paper. For example, the single-cell genomics analysis protocol should be expanded on so that readers unfamiliar with BD single-cell genomics handbooks could replicate the analysis. More detail is also needed on what criteria and cutoffs were used to calculate marker genes. Also, please be careful to cite the algorithms and software packages mentioned in the text.

Acknowledged, thank you for highlighting this. In essence, the workflow closely resembles that of the 10x Genomics workflow (despite the use of a different software, i.e., Cell Ranger). We better explain the workflow below, and also noting that this information may no longer be relevant for newer users of BD or individuals who are not acquainted with BD, given that the workflow underwent a complete overhaul in the summer of 2023.

References to lines

Line 32: typo "..uncovered unknown tissue heterogeny" should read "uncovering" or "and uncovered")

Overall abstract could include more detail of findings (ex: what are the "shifts in cell state" in line 36 that were observed)

We apologize for the mistakes, and have revised the manuscript accordingly.

Line 60: missing comma "...gill filament structure, but also"

We apologize for the mistakes, and have revised the manuscript accordingly.

Line 62-63: further discussion here, or in the relevant sections of the specific genes identified in the referenced bulk RNA-seq project could help strengthen confidence in annotation

We appreciate the comment, and have revised the manuscript accordingly.

Line 112: what bootstrapping strategy? Applied to what?

This is a bootstrap sampling algorithm to assess the robustness of each cell cluster developed in a recent biorxiv paper. (Singh, P. & Zhai, Y. Deciphering Hematopoiesis at single cell level through the lens of reduced dimensions. bioRxiv, 2022.2006.2007.495099 (2022). https://doi.org:10.1101/2022.06.07.495099)

Lines 127-129: What figures demonstrate the location of the inter lamina cells? Are there in situs that show this?

We apologize for any errors; the referencing of figures in the manuscript has been revised for clarity

Lines 185-190: does literature support these as markers of SMCs? Are they known smooth muscle markers in other systems?

We characterized the SMCs by the expression of LDL-associated protein, angiotensin-converting enzyme-like protein, and the "molecular spring" titin-like protein, all of which are commonly found in human vascular smooth muscle cells. Based on this analysis, we hypothesize that these cells belong to the smooth muscle cell category.

Line 201: What is meant by "regulatory roles"?

In this context, we are discussing the expression of genes encoding regulatory proteins, such as SOX transcription factors and secreted-frizzled proteins.

Line 211: which markers disappeared? What in situs show this?

We apologize for the mistakes, and have revised the manuscript accordingly.

Line 211: typo, "role" → "roll"

We apologize for the mistakes, and have revised the manuscript accordingly.

Line 214: what are these "hallmark genes"

We apologize for the mistakes, here we are referring to the genes listed in figure 4B. We have revised the manuscript accordingly.

Line 220: are there meristem-like cells in metazoans? If so, this would be preferable to a comparison with plants.

In this context, we are discussing the morphological characteristics of gill proliferative cell populations found in filibranch bivalves. These populations, namely PEPC, VEPC, and DEPC, consist of cells exhibiting morphological traits akin to those of plant cambial-zone meristem cells. These cells typically display small, round shapes with a high nucleus-to-plasma ratio. We acknowledge that while these terms are utilized in bivalve studies (citations below), they lack the robust support seen in model systems backed by molecular biology evidences. The present snRNA-seq data, however, may offer valuable cell markers for future comprehensive investigations.

Leibson, N. L. & Movchan, O. T. Cambial zones in gills of Bivalvia. Mar. Biol. 31, 175-180 (1975). https://doi.org:10.1007/BF00391629

Wentrup, C., Wendeberg, A., Schimak, M., Borowski, C. & Dubilier, N. Forever competent: deep-sea bivalves are colonized by their chemosynthetic symbionts throughout their lifetime. Environ. Microbiol. 16, 3699-3713 (2014). https://doi.org:10.1111/1462-2920.12597

Cannuel, R., Beninger, P. G., McCombie, H. & Boudry, P. Gill Development and its functional and evolutionary implications in the blue mussel Mytilus edulis (Bivalvia: Mytilidae). Biol. Bull. 217, 173-188 (2009). https://doi.org:10.1086/BBLv217n2p173

Line 335: what is slingshot trajectory analysis? Does this differ from the pseudotime analysis?

Slingshot is an algorithm that uses the principal graph of the cells to infer trajectories. It models trajectories as curves on the principal graph, capturing the progression and transitions between different cellular states.

Both Slingshot and pseudotime aim to infer cellular trajectories. Slingshot focuses on capturing branching patterns which is fully compatible with the graph generated using dimensionality reduction such as UMAP and PHATE, while pseudotime analysis aims to order cells along a continuous trajectory. It does not rely on dimensionality reduction graphs. We used both in the MS for different purposes.

Line 241: introduce FISH methodology earlier in the paper, when in situ images are first referenced

We appreciate the comment, and have revised the manuscript accordingly.

Line 246-249: can you quantify the decrease in signal or calculate the concentration of symbiotes in the cells? Was 5C imaged whole? This can impact the fluorescent intensity in tissues of different thicknesses.

We appreciate your comment. In Figure 5C, most of the typical gill filament region is visible (the ventral tip of the gill filament, and the mid part of the gill filament) except for the dorsal end. The gill filament of bathymodioline mussels exhibits a simple structure: a single layer of bacteriocytes grow on the basal membrane. Consequently, the gill slices have a fairly uniform thickness (with two layers of bacteriocytes and one layer of interlamina cells in between), minimizing any potential impact on fluorescent intensity. As of now, detailed quantification of intracellular symbionts may necessitate continuous TEM or ultra-resolution confocal sections to 3D reconstruct the bacteriocytes, which may exceed the scope of the current study. Therefore, fluorescent intensity remains the only method available to us for estimating bacterial density/distribution across the gill filament.

Line 249: What is meant by 'environmental gradient?'

Here we are refereeing the gases need for symbiont’s chemosynthesis. We have revised the manuscript to make it clear.

Lines 255-256: Were the results shown in the TEM images previously known? Not clear what novel information is conveyed in images Fig 5 C and D

In the Fig 5 C and D, we’ve delivered a high-quality SEM TEM image of a typical bacteriocyte, showcasing its morphology and subcellular machinery with clarity. These electron microscopy images offer the audience a comprehensive introduction to the cellular function of bacteriocytes. Additionally, they serve as supportive evidence for the bacteriocytes' snRNA-seq data.

Line 295-296: Can you elaborate on what types of solute carrier genes have been shown to be involved with symbioses?

We appreciate the comment, and have revised the manuscript accordingly. The putative functions of the solute carriers could be found in Figure 5I.

Line 297-301: Which genes from the bulk RNA-seq study? Adding more detail and references in cluster annotation would help readers better understand the justifications.

We appreciate the comment, and have revised the manuscript accordingly.

Line 316 -322: Can you provide the values of the distances?

We also provide values in the main text, in addition to the Fig6b. We also provide a supplementary Table (Supplementary Table S19).

Line 328: What are the gene expression patterns?

We observed genes that are up- and down-regulated in Starvation and reconstitution.

LIne 334-337: A visualization of the different expression levels of the specific genes in clusters between sites might be helpful to demonstrate the degree of difference between sites.

We have prepared a new supplementary file showing the different expression levels.

Line 337: Citation needed

We appreciate the comment. Here, we hypothesize the cellular responds based on the gene’s function and their expression patterns.

Line 402-403: Cannot determine lineages from data presented. Need lineage tracing over time to determine this

We acknowledge the necessity of conducting lineage tracing over time to validate this hypothesis. Nonetheless, in practical terms, it is difficult to obtain samples for testing this. Perhaps, it is easier to use their shallow sea relatives to test this hypothesis. However, in practice, it is very difficult.

413-414: What are the "cell-type specific responses to environmental change"? It could be interesting to present these results in the "results and discussion" section

These results are shown in Supplementary Figure S8.

Line 419-424: Sampling details might go better earlier on in the paper, when the sampling scheme is introduced.

We appreciate the comments. Here, we are discussing the limitations of our current study, not sampling details.

Line 552: What type of sequencing? Paired end? How long?

We conducted 150bp paired-end sequencing.

556-563: More detail here would be useful to readers not familiar with the BD guide. Also be careful to cite the software used in analysis!

The provided guide and handbook elucidate the intricacies of gene name preparation, data alignment to the genome, and the generation of an expression matrix. It is worth mentioning that we relied upon outdated versions of the aforementioned resources during our data analysis phase, as they were the only ones accessible to us at the time. However, we have since become aware of a newer pipeline available this year, rendering the information presented here of limited significance to other researchers utilizing BD.

Many thanks for your kind reminding. We have now included a reference for STAR. All other software was cited accordingly. There are no scholarly papers or publications to refer to for the BD pipeline that we can cite.

Line 577-578: How was the number of clusters determined? What is meant by "manually combine the clusters?" If cells were clustered by hand, more detail on the method is needed, as well as direct discussion and justification in the body of the paper.

It would be more appropriate to emphasize the determination of cell types rather than clusters. The clusters were identified using a clustering function, as mentioned in the manuscript. It's important to note that the clustering function (in our case, the FindClusters function of Seurat) provides a general overview based on diffuse gene expression. Technically speaking, there is no guarantee that one cluster corresponds to a single cell type. Therefore, it is crucial to manually inspect the clustering results to assign clusters to the appropriate cell types. In some cases, multiple clusters may be assigned to the same cell type, while in other cases, a single cluster may need to be further subdivided into two or more cell types or sub-cell types, depending on the specific circumstances.

For studies conducted on model species such as humans or mice, highly and specifically expressed genes within each cluster can be compared to known marker genes of cell types mentioned in previous publications, which generally suffices for annotation purposes. However, in the case of non-model species like Bathymodioline mussels, there is often limited information available about marker genes, making it challenging to confidently assign clusters to specific cell types. In such situations, in situ hybridisation proves to be incredibly valuable. In our study, WISH was employed to visualise the expression and morphology of marker genes within clusters. When WISH revealed the expression of marker genes from a cluster in a specific type of cell, we classified that cluster as a genuine cell type. Moreover, if WISH demonstrated uniform expression of marker genes from different clusters in the same cell, we assigned both clusters to the same cell type.

We expanded the description of the strategy in the Method section.

LIne 690-692: When slices were used, what part of the gill were they taken from?

We sectioned the gill around the mid part which could represent the mature bacteriocytes.

References to figures:

General

Please split the fluorescent images into different channels with an additional composite. It is difficult to see some of the expression patterns. It would also make it accessible to colorblind readers.

We appreciate the comments and suggestions from the reviewer. We have converted our figures to CMYK colour which will help the colorblind audiences to read our paper.

Please provide the number of replicates for each in situ and what proportion of those displayed the presented pattern.

We appreciate the reviewer’s comments. We have explained in the material and methods part of the manuscript.

Figure 2.C' is a fantastic summary and really helps the non-mussel audience understand the results. Adding schematics like this to Figures 3-5 would be helpful as well.

We value the reviewer's comments. We propose that Figures 3K, 4C, and 5A-D could offer similar schematic explanations to assist the audience.

Figure 2:

Figures 2.C-F, 2.C', 2.H-J are not referenced in the text. Adding in discussions of them would help strengthen your discussions on the cluster annotation

We appreciate the reviewer's comments. We have revise the manuscript accordingly.

In 2.B. 6 genes are highlighted in red and said to be shown in in situs, but only 5 are shown.

We apology for the mistake. We didn’t include the result 20639-0.0 WISH in present study. We have changed the label to black.

Figure 3:

FIg 2C-E not mentioned.

We appreciate the reviewer's comments. We have revise the manuscript accordingly.

In 3.B 8 genes are highlighted in red and said to be shown in in situs. Only 6 are.

The result of the WISH were provided in Supplementary Figures S4 and S5.

FIgure 3.K is not referenced in the legend.

We appreciate the comment, and have revised the manuscript accordingly.

Figure 4:

In Figure D, it might be helpful to indicate the growth direction.

We appreciate the comment, and have revised the manuscript accordingly by adding an arrow in panel D to indicate growth direction.

4F: A double in situ with the symbiote marker is needed to demonstrate the nucleolin-like positive cells are symbiote free.

We appreciate the comment. The symbiont free region could be found in Figure 5A.

Figure 5:

In 5.A, quantification of symbiote concentration would help support your conclusion that they are denser around the edges.

We appreciate the comment, as we mentioned above, detailed quantification of intracellular symbionts may necessitate continuous TEM or ultra-resolution confocal sections to 3D reconstruct the bacteriocytes, which may exceed the scope of the current study. Therefore, fluorescent intensity remains the only method available to us for estimating bacterial density/distribution across the gill filament.

In 5.D, the annotation is not clear. Adding arrows like in 5.C would be helpful.

We appreciate the comment, and have revised the manuscript accordingly.

A few genes in 5.F are not mentioned in the paper body when listing other genes. Mentioning them would help provide more support for your clustering.

We appreciate the comment, and have revised the manuscript accordingly.

Is 5.I meant to be color coded with the gene groups from 5.F? Color Coding the gene names, rather than organelles or cellular structures might portray this better and help visually strengthen the link between the diagram and your dot plot.

We appreciate the suggestions. We've experimented with color-coding the gene names, but some colors are less discernible against a white background.

Figure 6:

6.B Is there a better way to visualize this data? The color coding is confusing given the pairwise distances. Maybe heatmaps?

We attempted a heatmap, as shown in the figure below. However, all co-authors agree that a bar plot provides clearer visualization compared to the heatmap. We agree that the color scheme maya be confusing because they use the same color as for individual treatment. So we change the colors.

Author response image 1.

Figure 6.D: Why is the fanmao sample divided in the middle?

Fig6C show that single-cell trajectories include branches. The branches occur because cells execute alternative gene expression programs. Thus, in Fig 6D, we show changes for genes that are significantly branch dependent in both lineages at the same time. Specifically, in cluster 2, the genes are upregulated during starvation but downregulated during reconstitution. Conversely, genes in cluster 1 are downregulated during starvation but upregulated during reconstitution. It's of note that Fig 6D displays only a small subset of significantly branch-dependent genes.

FIgure 6.D: Can you visualize the expression in the same format as in figures 2-5?

We appreciate the comments from the reviewer. As far as we know, this heatmap are the best format to demonstrate this type of gene expression profile.

Supplementary Figure S2:

Please provide a key for the cell type abbreviations

We appreciate the comment, and have added the abbreviations of cell types accordingly.

Supplementary Figures S4 and S5:

What part of the larger images are the subsetted image taken from?

We appreciate the comment, these images were taken from the ventral tip and mid of the gill slices, respectively. We have revised the figure legends to make it clear.

Supplemental Figure S7:

If clusters 1 and 2 show genes up and downregulated during starvation, what do clusters 4 and 3 represent?

Cluster 1: Genes that are obviously upregulated during Starvation, and downregulated during reconstitution; luster4： genes are downregulated during reconstitution but not obviously upregulated during Starvation.

Cluster 2 show genes upregulated during reconstitution, and cluster 3 obviously downregulated during Starvation.

Author response table 1.

Supplemental Figure S8:

This is a really interesting figure that I think shows some of the results really well! Maybe consider moving it to the main figures of the paper?

We appreciate the comments and suggestions. We concur with the reviewer on the significance of the results presented. However, consider the length of this manuscript, we have prioritized the inclusion of the most pertinent information in the main figures. Supplementary materials containing additional figures and details on the genes involved in these pathways are provided for interested readers.

Supplemental Figure S11:

Switching the axes might make this image easier for the reader to interpret. Additionally, calculating the normalized contribution of each sample to each cluster could help quantify the extent to which bacteriocytes are reduced when starving.

Thank you for the insightful suggestion, which we have implemented as detailed below. We acknowledge the importance of understanding the changes in bacteriocyte proportions across different treatments. However, it's crucial to note that the percentage of cells per treatment is highly influenced by factors such as the location of digestion and sequencing, as previously mentioned.

Author response image 2.

Reviewer #2 (Recommendations For The Authors):

The following are minor recommendations for the text and figures that may help with clarity:

Fig. 3K: This figure describes water flow induced by different ciliary cells. It is not clear what the color of the arrows corresponds to, as they do not match the UMAP (i.e. the red arrow) and this is not indicated in the legend. Are these colours meant to indicate the different ciliary cell types? If so it would be helpful to include this in the legend.

We appreciate the reviewer's comments and suggestions. The arrows indicate the water flow that might be agitated by the certain types of cilium. We have revised our figure and figure legends to make it clear.

Line 369: The incorrect gene identifier is given for the mitochondrial trifunctional enzyme. This gene identifier is identical to the one given in line 366, which describes long-chain-fatty-acid-ligase ACSBG2-like (Bpl_scaf_28862-1.5).

We appreciate the reviewer's comments and suggestions. We have revised our manuscript accordingly.

Line 554: The Bioproject accession number (PRJNA779258) does not appear to lead to an existing page in any database.

We appreciate the reviewer's comments and suggestions. We have released this Bioproject to the public.

Line 597-598: it would be helpful to know the specific number of cells that the three sample types were downsampled to, and the number of cells remaining in each cluster, as this can affect the statistical interpretation of differential expression analyses.

The number of cells per cluster in our analysis ranged from 766 to 14633. To mitigate potential bias introduced by varying cell numbers, we implemented downsampling, restricting the number of cells per cluster to no more than 3500. This was done to ensure that the differences between clusters remained less than 5 times. We experimented with several downsampling strategies, exploring cell limits of 4500 and 2500, and consistently observed similar patterns across these variations.

Data and code availability:

The supplementary tables and supplementary data S1 appear to be the final output of the differential expression analyses. Including the raw data (e.g. reads) and/or intermediate data objects (e.g. count matrices, R objects), in addition to the code used to perform the analyses, may be very helpful for replication and downstream use of this dataset. As mentioned above, the Bioproject accession number appears to be incorrect.

We appreciate the reviewer's comments and suggestions. Regarding our sequencing data, we have deposited all relevant information with the National Center for Biotechnology Information (NCBI) under Bioproject PRJNA779258. Additionally, we have requested the release of the Bioproject. Furthermore, as part of this round of revision, we have included the count matrices for reference.

Reviewer #3 (Recommendations For The Authors):

As noted in the public review, my only major concerns are around the treatment of progenitor cell populations. I am sympathetic to the challenges of these experiments but suggest a few possible avenues to the authors.

First, there could be some demonstration that these cells in G. platifrons are indeed proliferative, using EdU incorporation labeling or a conserved epitope such as the phosphorylation of serine 10 in histone 3. It appears in Mytilus galloprovincialis that proliferating cell nuclear antigen (PCNA) and phospho-histone H3 have previously been used as good markers for proliferative cells (Maiorova and Odintsova 2016). The use of any of these markers along with the cell type markers the authors recover for PEBZCs for example would greatly strengthen the argument that these are proliferative cells.

If performing these experiments would not be currently possible, the authors could use some computation approaches to strengthen their arguments. Based on conserved cell cycle markers and the use of Cell-Cycle feature analysis in Seurat could the authors provide evidence that these progenitors occupy the G2/M phase at a greater percentage than other cells? Other than the physical position of the cells is there much that suggests that these are proliferative? While I am more convinced by markers in VEPCs the markers for PEBZCs and DEPCs are not particularly compelling.

While I do not think the major findings of the paper hinge on this, comments such as "the PBEZCs gave rise to new bacteriocytes that allowed symbiont colonization" should be taken with care. It is not clear that the PBEZCs are proliferative and there does not seem to be any direct evidence that PBEZCs (or DEPCs or VEPCS for that manner) are the progenitor cells through any sort of labeling or co-expression studies.

We appreciate the comments and suggestions from the reviewer. We have considered all the suggestions and have revised the manuscript accordingly. We especially appreciate the reviewer’s suggestions about the characterisations of the G. platifrons gill proliferative cell populations. In a separate research project, we have tested both cell division and cell proliferation markers on the proliferation cell populations. Though we are not able to include these results in the current manuscript, we are happy to share our preliminary results with the reviewer. Our results demonstrate the proliferative cell populations, particularly the VEPCs, are cell proliferation marker positive, and contains high amount of mitotic cells.

Author response image 3.

Finally, there is a body of literature that has examined cell proliferation and zones of proliferation in mussels (such as Piquet, B., Lallier, F.H., André, C. et al. Regionalized cell proliferation in the symbiont-bearing gill of the hydrothermal vent mussel Bathymodiolus azoricus. Symbiosis 2020) or other organisms (such as Bird, A. M., von Dassow, G., & Maslakova, S. A. How the pilidium larva grows. EvoDevo. 2014) that could be discussed.

We appreciate the comments and suggestions from the reviewer. We have considered all the suggestions and have revised the manuscript accordingly (line 226-229).

Minor comments also include:

Consider changing the orientation of diagrams in Figure 2C' in relationship to Figure 2C and 2D-K.

We appreciate the comments and suggestions from the reviewer. The Figure 2 has been reorganized.

For the diagram in Figure 3K, please clarify if the arrows drawn for the direction of inter lamina water flow is based on gene expression, SEM, or some previous study.

We are grateful for the reviewer's valuable feedback and suggestions. The arrows in the figure indicate the direction of water flow that could be affected by specific types of cilium. Our prediction is based on both gene expression and SEM results. To further clarify this point, we have revised the figure legend of Fig. 3.

Please include a label for the clusters in Figure 5E for consistency.

We have revised our Figure 5E to keep our figures consistent.

Please include a note in the Materials and Methods for Monocle analysis in Figure 6.

We conducted Monocle analyses using Monocle2 and Monocle 3 in R environment. We have revised our material and methods with further information of Figure 6.

In Supplement 2, the first column is labeled PEBC while the first row is labeled PEBZ versus all other rows and columns have corresponding names. I am guessing this is a typo and not different clusters?

We appreciate the great effort of the reviewer in reviewing our manuscript. We have corrected the typo in the revised version.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

The authors aim to consider the effects of phonotactics on the effectiveness of memory reactivation during sleep. They have created artificial words that are either typical or atypical and showed that reactivation improves memory for the latter but not the former.

Comment 1:

Strengths:

This is an interesting design and a creative way of manipulating memory strength and typicality. In addition, the spectral analysis on both the wakefulness data and the sleep data is well done. The article is clearly written and provides a relevant and comprehensive of the literature and of how the results contribute to it.

We thank the reviewer for his/her positive evaluation of our manuscript. 

Comment 2:

Weaknesses:

(1) Unlike most research involving artificial language or language in general, the task engaged in this manuscript did not require (or test) learning of meaning or translation. Instead, the artificial words were arbitrarily categorised and memory was tested for that categorisation. This somewhat limits the interpretation of the results as they pertain to language science, and qualifies comparisons with other language-related sleep studies that the manuscript builds on.

We thank the reviewer for this comment. We agree that we did not test for meaning or translation but used a categorization task in which we trained subjects to discriminate artificial words according to their reward associations (rewarded vs. non-rewarded). Previous language studies (Batterink et al., 2014; Batterink and Paller, 2017; Reber, 1967) used artificial words to investigate implicit learning of hidden grammar rules. Here, the language researchers studied generalization of the previously learned grammar knowledge by testing subject’s ability to categorize correctly a novel set of artificial words into rule-congruent versus rule-incongruent words. These differences to our study design might limit the comparability between the results of previous language studies of artificial grammar learning and our findings. We discussed now this aspect as a limitation of our novel paradigm. 

We added the following sentences to the discussion on p.14, ll. 481-488:

Based on our paradigm, we investigated categorization learning of artificial words according to their reward associations (rewarded vs. unrewarded) and did not studied aspects of generalization learning of artificial grammar rules (Batterink et al., 2014; Batterink and Paller, 2017; Reber, 1967). This difference might limit the comparability between these previous language-related studies and our findings. However, the usage of artificial words with distinct phonotactical properties provided a successful way to manipulate learning difficulty and to investigate word properties on TMR, whereas our reward categorization learning paradigm had the advantage to increase the relevance of the word learnings due to incentives.    

Comment 3:

(2) The details of the behavioural task are hard to understand as described in the manuscript. Specifically, I wasn't able to understand when words were to be responded to with the left or right button. What were the instructions? Were half of the words randomly paired with left and half with right and then half of each rewarded and half unrewarded? Or was the task to know if a word was rewarded or not and right/left responses reflected the participants' guesses as to the reward (yes/no)? Please explain this fully in the methods, but also briefly in the caption to Figure 1 (e.g., panel C) and in the Results section.

We thank the reviewer for this comment and added additional sentences into the document to provide additional explanations. We instructed the participants to respond to each word by left- and right-hand button presses, whereas one button means the word is rewarded and the other button means the word is unrewarded. The assignment of left- and right-hand button presses to their meanings (rewarded versus unrewarded) differed across subjects. In the beginning, they had to guess. Then over trial repetitions with feedback at the end of each trial, they learned to respond correctly according to the rewarded/unrewarded associations of the words.        

We added the following sentences to the results section on p.5, ll. 161-168: 

As a two alternative forced-choice task, we assigned left- and right-hand button presses to the rewarded and the unrewarded word category, counterbalanced across subjects. We instructed the participants to respond to each word by left- or right-hand button presses, whereas one button means the word is rewarded (gain of money points) and the other button means the word is unrewarded (avoid the loss of money points). In the beginning, they had to guess. By three presentations of each word in randomized order and by feedback at the end of each trial, they learned to respond correctly according to the rewarded/unrewarded associations of the words (Fig. 1c). 

We added the following sentences to the caption of Figure 1 on p.6, ll. 188-194:

As a two alternative forced-choice task, responses of left- and right-hand button presses were assigned to the rewarded and the unrewarded word category, respectively. The participants were instructed to respond to each word by left- or right-hand button presses, whereas one button means the word is rewarded (gain of money points) and the other button means the word is unrewarded (avoid the loss of money points). d) Feedback matrix with the four answer types (hits: rewarded and correct; CR, correct rejections: unrewarded and correct; misses: rewarded and incorrect; FA, false alarms: unrewarded and incorrect) regarding to response and reward assignment of the word.

We added the following sentences to the methods on p.19, ll. 687-692:  

As a two alternative forced-choice task, we assigned left- and right-hand button presses to the rewarded and the unrewarded word category, counterbalanced across subjects. We instructed the participants to respond to each word by left- or right-hand button presses, whereas one button means the word is rewarded (gain of money points) and the other button means the word is unrewarded (avoid the loss of money points).

Comment 4:  

(3) Relatedly, it is unclear how reward or lack thereof would translate cleanly into a categorisation of hits/misses/correct rejections/false alarms, as explained in the text and shown in Figure 1D. If the item was of the non-rewarded class and the participant got it correct, they avoided loss. Why would that be considered a correct rejection, as the text suggests? It is no less of a hit than the rewarded-correct, it's just the trial was set up in a way that limits gains. This seems to mix together signal detection nomenclature (in which reward is uniform and there are two options, one of which is correct and one isn't) and loss-aversion types of studies (in which reward is different for two types of stimuli, but for each type you can have H/M/CR/FA separably). Again, it might all stem from me not understanding the task, but at the very least this required extended explanations. Once the authors address this, they should also update Fig 1D. This complexity makes the results relatively hard to interpret and the merit of the manuscript hard to access. Unless there are strong hypotheses about reward's impact on memory (which, as far as I can see, are not at the core of the paper), there should be no difference in the manner in which the currently labelled "hits" and "CR" are deemed - both are correct memories. Treating them differently may have implications on the d', which is the main memory measure in the paper, and possibly on measures of decision bias that are used as well.

We thank the reviewer for this comment giving us the opportunity to clarify. As explained in the previous comment, for our two alternative forced-choice task, we instructed the participants to press one button when they were thinking the presented word is rewarded and the other button, when they were thinking the word is unrewarded. Based on this instruction, we applied the signal detection theory (SDT), because the subjects had the task to detect when reward was present or to reject when reward was absent. Therefore, we considered correct responses of words of the rewarded category as hits and words of the unrewarded category as correct rejections (see Table below). However, the reviewer is correct because in addition to false alarms, we punished here the incorrect responses by subtraction of money points to control for alternative task strategies of the participants instead of reward association learning of words. We agree that further explanation/argumentation to introduce our nomenclature is necessary.  

Author response table 1.

We adjusted the results section on p.5, ll. 169-177:

To obtain a measurement of discrimination memory with respect to the potential influence of the response bias, we applied the signal detection theory (Green and Swets, 1966). Because, we instructed the participants to respond to each word by left- or right-hand button presses and that one button means reward is present whereas the other button means reward is absent, we considered correct responses of words of the rewarded category as hits and words of the unrewarded category as correct rejections. Accordingly, we assigned the responses with regard to the reward associations of the words to the following four response types: hits (rewarded, correct); correct rejections (unrewarded, correct); misses (rewarded, incorrect); and false alarms (unrewarded, incorrect). Dependent on responses, subjects received money points (Fig. 1d). 

Comment 5:

(4) The study starts off with a sample size of N=39 but excludes 17 participants for some crucial analyses. This is a high number, and it's not entirely clear from the text whether exclusion criteria were pre-registered or decided upon before looking at the data. Having said that, some criteria seem very reasonable (e.g., excluding participants who were not fully exposed to words during sleep). It would still be helpful to see that the trend remains when including all participants who had sufficient exposure during sleep. Also, please carefully mention for each analysis what the N was.

Our study was not pre-registered. Including all the subjects independent of low prememory performance, but with respect to a decent number of reactivations (> 160 reactivations, every word at least 2 times), resulted in a new dataset with 15 and 13 participants of the high- and low-PP cueing condition, respectively. Here, statistical analyses revealed no significant overnight change anymore in memory performance in the high-PP cueing condition (Δ memory (d'): t(14) = 1.67, p = 0.12), whereas the increase of the bias in decision making towards risk avoidance still remained significant (Δ bias (c-criterion): t(14) = 3.36, p = 0.005).

We modified and added the following sentences to the discussion on p.13, ll. 456-458:

Our study has limitations due to a small sample size and between-subject comparisons. The criteria of data analyses were not pre-registered and the p-values of our behavior analyses were not corrected for multiple comparisons.

Comment 6:             

(5) Relatedly, the final N is low for a between-subjects study (N=11 per group). This is adequately mentioned as a limitation, but since it does qualify the results, it seemed important to mention it in the public review.

We agree with the reviewer that the small sample size and the between subject comparisons represent major limitations of our study. Accordingly, we now discussed these limitations in more detail by adding alternative explanations and further suggestions for future research to overcome these limitations.        

We added the following sentences to the discussion about the limitations on p.14, ll. 465-488: 

To control for potential confounders despite the influence of difficulty in word learning on TMR, we compared parameters of sleep, the pre-sleep memory performance and the vigilance shortly before the post-sleep memory test, revealing no significant group differences (see Table S1 and S2). Nevertheless, we cannot rule out that other individual trait factors differed between the groups, such as the individual susceptibility to TMR. To rule out these alternative explanations based on individual factors, we suggest for future research to replicate our study by conducting a within-subject design with cueing of subsets of previously learned low- and high-PP words providing all conditions within the same individuals as shown in other TMR studies (Cairney et al., 2018; Schreiner and Rasch, 2015).

Comment 7:

(6) The linguistic statistics used for establishing the artificial words are all based on American English, and are therefore in misalignment with the spoken language of the participants (which was German). The authors should address this limitation and discuss possible differences between the languages. Also, if the authors checked whether participants were fluent in English they should report these results and possibly consider them in their analyses. In all fairness, the behavioural effects presented in Figure 2A are convincing, providing a valuable manipulation test.

We thank the reviewer pointing to the misalignment between the German-speaking participants and the used artificial words based on American English. Further, we did not assessed the English language capability of the participants to control it as a potential confounder, whereas comparative control analyses revealed no significant differences between the both cueing groups in pre-sleep memory performance (see Table S1). 

We now discussed these comments as limitations on p.14, ll. 473-481: 

Further, we used artificial words based on American English in combination with German speaking participants, whereas language differences of pronunciation and phoneme structures might affect word perception and memory processing (Bohn and Best, 2012). On the other hand, both languages are considered to have the same language family (Eberhard et al., 2019) and the phonological distance between English and German is quite short compared for example to Korean (Luef and Resnik, 2023). Thus, major common phonological characteristics across both languages are still preserved. In addition, our behavior analyses revealed robust word discrimination learning and distinct memory performance according to different levels of phonotactic probabilities providing evidence of successful experimental manipulation. 

Comment 8:

(7) With regard to the higher probability of nested spindles for the high- vs low-PP cueing conditions, the authors should try and explore whether what the results show is a general increase for spindles altogether (as has been reported in the past to be correlated with TMR benefit and sleep more generally) or a specific increase in nested spindles (with no significant change in the absolute numbers of post-cue spindles). In both cases, the results would be interesting, but differentiating the two is necessary in order to make the claim that nesting is what increased rather than spindle density altogether, regardless of the SW phase.

We conducted additional analyses based on detected sleep spindles to provide additional data according to this question. 

We added the following section to the supplementary data on pp. 31-32, ll. 1007-1045:  

After conducting a sleep spindle detection (frequency range of 12-16Hz, see methods for details), we compared the sleep spindle density between the TMR conditions of high- and lowPP showing no significant difference (see Fig. S8a and Table S9). Next, we subdivided the detected sleep spindles into coupled and uncoupled sleep spindles with the previously detected slow waves (SW; analyses of Fig. 4). Sleep spindles were defined as coupled when their amplitude peak occurred during the SW up-state phase (0.3 to 0.8s time-locked to the SW troughs). A two-way mixed design ANOVA on the amplitude size of the sleep spindles with the cueing group as a between-subject factor (high-PP-cued vs. low-PP-cued) and SW-coupling as a within-subject factor (coupled vs. uncoupled) showed a significant interaction effect (cueing group × SW-coupling: F(1,20) = 4.51, p = 0.046, η2 = 0.18), a significant main effect of SW-coupling (F(1,20) = 85.02, p < 0.001, η2 = 0.81), and a trend of significance of the main effect of the cueing group (F(1,20) = 3.54, p = 0.08). Post-hoc unpaired t-tests revealed a significant higher amplitude size of the coupled sleep spindles of the cueing group of high- compared to low-PP (t(20) = 2.13, p = 0.046, Cohen’s d = 0.91; Fig. S8b) and no significant group difference of the uncoupled sleep spindles (t(20) = 1.62, p = 0.12). An additional comparison of the amount of coupled sleep spindles between the cueing groups revealed no significant difference (see Table S9). 

Here, we found that detected sleep spindles coupled to the SW up-state phase occurred with higher amplitude after TMR presentations of the high-PP words in comparison to the low-PP words, whereas the sleep spindle density and the amount of sleep spindles coupled to the SW up-state phase did not differed between the cueing conditions.     

We added the following sentences to the methods on pp. 22-23, ll. 822-839:  

Sleep spindle analyses 

We detected fast sleep spindles by band-pass filtering (12-16Hz) the signal of the Pz electrode during the auditory cueing trials in the time windows of -2 to 8s according to stimulus onsets. The amplitude threshold was calculated individually for each subject as 1.25 standard deviations (SDs) from the mean. The beginning and end times of the sleep spindles were then defined as the points at which the amplitude fell below 0.75 SDs before and after the detected sleep spindle. Only sleep spindles with a duration of 0.5-3 s were included in subsequent analyses. 

To compare the sleep spindle densities between the different cueing conditions of high- and low-PP, we computed the grand average sleep spindle density distribution in number per trial with a bin size of 0.5s from -0.5 to 6s time-locked to stimulus onset in each condition (see Fig. S8a and Table S9).     

Based on the detected slow waves and sleep spindles, we defined coupling events when the positive amplitude peak of a detected sleep spindle was occurring during the slow wave upstate phase in a time window of 0.3 to 0.8s according to the trough of a slow wave. 

We computed the averaged amplitude size of each detected sleep spindle by calculating the mean of the absolute amplitude values of all negative and positive peaks within a detected sleep spindle (see Fig. S8b).

We added the following sentences to the results on p.10, ll. 338-343:  

By conducting an additional analyses based on detection of fast sleep spindles (12-16Hz; see methods), we confirmed that fast sleep spindles during the SW up-states (from 0.3 to 0.8s after the SW trough) occurred with significantly higher amplitude after the cueing presentation of high- compared to low-PP words, whereas parameters of sleep spindle density and the amount sleep spindles coupled to the SW up-state did not differed between the cueing conditions (see Fig. S8 and Table S9).       

Reviewer #2 (Public Review):

Summary:

The work by Klaassen & Rasch investigates the influence of word learning difficulty on sleepassociated consolidation and reactivation. They elicited reactivation during sleep by applying targeted memory reactivation (TMR) and manipulated word learning difficulty by creating words more similar (easy) or more dissimilar (difficult) to our language. In one group of participants, they applied TMR of easy words and in another group of participants, they applied TMR of difficult words (between-subjects design). They showed that TMR leads to higher memory benefits in the easy compared to the difficult word group. On a neural level, they showed an increase in spindle power (in the up-state of an evoked response) when easy words were presented during sleep.

Comment 9:

Strengths:

The authors investigate a research question relevant to the field, that is, which experiences are actually consolidated during sleep. To address this question, they developed an innovative task and manipulated difficulty in an elegant way.

Overall, the paper is clearly structured, and results and methods are described in an understandable way. The analysis approach is solid.

We thank the reviewer for his/her positive evaluation of our manuscript.

Weaknesses:

Comment 10:

(1) Sample size

For a between-subjects design, the sample size is too small (N = 22). The main finding (also found in the title "Difficulty in artificial word learning impacts targeted memory reactivation") is based on an independent samples t-test with 11 participants/group.

The authors explicitly mention the small sample size and the between-subjects design as a limitation in their discussion. Nevertheless, making meaningful inferences based on studies with such a small sample size is difficult, if not impossible.

We agree with the reviewer that the small sample size and the between subject comparisons represent major limitations of our study. Accordingly, we now discussed these limitations in more detail by adding alternative explanations and further suggestions for future research to overcome these limitations.        

We added the following sentences to the discussion about the limitations on p.14, ll. 465-473: 

To control for potential confounders despite the influence of difficulty in word learning on TMR, we compared parameters of sleep, the pre-sleep memory performance and the vigilance shortly before the post-sleep memory test, revealing no significant group differences (see Table

S1 and S2). Nevertheless, we cannot rule out that other individual trait factors differed between the groups, such as the individual susceptibility to TMR. To rule out these alternative explanations based on individual factors, we suggest for future research to replicate our study by conducting a within-subject design with cueing of subsets of previously learned low- and high-PP words providing all conditions within the same individuals as shown in other TMR studies (Cairney et al., 2018; Schreiner and Rasch, 2015).

Comment 11:

(2) Choice of task

though the task itself is innovative, there would have been tasks better suited to address the research question. The main disadvantage the task and the operationalisation of memory performance (d') have is that single-trial performance cannot be calculated. Consequently, choosing individual items for TMR is not possible.

Additionally, TMR of low vs. high difficulty is conducted between subjects (and independently of pre-sleep memory performance) which is a consequence of the task design.

The motivation for why this task has been used is missing in the paper.

We used a reward task combined with TMR because previous studies revealed beneficial effects of reward related information on sleep dependent memory consolidation and reactivation (Asfestani et al., 2020; Fischer and Born, 2009; Lansink et al., 2009; Sterpenich et al., 2021). In addition, we wanted to increase the motivation of the participants, as they could receive additional monetary compensation according to their learning and memory task performances. Furthermore, we designed the task, with the overall possibility to translate this task to operant conditioning in rats (see research proposal: https://data.snf.ch/grants/grant/168602). However, the task turned out to be too difficult to translate to rats, whereas we developed a different learning paradigm for the animal study (Klaassen et al., 2021) of this cross-species research project.       

We added the following sentence to the introduction on p.4, ll. 134-137:

To consider the beneficial effect of reward related information on sleep dependent memory consolidation and reactivation (Asfestani et al., 2020; Fischer and Born, 2009; Lansink et al., 2009; Sterpenich et al., 2021), we trained healthy young participants to categorize these words into rewarded and unrewarded words to gain and to avoid losses of money points.  

Reviewer #3 (Public Review):

Summary:

In this study, the authors investigated the effects of targeted memory reactivation (TMR) during sleep on memory retention for artificial words with varying levels of phonotactical similarity to real words. The authors report that the high phonotactic probability (PP) words showed a more pronounced EEG alpha decrease during encoding and were more easily learned than the low PP words. Following TMR during sleep, participants who had been cued with the high PP TMR, remembered those words better than 0, whilst no such difference was found in the other conditions. Accordingly, the authors report higher EEG spindle band power during slow-wave up-states for the high PP as compared to low PP TMR trials. Overall, the authors conclude that artificial words that are easier to learn, benefit more from TMR than those which are difficult to learn.

Comment 12 & 13:

Strengths:

(1) The authors have carefully designed the artificial stimuli to investigate the effectiveness of TMR on words that are easy to learn and difficult to learn due to their levels of similarity with prior wordsound knowledge. Their approach of varying the level of phonotactic probability enables them to have better control over phonotactical familiarity than in a natural language and are thus able to disentangle which properties of word learning contribute to TMR success.

(2) The use of EEG during wakeful encoding and sleep TMR sheds new light on the neural correlates of high PP vs. low PP both during wakeful encoding and cue-induced retrieval during sleep.

We thank the reviewer for his/her positive evaluation of our manuscript.

Weaknesses:

Comment 14:

(1) The present analyses are based on a small sample and comparisons between participants. Considering that the TMR benefits are based on changes in memory categorization between participants, it could be argued that the individuals in the high PP group were more susceptible to TMR than those in the low PP group for reasons other than the phonotactic probabilities of the stimuli (e.g., these individuals might be more attentive to sounds in the environment during sleep). While the authors acknowledge the small sample size and between-subjects comparison as a limitation, a discussion of an alternative interpretation of the data is missing.

We agree with the reviewer that the small sample size and the between subject comparisons represent major limitations of our study. We thank the reviewer for this helpful comment and now discussed these limitations in more detail by adding alternative explanations and further suggestions for future research to overcome these limitations.

We added the following sentences to the discussion on p.14, ll. 465-473: 

To control for potential confounders despite the influence of difficulty in word learning on TMR, we compared parameters of sleep, the pre-sleep memory performance and the vigilance shortly before the post-sleep memory test, revealing no significant group differences (see Table S1 and S2). Nevertheless, we cannot rule out that other individual trait factors differed between the groups, such as the individual susceptibility to TMR. To rule out these alternative explanations based on individual factors, we suggest for future research to replicate our study by conducting a within-subject design with cueing of subsets of previously learned low- and high-PP words providing all conditions within the same individuals as shown in other TMR studies (Cairney et al., 2018; Schreiner and Rasch, 2015).

Comment 15:

(2) While the one-tailed comparison between the high PP condition and 0 is significant, the ANOVA comparing the four conditions (between subjects: cued/non-cued, within-subjects: high/low PP) does not show a significant effect. With a non-significant interaction, I would consider it statistically inappropriate to conduct post-hoc tests comparing the conditions against each other. Furthermore, it is unclear whether the p-values reported for the t-tests have been corrected for multiple comparisons. Thus, these findings should be interpreted with caution.

We thank the reviewer for this comment giving us the opportunity to correct our analyses and clarify with additional description. Indeed, we investigated at first overnight changes in behavior performance within the four conditions, conducting t-tests against 0 of Δ-values of d' and c-criterion. Whereas for all our statistical analyses the p-value was set at p < 0.05 for two-tailed testing, we did not corrected the p-value of our behavior analyses for multiple comparisons. To investigate subsequently differences between conditions, we conducted additional ANOVAs. We agree with the reviewer that without significant of results of the ANOVA, post-hoc analyses should not be conducted. Taken in account as well the recommendation of reviewer 1, we included now only post-hoc pairwise comparisons when the interaction effect of the ANOVA revealed at least a trend of significance (p < 0.1). 

We removed the following post-hoc analyses from the results section on p.9, ll. 291-295: 

Additional post-hoc pairwise comparisons revealed a significant difference between the highPP cued and low-PP uncued (high-PP cued vs. low-PP uncued: t(10) = 2.43, p = 0.04), and no difference to other conditions (high-PP cued vs.: high-PP uncued t(20) = 1.28, p = 0.22; lowPP cued t(20) = 1.57, p = 0.13).  

Further, we mentioned the lack of correction for multiple comparisons as a limitation of our results in the discussion on p.13, ll. 456-458:  

The criteria of data analyses were not pre-registered and the p-values of our behavior analyses were not corrected for multiple comparisons.

We added the following sentences to the methods p.23, ll. 842-849:

To analyze overnight changes of sleep behavioral data within TMR conditions, we conducted at first dependent sample t-tests against 0 of Δ-values (post-sleep test minus pre-sleep test) of d' and c-criterion (see Fig. 3). Two-way mixed design ANOVAs were computed to compare Δvalues between TMR conditions. After confirming at least a trend of significance (p < 0.1) for the interaction effect, we conducted post-hoc pairwise comparisons by independent and dependent sample t-tests. For all behavior statistical analyses, the p-value was set at p < 0.05 for two-tailed testing. A p-value < 0.1 and > 0.05 was reported as a trend of significance.

Comment 16:

(3) With the assumption that the artificial words in the study have different levels of phonotactic similarity to prior word-sound knowledge, it was surprising to find that the phonotactic probabilities were calculated based on an American English lexicon whilst the participants were German speakers. While it may be the case that the between-language lexicons overlap, it would be reassuring to see some evidence of this, as the level of phonotactic probability is a key manipulation in the study.

We thank the reviewer pointing to the misalignment between the German-speaking participants and the used artificial words based on American English. In line with this recommendation, we added a more outlined argumentation to the manuscript about the assumption of our study that major common phonetic characteristics across both languages are still preserved.       

We now discussed these aspects on p.14, ll. 473-481:

Further, we used artificial words based on American English in combination with German speaking participants, whereas language differences of pronunciation and phoneme structures might affect word perception and memory processing (Bohn and Best, 2012). On the other hand, both languages are considered to have the same language family (Eberhard et al., 2019) and the phonological distance between English and German is quite short compared for example to Korean (Luef and Resnik, 2023). Thus, major common phonological characteristics across both languages are still preserved. In addition, our behavior analyses revealed robust word discrimination learning and distinct memory performance according to different levels of phonotactic probabilities providing evidence of successful experimental manipulation. 

Comment 17:

(4) Another manipulation in the study is that participants learn whether the words are linked to a monetary reward or not, however, the rationale for this manipulation is unclear. For instance, it is unclear whether the authors expect the reward to interact with the TMR effects.

We used a reward task combined with TMR because previous studies revealed beneficial effects of reward related information on sleep dependent memory consolidation and reactivation (Asfestani et al., 2020; Fischer and Born, 2009; Lansink et al., 2009; Sterpenich et al., 2021). In addition, we wanted to increase the motivation of the participants, as they could receive additional monetary compensation according to their learning and memory task performances. Furthermore, we designed the task, with the overall possibility to translate this task to operant conditioning in rats (see research proposal: https://data.snf.ch/grants/grant/168602). However, the task turned out to be too difficult to translate to rats, whereas we developed a different learning paradigm for the animal study (Klaassen et al., 2021) of this cross-species research project.       

We added the following sentence to the introduction on p.4, ll. 134-137:

To consider the beneficial effect of reward related information on sleep dependent memory consolidation and reactivation (Asfestani et al., 2020; Fischer and Born, 2009; Lansink et al., 2009; Sterpenich et al., 2021), we trained healthy young participants to categorize these words into rewarded and unrewarded words to gain and to avoid losses of money points.  

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Comment 18:

(1) Please clearly define all linguistics terms - and most importantly the term "phonotactics" - at first use.

We thank the reviewer for this recommendation and we added the definition of phonotactics and further reduced the diversity of linguistic terms to improve readability. 

We added the following sentences to the beginning of the introduction on p.3, ll. 72-76:

One critical characteristic of similarity to pre-existing knowledge in auditory word processing is its speech sound (phoneme) pattern. In phonology as the field of language specific phoneme structures, phonotactics determines the constraints of word phoneme composition of a specific language.

Comment 19:

(2) Some critical details about the methods should be included in the Results section to make it comprehensible. For example, the way the crucial differences between G1-4 words should be addressed in the Results, not only in Figure 1.

According to the recommendation, we added this information to the results section.  We added the following sentences to the results section on p.4, ll. 145-154:

To study the impact of difficulty in word learning on TMR, we developed a novel learning paradigm. We formed four sets of artificial words (40 words per set; see Table S3 and S4) consisting of different sequences of two vowels and two consonants. Here, we subdivided the alphabet into two groups of consonants (C1: b, c, d, f, g, h, j, k, l, m; C2: n, p, q, r, s, t, v, w, x, z) and vowels (V1: a, e, I; V2: o, u, y). Four-letter-words were created by selecting letters from the vowel and consonant groups according to four different sequences (G1:C1, V1, V2, C2; G2: C1, V1, C2, V2; G3: V1, C1, C2, V2; G4: V1, C1, V2, C2; Fig. 1a; see methods for further details). Comparison analyses between the sets revealed significant differences in phonotactic probability (PP; Fig. 1b; unpaired t-tests: G1 / G2 > G3 / G4, p < 0.005, values of Cohen’s d > 0.71).

Comment 20

(3) Was scoring done both online and then verified offline? If so, please note that.

We included now this information.  

We adjusted the method section on p.21, ll. 765-769:   

The sleep stages of NREM 1 to 3 (N1 to N3), wake, and REM sleep were scored offline and manually according to the criteria of the American Academy of Sleep Medicine (AASM) by visual inspection of the signals of the frontal, central, and occipital electrodes over 30s epochs (Iber et al., 2007). Based on offline scoring, we confirmed TMR exposure during N2 and N3 and no significant differences (p-values > 0.05) of sleep parameters between the cueing groups (see Table S2).  

Comment 21:

(4) In Figure 2, please arrange the panel letters in an easier-to-read way (e.g., label upper right panel b with a different letter).

Now we rearranged the panel letters according to the recommendation.

We adjusted Figure 2 on p.8, ll. 242-258:     

Comment 22

(5) In the first paragraph on TMR effects, please note which memory measure you are comparing (i.e., d').

We added this information according to the recommendation.  

We adjusted the sentence of the results on p.8, ll. 260-263:

To examine whether TMR during sleep impacts memory consolidation of discrimination learning with respect to learning difficulty, we calculated the overnight changes by subtracting the pre- from the post-sleep memory performance based on d'-values of the reactivated sequences (cued) and non-reactivated sequences (uncued).

Comment 23:

(6) Please show the pre-sleep and post-sleep test scores for both word categories (not only the delta). It may be best to show this as another data point in Fig 2a, but it may be helpful to also see this split between cued and uncued.

We added the pre-sleep and post-sleep test scores with the individual data points as an additional figure. 

We added the following figure to the supplementary data on p.28, ll. 936-940:  

Comment 24:

(7) In the sentence "An additional two-way mixed design ANOVA on the same values with cueing as a between-subject factor (cued vs. uncued) ...", a more exact phrasing for the last parentheses would probably be "(high-PP-Cued vs Low-PP-Cued)". Both groups were cued.

We thank the reviewer pointing this out. According to the recommendation, we corrected the descriptions of the two-way mixed design ANOVAs. In addition, we detected a mistake of wrong assignments of the conditions to ANOVAs and corrected the reported values.   

We adjusted the sentences and corrected the values on p.9, ll. 271-275 and ll. 289-291: 

An additional two-way mixed design ANOVA on the same values with the factor cueing (cued vs. uncued) as a within-subject factor and group as a between-subject factor revealed trends of significance (p < 0.1) for the interaction (cueing × group: F(1,20) = 3.47, p = 0.08) and the main effect of group (F(1,20) = 3.28, p = 0.09). The main effect of cueing was not significant (F(1,20) = 0.58, p = 0.46).

An ANOVA on c-criterion changes showed no significant effects (interaction cueing × group: F(1,20) = 2.66, p = 0.12; main effect cueing  F(1,20) = 2.08, p = 0.17; main effect group F(1,20) = 0.38, p = 0.55).

Comment 25:

(8) In the same ANOVA, please mention that there is a trend toward an interaction effect. If there wasn't one, the post-hoc comparison would be unwarranted. Please consider noting other p<0.1 pvalues as a trend as well, for consistency.

Regarding this recommendation, we included now only post-hoc pairwise comparisons after confirming at least a trend toward an interaction effect of these ANOVAs and reported consistently a p-value < 0.1 and > 0.05 as a trend of significance.

We added the following sentences to the methods p.23, ll. 844-849:

Two-way mixed design ANOVAs were computed to compare Δ-values between TMR conditions. After confirming at least a trend of significance (p < 0.1) for the interaction effect, we conducted post-hoc pairwise comparisons by independent and dependent sample t-tests. For all behavior statistical analyses, the p-value was set at p < 0.05 for two-tailed testing. A p-value < 0.1 and > 0.05 was reported as a trend of significance.

We removed the following post-hoc analyses from the results section on p.9, ll. 291-295: 

Additional post-hoc pairwise comparisons revealed a significant difference between the highPP cued and low-PP uncued (high-PP cued vs. low-PP uncued: t(10) = 2.43, p = 0.04), and no difference to other conditions (high-PP cued vs.: high-PP uncued t(20) = 1.28, p = 0.22; lowPP cued t(20) = 1.57, p = 0.13).          

Comment 26:      

(9) Please consider adding an analysis correlating spindle power with memory benefit across participants. Even if it is non-significant, it is important to report given that some studies have found such a relationship.

According to this recommendation, we conducted an additional correlation analyses.

We added the following sentences to the manuscript into the results (pp. 10-11, ll. 346-349), the discussion (p.12, ll. 413-417), and the methods (p.23, ll. 864-867):   

Whereas we found a significant group difference in spindle power nested during SW up-states,   conducting further whole sample (n = 22) correlation analyses between the individual spindle power values of the significant cluster and the overnight changes of behavior measurements revealed no significant correlations (Δ d': r = 0.16, p = 0.48; Δ c-criterion: r = 0.19, p = 0.40).

In addition to our result of the significant group difference, we failed to find significant correlations between SW nested spindle power values and overnight changes in behavior measurements, whereas previous studies reported associations of SW and spindle activities during sleep with the integration of new memories in pre-existing knowledge networks (Tamminen et al., 2013, 2010).

By using the same extracted power values (0.3 to 0.8s; 11-14Hz; Pz, P3, P4, O2, P7) per subject, we performed whole sample (n = 22) Pearson correlation analyses between these power values and the overnight changes of behavior measurements of the cued condition (Δ d' and Δ ccriterion).

Reviewer #2 (Recommendations For The Authors):

(1) Choice of task

Comment 27:      

In general, I find your task well-designed and novel. In light of your research question, however, I wonder why you chose this task. When you outlined the research question in the introduction, I expected a task similar to Schreiner et al. (2015). For example, participants have to associate high PP words with each other and low PP words. The advantage here would be that you could test the benefits of TMR in a within-subjects design (for example, cueing half of the remembered high and half of the remembered low PP words).

Please see our previous response at comment 14.    

Comment 28:

Why did you decide to introduce a reward manipulation?

Please see our previous response at comment 11.    

Comment 29:

Why did you do the cueing on a category level (cueing all high PP or all low PP words instead of single word cueing or instead of cueing 20 reward high-PP, 20 unrewarded high-PP plus 20 reward low-PP and 20 unrewarded low-PP)? Both alternatives would have provided you the option to run your statistics within participants.

Please see our previous response at comment 14.    

Comment 30:

(2) Between-subjects design and small sample size.

Why did you decide on a between-subjects design that severely reduces your power?

Why did you just collect 22 participants with such a design? Were there any reasons for this small sample size? Honestly, I think publishing a TMR study with healthy participants and such a small sample size (11 participants for some comparisons) is not advisable.

Please see our previous response at comment 14.

Comment 31:

(3) Encoding performance.

Is d' significantly above 0 in the first repetition round? I would assume that the distinction between rewarded and non-rewarded words is just possible after the first round of feedback.

Indeed, conducting t-tests against 0 revealed significantly increased d'-values in the first repetition round (2nd presentation) in both PP conditions (high-PP: 0.85 ± 0.09, t(32) = 9.17, p < 0.001; low-PP: 0.62 ± 0.09, t(32) = 6.83, p < 0.001).  

Comment 32:

(4) Encoding response options

If you want to you could make it more explicit what exactly the response options are. I assume that one button means a word has a high reward and the other button means a word has a low reward. Making it explicit increases the understanding of the results section.

Please see our previous response at comment 3.

Comment 33:           

(5) Alpha desynchronisation.

Relative change

Why did you subtract alpha power during the 1st presentation from alpha power during 2nd and 3rd presentation? You baseline-corrected already and individually included the 1st, 2nd, and 3rd repetition in your behavioural analysis.

Based on this analysis, we aimed to examine the relative change in alpha power between PP-conditions of memory-relevant word repetitions. Therefore, to extract memory relevant changes of EEG activities, the first word presentation of naive stimulus processing could serve as a more representative baseline condition covering the time-window of interest of 0.7 to 1.9 s after the stimulus onset compared to a baseline condition before stimulus onset (-1 to -0.1s). 

To explain the rational of the analyses with the baseline condition more clearly, we added this information to the results section on p.7, ll. 222-226: 

We obtained the changes in power values by subtracting the first from the second and third presentation for the high- and low-PP condition, respectively. Here, the first word presentation of naive stimulus processing served us with a more representative baseline condition covering the time-window of interest of 0.7 to 1.9 s after the stimulus onset to examine relevant changes of encoding.  

Comment 34:

(6) Alpha desynchronisation as a neural correlate of encoding depth & difficulty?

"In addition to the behavior results, these EEG results indicate differences between PP conditions in desynchronization of alpha oscillations, as an assumed neural correlate of encoding depth. In addition to the behavior results, these EEG results indicate differences between PP conditions in desynchronization of alpha oscillations, as an assumed neural correlate of encoding depth."

Given that the low-PP words are more difficult to learn, I was expecting to see higher alpha desynchronisation in the low-PP relative to the high-PP words. Could you outline in a bit more detail how your findings fit into the literature (e.g., Simon Hanslmayr did a lot of work on this)?

I would also advise you to add citations e.g., after your sentence in the quote above ("as an assumed neural correlate of encoding depth").

We thank the reviewer for the recommendation giving us the opportunity to discuss in more detail how our results relate to previous findings. 

We added additional sentences to the discussion on p.13, ll. 441-455:    

Additional studies linked alpha desynchronization to cognitive effort and cognitive load (Proskovec et al., 2019; Zhu et al., 2021). So, one could assume to observe higher alpha desynchronization in the more difficult to learn condition of low-PP compared to high-PP. On the other hand numerous studies investigating oscillatory correlates of learning and memory showed that alpha desynchronization is associated with memory across different tasks, modalities and experimental phases of encoding and retrieval (Griffiths et al., 2016, 2021, 2019a, 2019b; Hanslmayr et al., 2009; Michelmann et al., 2016). Strikingly, Griffith and colleagues (Griffiths et al., 2019a) revealed by simultaneous EEG-fMRI recordings a negative correlation between the occurrence of patterns of stimulus-specific information detected by fMRI and cortical alpha/beta suppression. Here, the authors suggested that a decrease of alpha/beta oscillations might represent the neuronal mechanism of unmasking the task-critical signal by simultaneous suppression of task-irrelevant neuronal activities to promote information processing. Following this interpretation, we assume that over the course of learning elevated memory processing of the easier to learn stimuli is associated with enhanced information processing and thus accompanied by higher cortical alpha desynchronization in comparison of the more difficult to learn stimuli.

In addition, we added the mentioned quote on p.7, ll. 239-240:

In addition to the behavior results, these EEG results indicate differences between PP conditions in desynchronization of alpha oscillations, as an assumed neural correlate of encoding depth (Griffiths et al., 2021; Hanslmayr et al., 2009).

Comment 35:

(7) Exclusion criterion.

Why did you use a d' > 0.9 as a criterion for data inclusion?

This criterion ensured that each included subject had at least in one PP-condition a d' > 1.05 of pre-sleep memory performance, which corresponds to a general accuracy rate of 70%. 

Accordingly, we adjusted these sentences of the method section on p.19, ll. 677-680: 

Data were excluded from subjects who did not reach the minimal learning performance of d' > 1.05 during the pre-sleep memory test in at least one of the two PP conditions, whereas this threshold value corresponds to accuracy rates of 70% (n = 5). In addition, we excluded one subject who showed a negative d' in one PP condition of the pre-sleep memory test (n = 1). 

Comment 36:

(8) Coherence of wording.

When you talk about your dependent variable (d') you sometimes use sensitivity. I would stick to one term.

We replaced the word sensitivity with d'.    

(9) Criterion

Comment 37:

Why do you refer to a change in criterion (Figure 3b, axis labels) as a change in memory? Do you think the criterion says something about memory?

We corrected the axis label of Figure 3b and deleted here the word memory.

Comment 38:

Additionally, why did you analyse the effect of TMR on the criterion? Do you expect the criterion to change due to sleep-dependent memory consolidation? This section would benefit from more explanation. Personally, I am very interested in your thoughts and your hypothesis (if you had one, if not that is also fine but then, make it explicit that it was an exploratory analysis).

By conducting exploratory analyses of overnight changes of the c-criterion measurements, we aimed to examine the bias of decision-making to provide comprehensive data according to the framework of the signal detection theory. Regarding the previous literature showing mainly beneficial effects of sleep on learning and memory, we focused with our hypothesis on d' and explored additionally the c-criterion.

Despite our task design with gains/hits of +10 money points and losses/FAs of -8 (instead of -10), the subjects showed already during the pre-sleep memory task significant biases towards loss avoidance in both PP conditions (t-tests against 0: high-PP: 0.44 ± 0.07, t(21) = 5.63, p < 0.001; low-PP: 0.47 ± 0.09, t(21) = 5.51, p < 0.001). As already reported in the preprint, we found an additional significant increase of c-criterion by TMR solely for the high-PP words (see Fig. 3b). Even by integrating subjects with poor pre-sleep memory performance (high-PP-cueing group: n = 15; low-PP-cueing group: n = 13), t-tests against 0 revealed a significant increase of the high-PP cueing condition (t(14) = 3.36, p = 0.005) and no significant overnight changes in the other conditions (high-PP uncued: t(12) = 1.39, p = 0.19; low-PP cued: t(12) = 1.47, p = 0.17; low-PP uncued: t(14) = -0.20, p = 0.84). These exploratory findings on c-criterion suggest potential applications of TMR to affect decision-making biases in combination with reward learning.      

We revised the manuscript mentioning the exploratory character of the c-criterion analyses of the results on p.9, ll. 282-283 and of the discussion on p.12, ll. 400-402:  

We examined next as an exploratory analysis whether TMR conditions influence biases in decision-making.

By conducting an additional exploratory analysis, we observed a significant change of the decision bias in the cueing condition of the easy to learn words and no overnight changes in the other conditions.

Comment 39:

(10) You detected SWs in the time range of 0-6 sec post sound stimulation. How was the distribution of all detected SW down-states in this time range? (You could plot a histogram for this.)

We illustrated now the detected SWs in the time range of 0 to 6 s after stimulus onset. 

We added a histogram to the supplementary section on p.30, ll. 982-986:  

Reviewer #3 (Recommendations For The Authors):

Comment 40:

(1) In line with the weakness outlined above, I would recommend including a discussion of how the between-subject comparison and small sample size could affect the results and provide alternative interpretations.

Please see our previous response at comment 14.

Comment 41:

(2) Regarding my point about statistical comparisons, I would recommend that the authors follow best practice guidelines for post-hoc tests and multiple comparisons. In Figures 3a and b, I would also recommend removing the stars indicating significance from the post-hoc tests (if this is what they reflect). Perhaps this link will be useful: https://www.statology.org/anova-post-hoc-tests/

Please see our previous response at comment 15.    

Comment 42:

(3) Furthermore, to address any doubts about the possible phonotactic probability differences between languages, I would recommend that the authors show whether the languages overlap, the level of English fluency in the German-speaking participants, and/or another way of reassuring that this is unlikely to have affected the results.

Please see our previous response at comment 7.    

Comment 43:

(4) In the introduction, I would recommend that the authors outline a clear rationale for the reward/no reward manipulation.

Please see our previous response at comment 11.    

Comment 44:

(5) Figure 1c: Please include what response options participants had, e.g., 'rewarded/not rewarded'. This would make the type of categorization clearer to the reader.

Please see our previous response at comment 3.

Comment 45:

(6) It is unclear whether the additional ANOVA conducted on the time and frequency of the identified clusters included all channels or only the channels contributing to the cluster. Consider clarifying this in the relevant methods and results. Furthermore, I would recommend labelling this as a posthoc test as this analysis was guided by an initial peak at the data and the timings, frequencies, and channels of interest were not selected a-priori.

We thank the reviewer for this recommendation and labelled the additional repeatedmeasure ANOVA as a post-hoc test. Further, we mentioned the used channels (Pz and Cz) for this analyses.

We adjusted the results section on p.7, ll. 230-233 and the methods section on p.23, ll. 858-860:            

A post-hoc repeated-measure ANOVA on alpha power changes (merged over Pz and Cz electrodes) with PP (high vs. low) and presentations (2 to 3) as within-subjects factors revealed a main effect of PP (F(1,32) = 5.42, p = 0.03, η2 = 0.15), and a significant interaction (F(1,32)  = 7.38, p = 0.01, η2 = 0.19; Fig. 2e).

After confirming the existence of a significant cluster, we conducted an additional post-hoc repeated-measure ANOVA with averaged values of the identified time and frequency range of interest and merged over the Pz and Cz electrodes (see Fig. 2e).

Comment 46:

(7) Figure 3: To better illustrate within- vs. between-subjects comparisons and promote transparency, please add individual points and lines between the within-subjects conditions.

According to this recommendation, we changed Figure 3 to add the individual data points by lines.  

We modified Figure 3 on p.9, ll. 299-303:  

Comment 47:

(8) For the SW density time-bin analyses, please include statistics for all comparisons (i.e., through 0 s to 3 s) and say whether these were corrected for multiple comparisons.

According to this recommendation, we included now statistics for all comparisons. 

We added table S6 table to the supplementary data on p.29, l.962:     

Comment 48:

(9) Consider reporting effect sizes.

We thank the reviewer for this recommendation and we added now effect sizes of significant results. 

Comment 49:

(10) For transparency and replicability, consider including a list of the four stimulus sets including their phoneme and biphone probabilities.

We included a list of the four stimulus sets with their phoneme and biphone probabilities  

We added table S3 and table S4 to the supplementary data on pp. 26-27:       

References

Asfestani MA, Brechtmann V, Santiago J, Peter A, Born J, Feld GB. 2020. Consolidation of Reward Memory during Sleep Does Not Require Dopaminergic Activation. J Cogn Neurosci 32:1688– 1703. doi:10.1162/JOCN_A_01585

Batterink LJ, Oudiette D, Reber PJ, Paller KA. 2014. Sleep facilitates learning a new linguistic rule.

Neuropsychologia 65:169–79. doi:10.1016/j.neuropsychologia.2014.10.024

Batterink LJ, Paller KA. 2017. Sleep-based memory processing facilitates grammatical generalization: Evidence from targeted memory reactivation. Brain Lang 167:83–93. doi:10.1016/J.BANDL.2015.09.003

Bohn OS, Best CT. 2012. Native-language phonetic and phonological influences on perception of American English approximants by Danish and German listeners. J Phon 40:109–128. doi:10.1016/J.WOCN.2011.08.002

Cairney SA, Guttesen A á. V, El Marj N, Staresina BP. 2018. Memory Consolidation Is Linked to Spindle-Mediated Information Processing during Sleep. Curr Biol 28:948-954.e4. doi:10.1016/j.cub.2018.01.087

Eberhard DM, Simons GF, Fennig CD. 2019. Ethnologue: Languages of the world . SIL International. Online version: http://www.ethnologue.com.

Fischer S, Born J. 2009. Anticipated reward enhances offline learning during sleep. J Exp Psychol Learn Mem Cogn 35:1586–1593. doi:10.1037/A0017256

Green DM, Swets JA. 1966. Signal detection theory and psychophysics., Signal detection theory and psychophysics. Oxford,  England: John Wiley.

Griffiths B, Mazaheri A, Debener S, Hanslmayr S. 2016. Brain oscillations track the formation of episodic memories in the real world. Neuroimage 143:256–266. doi:10.1016/j.neuroimage.2016.09.021

Griffiths BJ, Martín-Buro MC, Staresina BP, Hanslmayr S, Staudigl T. 2021. Alpha/beta power decreases during episodic memory formation predict the magnitude of alpha/beta power decreases during subsequent retrieval. Neuropsychologia 153. doi:10.1016/j.neuropsychologia.2021.107755

Griffiths BJ, Mayhew SD, Mullinger KJ, Jorge J, Charest I, Wimber M, Hanslmayr S. 2019a. Alpha/beta power decreases track the fidelity of stimulus specific information. Elife 8. doi:10.7554/eLife.49562

Griffiths BJ, Parish G, Roux F, Michelmann S, van der Plas M, Kolibius LD, Chelvarajah R, Rollings DT, Sawlani V, Hamer H, Gollwitzer S, Kreiselmeyer G, Staresina B, Wimber M, Hanslmayr S. 2019b. Directional coupling of slow and fast hippocampal gamma with neocortical alpha/beta oscillations in human episodic memory. Proc Natl Acad Sci U S A 116:21834–21842. doi:10.1073/pnas.1914180116

Hanslmayr S, Spitzer B, Bäuml K-H. 2009. Brain oscillations dissociate between semantic and nonsemantic encoding of episodic memories. Cereb Cortex 19:1631–40. doi:10.1093/cercor/bhn197

Iber C, Ancoli‐Israel S, Chesson AL, Quan SF. 2007. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. Westchester, IL: American Academy of Sleep Medicine.

Klaassen AL, Heiniger A, Sánchez PV, Harvey MA, Rainer G. 2021. Ventral pallidum regulates the default mode network, controlling transitions between internally and externally guided behavior. Proc Natl Acad Sci U S A 118:1–10. doi:10.1073/pnas.2103642118

Lansink CS, Goltstein PM, Lankelma J V., McNaughton BL, Pennartz CMA. 2009. Hippocampus leads ventral striatum in replay of place-reward information. PLoS Biol 7. doi:10.1371/JOURNAL.PBIO.1000173

Luef EM, Resnik P. 2023. Phonotactic Probabilities and Sub-syllabic Segmentation in Language

Learning. Theory Pract Second Lang Acquis 9:1–31. doi:10.31261/TAPSLA.12468

Michelmann S, Bowman H, Hanslmayr S. 2016. The Temporal Signature of Memories: Identification of a General Mechanism for Dynamic Memory Replay in Humans. PLoS Biol 14:e1002528. doi:10.1371/journal.pbio.1002528

Proskovec AL, Heinrichs-Graham E, Wilson TW. 2019. Load Modulates the Alpha and Beta Oscillatory Dynamics Serving Verbal Working Memory. Neuroimage 184:256. doi:10.1016/J.NEUROIMAGE.2018.09.022

Reber AS. 1967. Implicit learning of artificial grammars. J Verbal Learning Verbal Behav 6:855–863.

doi:10.1016/S0022-5371(67)80149-X

Schreiner T, Rasch B. 2015. Boosting vocabulary learning by verbal cueing during sleep. Cereb Cortex 25:4169–4179. doi:10.1093/cercor/bhu139

Sterpenich V, van Schie MKM, Catsiyannis M, Ramyead A, Perrig S, Yang H-D, Van De Ville D, Schwartz S. 2021. Reward biases spontaneous neural reactivation during sleep. Nat Commun 2021 121 12:1–11. doi:10.1038/s41467-021-24357-5

Tamminen J, Lambon Ralph MA, Lewis PA. 2013. The role of sleep spindles and slow-wave activity in integrating new information in semantic memory. J Neurosci 33:15376–15381. doi:10.1523/JNEUROSCI.5093-12.2013

Tamminen J, Payne JD, Stickgold R, Wamsley EJ, Gaskell MG. 2010. Sleep spindle activity is associated with the integration of new memories and existing knowledge. J Neurosci 30:14356–60. doi:10.1523/JNEUROSCI.3028-10.2010

Zhu Y, Wang Q, Zhang L. 2021. Study of EEG characteristics while solving scientific problems with different mental effort. Sci Rep 11. doi:10.1038/S41598-021-03321-9

Author response:

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

Reviewer #1 (Public review):

Summary:

In this manuscript, Herrmannova et al explore changes in translation upon individual depletion of three subunits of the eIF3 complex (d, e and h) in mammalian cells. The authors provide a detailed analysis of regulated transcripts, followed by validation by RT-qPCR and/or Western blot of targets of interest, as well as GO and KKEG pathway analysis. The authors confirm prior observations that eIF3, despite being a general translation initiation factor, functions in mRNA-specific regulation, and that eIF3 is important for translation re-initiation. They show that global effects of eIF3e and eIF3d depletion on translation and cell growth are concordant. Their results support and extend previous reports suggesting that both factors control translation of 5'TOP mRNAs. Interestingly, they identify MAPK pathway components as a group of targets coordinately regulated by eIF3 d/e. The authors also discuss discrepancies with other reports analyzing eIF3e function.

Strengths:

Altogether, a solid analysis of eIF3 d/e/h-mediated translation regulation of specific transcripts. The data will be useful for scientists working in the Translation field.

Weaknesses:

The authors could have explored in more detail some of their novel observations, as well as their impact on cell behavior.

The manuscript has improved with the new corrections. I appreciate the authors' attention to the minor comments, which have been fully solved. The authors have not, however, provided additional experimental evidence that uORF-mediated translation of Raf-1 mRNA depends on an intact eIF3 complex, nor have they addressed the consequences of such regulation for cell physiology. While I understand that this is a subject of follow-up research, the authors could have at least included their explanations/ speculations regarding major comments 2-4, which in my opinion could have been useful for the reader.

Our explanations/speculations regarding major comments 2 and 3 were included in the Discussion. We apologize for this misunderstanding as we thought that we were supposed to explain our ideas only in the responses. We did not discuss the comment 4, however, as we are really not sure what is the true effect and did not want to go into wild speculations in our manuscript. We thank this reviewer for his insightful comments and understanding.

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

Reviewer #1 (Recommendations For The Authors):

Major comments:

(1) The authors report the potential translational regulation of Raf kinase by re-initiation. It would be interesting to show that Raf is indeed regulated by uORF-mediated translation, and that this is dependent on an intact eIF3 complex. Analyzing the potential consequences of Raf1 regulation for cancer cell proliferation or apoptosis would be a plus.

We agree that this is an interesting and likely possibility. In fact, another clue that translation of Raf1 is regulated by uORFs comes from Bohlen et al. 2023 (PMID: 36869665) where they showed that RAF1 translation is dependent on PRRC2 proteins (that promote leaky scanning through these uORFs). We noted in the discussion that our results from eIF3d/e/hKD and the PRRC2A/B/CKD partly overlap. It is a subject of our follow-up research to investigate whether eIF3 and PRRC2 co-operate together to regulate translation of this important mRNA. 

(2) The authors show that eIF3 d/e -but not 3h- has an effect on cell proliferation. First, this indicates that proliferation does not fully correlate with eIF3 integrity. Depletion of eIF3d does not affect the integrity of eIF3, yet the effects on proliferation are similar to those of eIF3e. What is the possibility that changes in proliferation reflect functions of eIF3d outside the eIF3 complex? What could be the real consequences of disturbing eIF3 integrity for the mammalian cell? Please, discuss.

Yes, proliferation does not fully correlate with eIF3 integrity. Downregulation of eIF3 subunits that lead to disintegration of eIF3 YLC core (a, b, c, g, i) have more detrimental effect on growth and translation than downregulation of the peripheral subunits (e, k, l, f, h, m). Our previous studies (Wagner et al. 2016, PMID: 27924037 and Herrmannová et al. 2020, PMID: 31863585) indicate that the YLC core of eIF3 can partially support translation even without its peripheral subunits. In this respect eIF3d (as a peripheral subunit) is an amazing exception, suggesting it may have some specialized function(s). Whether this function resides outside of the eIF3 complex or not we do not know, but do not think so. Mainly because in the absence of eIF3e – its interaction partner, eIF3d gets rapidly degraded. Therefore, it is not very likely that eIF3d exists alone outside of eIF3 complex with moonlighting functions elsewhere. We think that eIF3d, as a head-interacting subunit close to an important head ribosomal protein RACK1 (a landing pad for regulatory proteins), is a target of signaling pathways, which may make it important for translation of specific mRNAs. In support is these thoughts, eIF3d (in the context of entire eIF3) together with DAP5 were shown to promote translation by an alternate capdependent (eIF4F-independent) mechanism (Lee et al. 2016, PMID: 27462815; de la Parra et al. 2018, PMID:30076308). In addition, the eIF3d function (also in the context of entire eIF3) was proved to be regulated by stress-triggered phosphorylation (Lamper et al. 2020, PMID: 33184215). 

(3) Figure 6D: Surprisingly, reduced levels of ERK1/2 upon eIF3d/e-KD are compensated by increased phosphorylation of ERK1/2 and net activation of c-Jun. Please comment on the functional consequences of buffering mechanisms that the cell deploys in order to counteract compromised eIF3 function. Why would the cell activate precisely the MAPK pathway to compensate for a compromised eIF3 function?

This we do not know. We can only speculate that when translation is compromised, cells try to counteract it in two ways: 1) they produce more ribosomes to increase translational rates and 2) activate MAPK signaling to send pro-growth signals, which can in the end further boost ribosome biogenesis.

(4) Regarding DAP-sensitive transcripts, can the authors discuss in more detail the role of eIF3d in alternative cap-dependent translation versus re-initiation? Are these transcripts being translated by a canonical cap- and uORF-dependent mechanism or by an alternative capdependent mechanism?

This is indeed not an easy question. On one hand, it was shown that DAP5 facilitates translation re-initiation after uORF translation in a canonical cap-dependent manner. This mechanism is essential for translation of the main coding sequence (CDS) in mRNAs with structured 5' leaders and multiple uORFs. (Weber et al. 2022, PMID: 36473845; David et al., 2022, PMID: 35961752). On the other hand, DAP5 was proposed to promote alternative, eIF4F-independent but cap-dependent translation, as it can substitute the function of the eIF4F complex in cooperation with eIF3d (de la Parra et al., 2018, PMID: 30076308; Volta et al., 2021 34848685). Overall, these observations paint a very complex picture for us to propose a clear scenario of what is going on between these two proteins on individual mRNAs. We speculate that both mechanisms are taking place and that the specific mechanism of translation initiation differs for differently arranged mRNAs.

Minor comments:

(5) Figure S2C: why is there a strong reduction of the stop codon peak for 3d and 3h KDs?

We have checked the Ribowaltz profiles of all replicates (in the Supplementary data we are showing only a representative replicate I) and the stop codon peak differs a lot among the replicates. We think that this way of plotting was optimized for calculation and visualization of P-sites and triplet periodicity and thus is not suitable for this type of comparison among samples. Therefore, we have performed our own analysis where the 5’ ends of reads are used instead of P-sites and triplicates are averaged and normalized to CDS (see below please), so that all samples can be compared directly in one plot (same as Fig. S13A but for stop codon). We can see that the stop codon peak really differs and is the smallest for eIF3hKD. However, these changes are in the range of 20% and we are not sure about their biological significance. We therefore refrain from drawing any conclusions. In general, reduced stop codon peak may signal faster termination or increased stop codon readthrough, but the latter should be accompanied by an increased ribosome density in the 3’UTR, which is not the case. A defect in termination efficiency would be manifested by an increased stop codon peak, instead.

Author response image 1.

(6) Figures 5 and S8: Adding a vertical line at 'zero' in all cumulative plots will help the reader understand the author's interpretation of the data. 

We have added a dashed grey vertical line at zero as requested. However, for interpretation of these plots, the reader should focus on the colored curve and whether it is shifted in respect to the grey curve (background) or not. Shift to the right indicates increased expression, while shift to the left indicates decreased expression. The reported p-value then indicates the statistical significance of the shift.

(7) The entire Figure 2 are controls that can go to Supplementary Material. The clustering of Figure S3B could be shown in the main Figure, as it is a very easy read-out of the consistent effects of the KDs of the different eIF3 subunits under analysis.

We have moved the entire Figure 2 to Supplementary Material as suggested (the original panels can be found as Supplementary Figures 1B, 1C and 3A). Figure S3B is now the main Figure 2E. 

(8) There are 3 replicates for Ribo-Seq and four for RNA-Seq. Were these not carried out in parallel, as it is usually done in Ribo-seq experiments? Why is there an extra replicate for RNASeq?

Yes, the three replicates were carried out in parallel. We have decided to add the fourth replicate in RNA-Seq to increase the data robustness as the RNA-Seq is used for normalization of FP to calculate the TE, which was our main analyzed metrics in this article. We had the option to add the fourth replicate as we originally prepared five biological replicates for all samples, but after performing the control experiments, we selected only the 3 best replicates for the Ribo-Seq library preparation and sequencing.  

(9) Please, add another sheet in Table S2 with the names of all genes that change only at the translation (RPF) levels.

As requested, we have added three extra sheets (one for each downregulation) for differential FP with Padjusted <0.05 in the Spreadsheet S2. We also provide a complete unfiltered differential expression data (sheet named “all data”), so that readers can filter out any relevant data based on their interest.

(10) Page 5, bottom: ' ...we showed that the expression of all 12 eIF3 subunits is interconnected such that perturbance of the expression of one subunit results in the down-regulation of entire modules...'. This is not true for eIF3d, as shown in Fig1B and mentioned in Results.

This reviewer is correct. By this generalized statement, we were trying to summarize our previous results from Wagner et al., 2014, PMID: 24912683; Wagner et al.,2016, PMID: 27924037 and Herrmannova et al.,2020, PMID: 31863585. The eIF3d downregulation is the only exception that does not affect expression of any other eIF3 subunit. Therefore, we have rewritten this paragraph accordingly: “We recently reported a comprehensive in vivo analysis of the modular dynamics of the human eIF3 complex (Wagner et al, 2020; Wagner et al, 2014; Wagner et al., 2016). Using a systematic individual downregulation strategy, we showed that the expression of all 12 eIF3 subunits is interconnected such that perturbance of the expression of one subunit results in the down-regulation of entire modules leading to the formation of partial eIF3 subcomplexes with limited functionality (Herrmannova et al, 2020). eIF3d is the only exception in this respect, as its downregulation does not influence expression of any other eIF3 subunit.”

(11) Page 10, bottom: ' The PCA plot and hierarchical clustering... These results suggest that eIF3h depletion impacts the translatome differentially than depletion of eIF3e or eIF3d.' This is already obvious in the polysome profiles of Figure S2C.

We agree that this result is surely not surprising given the polysome profile and growth phenotype analyses of eIF3hKD. But still, we think that the PCA plot and hierarchical clustering results represent valuable controls. Nonetheless, we rephrased this section to note that this result agrees with the polysome profiles analysis: “The PCA plot and hierarchical clustering (Figure 2A and Supplementary Figure 4A) showed clustering of the samples into two main groups: Ribo-Seq and RNA-seq, and also into two subgroups; NT and eIF3hKD samples clustered on one side and eIF3eKD and eIF3dKD samples on the other. These results suggest that the eIF3h depletion has a much milder impact on the translatome than depletion of eIF3e or eIF3d, which agrees with the growth phenotype and polysome profile analyses (Supplementary Figure 1A and 1D).”

(12) Page 12: ' As for the eIF3dKD "unique upregulated" DTEGs, we identified one interesting and unique KEGG pathway, the ABC transporters (Supplementary Figure 5A, in green).' This sentence is confusing, as there are more pathways that are significant in this group, so it is unclear why the authors consider it 'unique'.

The eIF3dKD “unique upregulated” group comprises genes with increased TE only in eIF3dKD but not in eIF3eKD or eIF3hKD (500 genes, Fig 2G). All these 500 genes were examined for enrichment in the KEGG pathways, and the top 10 significant pathways were reported (Fig S6A). However, 8 out of these 10 pathways were also significantly enriched in other gene groups examined (e.g. eIF3d/eIF3e common). Therefore, the two remaining pathways (“ABC transporters” and “Other types of O-glycan biosynthesis”) are truly unique for eIF3dKD. We wanted to highlight the ABC transporters group in particular because we find it rather interesting (for the reasons mentioned in the article). We have corrected the sentence in question to avoid confusion: “Among the eIF3dKD “unique upregulated” DTEGs, we identified one interesting KEGG pathway, the ABC transporters, which did not show up in other gene groups (Supplementary Figure 6A, in green). A total of 12 different ABC transporters had elevated TE (9 of them are unique to eIF3dKD, while 3 were also found in eIF3eKD), 6 of which (ABCC1-5, ABCC10) belong to the C subfamily, known to confer multidrug resistance with alternative designation as multidrug resistance protein (MRP1-5, MRP7) (Sodani et al, 2012).

Interestingly, all six of these ABCC transporters were upregulated solely at the translational level (Supplementary Spreadsheet S2).”    

(13) Note typo ('Various') in Figure 4A.

Corrected

(14) The introduction could be shortened.

This is a very subjective requirement. In fact, when this manuscript was reviewed in NAR, we were asked by two reviewers to expand it substantially. Because a number of various research topics come together in this work, e.g. translational regulation, the eIF3 structure and function, MAPK/ERK signaling, we are convinced that all of them demand a comprehensive introduction for non-experts in each of these topics. Therefore, with all due respect to this reviewer, we did not ultimately shorten it.

Reviewer #2 (Recommendations For The Authors):

- In Figure 2, it would be useful to know why eIF3d is destabilized by eIF3e knockdown - is it protein degradation and why do the eIF3d/e knockdowns not more completely phenocopy each other when there is the same reduction to eIF3d as in the eIF3d knockdown sample?

Yes, we do think that protein degradation lies behind the eIF3d destabilization in the eIF3eKD, but we have not yet directly demonstrated this. However, we have shown that eIF3d mRNA levels are not altered in eIF3eKD and that Ribo-Seq data indicate no change in TE or FP for eIF3d-encoding mRNA in eIF3eKD. Nonetheless, it is important to note (and we discuss it in the article) that eIF3d levels in eIF3dKD are lower than eIF3d levels in eIF3eKD (please see Supplementary Figure 1C). In fact, we believe that this is one of the main reasons for the eIF3d/e knockdowns differences.

- The western blots in Figures 4 and 6 show modest changes to target protein levels and would be strengthened by quantification.

We have added the quantifications as requested by this reviewer and the reviewer 3.

- For Figure 4, this figure would be strengthened by experiments showing if the increase in ribosomal protein levels is correlated with actual changes to ribosome biogenesis.

As suggested, we performed polysome profiling in the presence of EDTA to monitor changes in the 60S/40S ratio, indicating a potential imbalance in the biogenesis of individual ribosome subunits. We found that it was not affected (Figure 3G). In addition, we performed the same experiment, normalizing all samples to the same number of cells (cells were carefully counted before lysis). In this way, we confirmed that eIF3dKD and eIF3eKD cells indeed contain a significantly increased number of ribosomes, in agreement with the western blot analysis (Figure 3H).

- In Figure 6, there needs to be a nuclear loading control.

This experiment was repeated with Lamin B1 used as a nuclear loading control – it is now shown as Fig. 5F.

- For Figure 8, these findings would be strengthened using luciferase reporter assays where the various RNA determinants are experimentally tested. Similarly, 5′ TOP RNA reporters would have been appreciated in Figure 4.

This is indeed a logical continuation of our work, which represents the current work in progress of one of the PhD students. We apologize, but we consider this time- and resource-demanding analysis out of scope of this article.

Reviewer #3 (Recommendations For The Authors):

(1) Within the many effects observed, it is mentioned that eIF3d is known to be overexpressed while eIF3e is underexpressed in many cancers, but knockdown of either subunit decreases MDM2 levels, which would be expected to increase P53 activity and decrease tumor cell transformation. In contrast, they also report that 3e/3d knockdown dramatically increases levels of cJUN, presumably due to increased MAPK activity, and is expected to increase protumor gene expression. Additional discussion is needed to clarify the significance of the findings, which are a bit confusing.

This is indeed true. However, considering the complexity of eIF3, the largest initiation factor among all, as well as the broad portfolio of its functions, it is perhaps not so surprising that the observed effects are complex and may seem even contradictory in respect to cancer. To acknowledge that, we expanded the corresponding part of discussion as follows: “Here, we demonstrate that alterations in the eIF3 subunit stoichiometry and/or eIF3 subcomplexes have distinct effects on the translatome; for example, they affect factors that play a prominent (either positive or negative) role in cancer biology (e.g., MDM2 and cJUN), but the resulting impact is unclear so far. Considering the complex interactions between these factors as well as the complexity of the eIF3 complex per se, future studies are required to delineate the specific oncogenic and tumor suppressive pathways that play a predominant role in mediating the effects of perturbations in the eIF3 complex in the context of neoplasia.”

(2) There are places in the text where the authors refer to changes in transcriptional control when RNA levels differ, but transcription versus RNA turnover wasn't tested, e.g. page 16 and Figure S10, qPCR does not confirm "transcriptional upregulation in all three knockdowns" and page 19 "despite apparent compensatory mechanisms that increase their transcription."

This is indeed true, the sentences in question were corrected. The term “increased mRNA levels” was used instead of transcriptional upregulation (increased mRNA stabilization is also possible).

(3) Similarly, the authors suggest that steady-state LARP1 protein levels are unaffected based on ribosome footprint counts (page 21). It is incorrect to assume this, because ribosome footprints can be elevated due to stalling on RNA that isn't being translated and doesn't yield more protein, and because levels of translated RNA/synthesized proteins do not always reflect steady-state protein levels, especially in mutants that could affect lysosome levels and protein turnover. Also page 12, 1st paragraph suggests protein production is down when ribosome footprints are changed.

Yes, we are well-aware of this known limitation of Ribo-seq analysis. Therefore, the steadystate protein levels of our key hits were verified by western blotting. In addition, we have removed the sentence about LARP1 because it was based on Ribo-Seq data only without experimental evaluation of the steady-state LARP1 protein levels.

(4) The translation buffering effect is not clear in some Figures, e.g. S6, S8, 8A, and B. The authors show a scheme for translationally buffered RNAs being clustered in the upper right and lower left quadrants in S4H (translation up with transcript level down and v.v.), but in the FP versus RNA plots, the non-TOP RNAs and 4E-P-regulated RNAs don't show this behavior, and appear to show a similar distribution to the global changes. Some of the right panels in these figures show modest shifts, but it's not clear how these were determined to be significant. More information is needed to clarify, or a different presentation, such as displaying the RNA subsets in the left panels with heat map coloring to reveal whether RNAs show the buffered translation pattern defined in purple in Figure S4H, or by reporting a statistical parameter or number of RNAs that show behavior out of total for significance. Currently the conclusion that these RNAs are translationally buffered seems subjective since there are clearly many RNAs that don't show changes, or show translation-only or RNA-only changes.

We would like to clarify that S4H does not indicate a necessity for changes in FPs in the buffered subsets. Although opposing changes in total mRNA and FPs are classified as buffering, often we also consider the scenario where there are changes to the total mRNA levels not accompanied by changes in ribosome association.

In figure S6, the scatterplots indicate a high density of genes shifted towards negative fold changes on the x-axis (total mRNA). This is also reflected in the empirical cumulative distribution functions (ecdfs) for the log2 fold changes in total mRNA in the far right panels of A and B, and the lack of changes in log2 fold change for FPs (middle panels). Similarly, in figure S8, the scatterplots indicate a density of genes shifted towards positive fold changes on the x-axis for total mRNA. The ecdfs also demonstrate that there is a significant directional shift in log2 fold changes in the total mRNA that is not present to a similar degree in the FPs, consistent with translational offsetting. It is rightly pointed out that not all genes in these sets follow the same pattern of regulation. We have revised the title of Supplementary Figure S6 (now S7) to reflect this. However, we would like to emphasize that these figures are not intended to communicate that all genes within these sets of interest are regulated in the same manner, but rather that when considered as a whole, the predominant effect seen is that of translational offsetting (directional shifts in the log2 fold change distribution of total mRNA that are not accompanied by similar shifts in FP mRNA log2 fold changes).

The significance of these differences was determined by comparing the ecdfs of the log2 fold changes for the genes belonging to a particular set (e.g. non-TOP mTOR-sensitive, p-eIF4E-sensitive) against all other expressed genes (background) using a Wilcoxan rank sum test. This allows identification of significant shifts in the distributions that have a clear directionality (if there is an overall increase, or decrease in fold changes of FPs or total mRNA compared to background). If log2 fold changes are different from background, but without a clear directionality (equally likely to be increased or decreased), the test will not yield a significant result. This approach allows assessment of the overall behavior of gene signatures within a given dataset in a manner that is completely threshold-independent, such that it does not rely on classification of genes into different regulatory categories (translation only, buffering, etc.) based on significance or fold-change cut-offs (as in S4H). Therefore, we believe that this unbiased approach is well-suited for identifying cases when there are many genes that follow similar patterns of regulation within a given dataset.

(5) Page 10-"These results suggest that eIF3h depletion impacts the translatome differentially than depletion of eIF3e or eIF3d" ...These results suggest that eIF3h has less impact on the translatome, not that it does so differently. If it were changing translation by a different mechanism, I would not expect it to cluster with control.

This sentence was rewritten as follows: “The PCA plot and hierarchical clustering (Figure 2A and Supplementary Figure 4A) showed clustering of the samples into two main groups: RiboSeq and RNA-seq, and also into two subgroups; NT and eIF3hKD samples clustered on one side and eIF3eKD and eIF3dKD samples on the other. These results suggest that the eIF3h depletion has a much milder impact on the translatome than depletion of eIF3e or eIF3d, which agrees with the growth phenotype and polysome profile analyses (Supplementary Figure 1A and 1D).”

Other minor issues:

(1) There are some typos: Figure 2 leves, Figure 4 variou,

Corrected.

(2) Figure 3, font for genes on volcano plot too small

Yes, maybe, however the resolution of this image is high enough to enlarge a certain part of it at will. In our opinion, a larger font would take up too much space, which would reduce the informativeness of this graph.

(3) Figure S5, highlighting isn't defined.

The figure legend for S5A (now S6A) states: “Less significant terms ranking 11 and below are in grey. Terms specifically discussed in the main text are highlighted in green.” Perhaps it was overlooked by this reviewer.

(4) At several points the authors refer to "the MAPK signaling pathway", suggesting there is a single MAPK that is affected, e.g in the title, page 3, and other places when it seems they mean "MAPK signaling pathways" since several MAPK pathways appear to be affected.

We apologize for any terminological inaccuracies. There are indeed several MAPK pathways operating in cells. In our study, we focused mainly on the MAPK/ERK pathway. The confusion probably stems from the fact that the corresponding term in the KEGG pathway database is labeled "MAPK signaling pathway" and this term, although singular, includes all MAPK pathways. We have carefully reviewed the entire article and have corrected the term used accordingly to either: 1) MAPK pathways in general, 2) the MAPK/ERK pathway for this particular pathway, or 3) "MAPK signaling pathway", where the KEGG term is meant.

(5) Some eIF3 subunit RNAs have TOP motifs. One might expect 3e and 3h levels to change as a function of 3d knockdown due to TOP motifs but this is not observed. Can the authors speculate why the eIF3 subunit levels don't change but other TOP RNAs show TE changes? Is this true for other translation factors, or just for eIF3, or just for these subunits? Could the Western blot be out of linear range for the antibody or is there feedback affecting eIF3 levels differently than the other TOP RNAs, or a protein turnover mechanism to maintain eIF3 levels?

This is indeed a very interesting question. In addition to the mRNAs encoding ribosomal proteins, we examined all TOP mRNAs and added an additional sheet to the S2 supplemental spreadsheet with all TOP RNAs listed in (Philippe et al., 2020, PMID: 32094190). According to our Ribo-Seq data, we could expect to see increased protein levels of eIF3a and eIF3f in eIF3dKD and eIF3eKD, but this is not the case, as judged from extensive western blot analysis performed in (Wagner et. al 2016, PMID: 27924037). Indeed, we cannot rule out the involvement of a compensatory mechanism monitoring and maintaining the levels of eIF3 subunits at steady-state – increasing or decreasing them if necessary, which could depend on the TOP motif-mediated regulation. However, we think that in our KDs, all non-targeted subunits that lose their direct binding partner in eIF3 due to siRNA treatment become rapidly degraded. For example, co-downregulation of subunits d, k and l in eIF3eKD is very likely caused by protein degradation as a result of a loss of their direct binding partner – eIF3e. Since we showed that the yeast eIF3 complex assembles co-translationally (Wagner et. al 2020, PMID: 32589964), and there is no reason to think that mammalian eIF3 differs in this regard, our working hypothesis is that free subunits that are not promptly incorporated into the eIF3 complex are rapidly degraded, and the presence or absence of the TOP motif in the 5’ UTR of their mRNAs has no effect. As for the other TOP mRNAs, translation factors eEF1B2, eEF1D, eEF1G, eEF2 have significantly increased FPs in both eIF3dKD and eIF3eKD, but we did not check their protein levels by western blotting to conclude anything specific.

Author response:

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

The detailed, thorough critique provided by the three reviewers is very much appreciated. We believe the manuscript is greatly improved by the changes we have made based on those reviews. The major changes are described below, followed by a point by point response.

Major Changes:

(1) We revised our model (old Fig. 10; new Fig. 9) to keep the explanation focused on the data shown in the current study. Specifically, references to GTP/GDP states of Rab3A and changes in the presynaptic quantum have been removed and the mechanisms depicted are confined to pre- or post-synaptic Rab3A participating in either controlling release of a trophic factor that regulates surface GluA2 receptors (pre- or postsynaptic) or directly affecting fusion of GluA2-receptor containing vesicles (postsynaptic).

(2) We replaced all cumulative density function plots and ratio plots, based on multiple quantile samples per cell, with box plots of cell means. This affects new Figures 1, 2, 3, 5, 6, 7 and 8. All references to “scaling,” “divergent scaling,” or “uniform scaling,” have been removed. New p values for comparison of means are provided above every box plot in Figures 1, 2, 3, 5, 6, 7 and 8. The number of cultures is provided in the figure legends.

(3) We have added frequency to Figures 1, 2 and 8. Frequency values overall are more variable, and the effect of activity blockade less robust, than for mEPSC amplitudes. We have added text indicating that the increase in frequency after activity blockade was significant in neurons from cultures prepared from WT in the Rab3A+/- colony but not cultures prepared from KO mice (Results, lines 143 to 147, new Fig. 1G. H). The TTX-induced increase in frequency was significant in the NASPM experiments before NASPM, but not after NASPM (Results, lines 231 to 233, new Fig. 3, also cultures from WT in Rab3A+/- colony). The homeostatic plasticity effect on frequency did not reach significance in WT on WT glia cultures or

WT on KO glia cultures, possibly due to the variability of frequency, combined with smaller sample sizes (Results, lines 400 to 403, new Fig. 8). In the cultures prepared from WT mice in the Rab3A+/Ebd colony, there was a trend towards higher frequency after TTX that did not reach statistical significance, and in cultures prepared from mutant mice, the p value was large, suggesting disruption of the effect, which appears to be due to an increase in frequency in untreated cultures, similar to the behavior of mEPSC amplitudes in neurons from mutant mice (Results, lines 161-167). In sum, the effect of activity on frequency requires Rab3A and Ca2+-permeable receptors, and is mimicked by the presence of the Rab3A Earlybird mutant. We have also added a discussion of these results (Discussion, lines 427-435). 

(4) In the revised manuscript we have added analysis of VGLUT1 levels for the same synaptic sites that we previously analyzed GluA2 levels, and these data are described in Results, lines 344 to 371, and appear in new Table 2. In contrast to previous studies, we did not find any evidence for an increase in VGLUT1 levels after activity blockade. We reviewed those studies to determine whether there might be differences in the experimental details that could explain the lack of effect we observed. In (De Gois et al., 2005), the authors measured mRNA and performed western blots to show increases in VGLUT1 after TTX treatment in older rat cortical cultures (DIV 19). The study performs immunofluorescence imaging of VGLUT1 but only after bicuculline treatment (it decreases), not after TTX treatment. In (Wilson et al.,

2005), the hippocampal cultures are treated with AP5, not TTX, and the VGLUT1 levels in immunofluorescence images are reported relative to synapsin I. That the type of activity blockade matters is illustrated by the failure of Wilson and colleagues to observe a consistent increase in VGLUT1/Synapsin ratio in cultures treated with AMPA receptor blockade (NBQX; supplementary information). These points have been added to the Discussion, lines 436 to 447.)

Reviewer #1:

(1) (model…is not supported by the data), (2) (The analysis of mEPSC data using quantile sampling…), (3) (…statistical analysis of CDFs suffers from n-inflation…), (4) (How does recording noise and the mEPSC amplitude threshold affect “divergent scaling?”) (5) (…justification for the line fits of the ratio data…), (7) (A comparison of p-values between conditions….) and (10) (Was VGLUT intensity altered in the stainings presented in the manuscript?)

The major changes we made, described above, address Reviewer #1’s points. The remaining points are addressed below.

(6) TTX application induces a significant increase in mEPSC amplitude in Rab3A-/- mice in two out of three data sets (Figs. 1 and 9). Hence, the major conclusion that Rab3A is required for homeostatic scaling is only partially supported by the data. 

The p values based on CDF comparisons were problematic, but the point we were making is that they were much larger for amplitudes measured in cultures prepared from Rab3A-/- mice (Fig. 1, p = 0.04) compared to those from cultures prepared from Rab3A+/+ mice (Fig. 1, p = 4.6 * 10-4). Now that we are comparing means, there are no significant TTX-induced effects on mEPSC amplitudes for Rab3A-/- data. However, acknowledging that some increase after activity blockade remains, we describe homeostatic plasticity as being impaired or not significant, rather than abolished, by loss of Rab3A, (Abstract, lines 37 to 39; Results, lines 141 to 143; Discussion, lines 415 to 418).

(8) There is a significant increase in baseline mEPSC amplitude in Rab3AEbd/Ebd (15 pA) vs. Rab3AEbd/+ (11 pA) cultures, but not in Rab3A-/- (13.6 pA) vs. Rab3A+/- (13.9 pA). Although the nature of scaling was different between Rab3AEbd/Ebd vs. Rab3AEbd/+ and Rab3AEbd/Ebd with vs. without TTX, the question arises whether the increase in mEPSC amplitude in Rab3AEbd/Ebd is Rab3A dependent. Could a Rab3A independent mechanism occlude scaling?

The Reviewer is concerned that the increase in mEPSC amplitude in the presence of the Rab3A point mutant may be through a ‘non-Rab3A’ mechanism (a concern raised by the lack of such effect in cultures from the Rab3A-/- mice), and secondly, that the already large mEPSC cannot be further increased by the homeostatic plasticity mechanism. It must always be considered that a mutant with an altered genetic sequence may bind to novel partners, causing activities that would not be either facilitated or inhibited by the original molecule. We have added this caveat to Results, lines 180 to 186 We added that a number of other manipulations, implicating individual molecules in the homeostatic mechanism, have caused an increase in mEPSC amplitude at baseline, potentially nonspecifically occluding the ability of activity blockade to induce a further increase (Results lines 186 to 189). Still, it is a strong coincidence that the novel activity of the mutant Rab3A would affect mEPSC amplitude, the same characteristic that is affected by activity blockade in a Rab3A dependent manner, a point which we added to Results, lines 189 to 191.

(9) Figure 4: NASPM appears to have a stronger effect on mEPSC frequency in the TTX condition vs. control (-40% vs -15%). A larger sample size might be necessary to draw definitive conclusions on the contribution of Ca2+-permeable AMPARs.

Our results, even with the modest sample size of 11 cells, are clear: NASPM does not disrupt the effect of TTX treatment on mEPSC amplitude (new Fig. 3A). It also looks like there is a greater magnitude effect of NAPSM on frequency in TTX-treated cells; we note this, but point out that nevertheless, these mEPSCs are not contributing to the increase in mEPSC amplitude (Results, lines 238-241). 

(11) The change in GluA2 area or fluorescence intensity upon TTX treatment in controls is modest. How does the GluA2 integral change?

We had reported that GluA2 area showed the most prominent increase following activity blockade, with intensity changing very little. When we examined the integral, it closely matched the change in area. We have added the values for integral to new Fig. 5 D, H; new Fig. 6 A-C; new Fig. 7 A-C and new Table 1 (for GluA2) and new Table 2 (for VGLUT1). These results are described in the text in the following places: Results, lines 289-292; 298-299; 311-319; 328-324). For VGLUT1, both area and intensity changed modestly, and the integral appeared to be a combination of the two, being higher in magnitude and resulting in smaller p values than either area or intensity (Results, lines 344-348; 353-359; new Table 2).

(12) The quantitative comparison between physiology and microscopy data is problematic. The authors report a mismatch in ratio values between the smallest mEPSC amplitudes and the smallest GluA2 receptor cluster sizes (l. 464; Figure 8). Is this comparison affected by the fluorescence intensity threshold? What was the rationale for a threshold of 400 a.u. or 450 a.u.? How does this threshold compare to the mEPSC threshold of 3 pA.

This concern is partially addressed by no longer comparing the rank ordered mEPSC amplitudes with the rank ordered GluA2 receptor characteristics. We had used multiple thresholds in the event that an experiment was not analyzable with the chosen threshold (this in fact happened for VGLUT1, see end of this paragraph). We created box plots of the mean GluA2 receptor cluster size, intensity and integral, for experiments in which we used all three thresholds, to determine if the effect of activity blockade was different depending on which threshold was applied, and found that there was no obvious difference in the results (Author response image 1). Nevertheless, since there is no need to use a different threshold for any of the 6 experiments (3 WT and 3KO), for new Figures 5, 6 and 7 we used the same threshold for all data, 450; described in Methods, lines 746 to 749. For VGLUT1 levels, it was necessary to use a different threshold for Rab3A+/+ Culture #1 (400), but a threshold of 200 for the other five experiments (Methods, lines 751-757). The VGLUT1 immunofluorescent sites in Culture #1 had higher levels overall, and the low threshold caused the entire AOI to be counted as the synapse, which clearly included background levels outside of the synaptic site. Conversely, to use a threshold of 400 on the other experiments meant that the synaptic site found by the automated measurement tool was much smaller that what was visible by eye. In our judgement it would have been meaningless to adhere to a single threshold for VGLUT1 data.

Author response image 1.

Using different thresholds does not substantially alter GluA2 receptor cluster size data. A) Rab3A+/+ Culture #1, size data for three different thresholds, depicted above each graph. B) Rab3A+/+ Culture #2, size data for three different thresholds, depicted above each graph. Note scale bar in A is different from B, to highlight differences for different thresholds. (Culture #3 was only analyzed with 450 threshold).

The conclusion that an increase in AMPAR levels is not fully responsible for the observed mEPSC increase is mainly based on the rank-order analysis of GluA2 intensity, yielding a slope of ~0.9. There are several points to consider here: (i) GluA2 fluorescence intensity did increase on average, as did GluA2 cluster size.

(ii) The increase in GluA2 cluster size is very similar to the increase in mEPSC amplitude (each approx. 1820%). (iii) Are there any reports that fluorescence intensity values are linearly reporting mEPSC amplitudes (in this system)? Antibody labelling efficiency, and false negatives of mEPSC recordings may influence the results. The latter was already noted by the authors.

Our comparison between mEPSC amplitude and GluA2 receptor cluster characteristics has been reexamined in the revised version using means rather than rank-ordered data in rank-order plots or ratio plots. Importantly, all of these methods revealed that in one out of three WT cultures (Culture #3) GluA2 receptor cluster size (old Fig. 8, old Table 1; new Fig. 6, new Table 1), intensity and integral (new Fig. 6, new Table 1) values decreased following activity blockade while in the same culture, mEPSC amplitudes increased. It is based on this lack of correspondence that we conclude that increases in mEPSC amplitude are not fully explained by increases in GluA2 receptors, and suggest there may be other contributors. These points are made in the Abstract (lines 108-110); Results (lines 319 to 326; 330337; 341-343) and the Discussion (lines 472 to 474). To our knowledge, there are not any reports that quantitatively compare receptor levels (area, intensity or integrals) to mEPSC amplitudes in the same cultures. We examined the comparisons very closely for 5 studies that used TTX to block activity and examined receptor levels using confocal imaging at identified synapses (Hou et al., 2008; Ibata et al., 2008; Jakawich et al., 2010a; Xu and Pozzo-Miller, 2017; Dubes et al., 2022). We were specifically looking for whether the receptor data were more variable than the mEPSC amplitude data, as we found. However, for 4 of the studies, sample sizes were very different so that we cannot simply compare the p values. Below is a table of the comparisons.

Author response table 1.

In Xu 2017 the sample sizes are close enough that we feel comfortable concluding that the receptor data were slightly more variable (p < 0.05) than mEPSC data (p<0.01) but recognize that it is speculative to say our finding has been confirmed. A discussion of these articles is in Discussion, lines 456-474.

(iv) It is not entirely clear if their imaging experiments will sample from all synapses. Other AMPAR subtypes than GluA2 could contribute, as could kainite or NMDA receptors.

While our imaging data only examined GluA2, we used the application of NASPM to demonstrate Ca2+permeable receptors did not contribute quantitatively to the increase in mEPSC amplitude following TTX treatment. Since GluA3 and GluA4 are also Ca2+-permeable, the findings in new Figure 3 (old Fig. 4) likely rule out these receptors as well.  There are also reports that Kainate receptors are Ca2+-permeable and blocked by NASPM (Koike et al., 1997; Sun et al., 2009), suggesting the NASPM experiment also rules out the contribution of Kainate receptors. Finally, given our recording conditions, which included normal magnesium levels in the extracellular solution as well as TTX to block action-potential evoked synaptic transmission, NMDA receptors would not be available to contribute currents to our recordings due to block by magnesium ions at resting Vm. These points have been added to the Methods section, lines 617 to 677 (NMDA); 687-694 (Ca2+-permeable AMPA receptors and Kainate receptors).

Furthermore, the statement “complete lack of correspondence of TTX/CON ratios” is not supported by the data presented (l. 515ff). First, under the assumption that no scaling occurs in Rab3A-/-, the TTX/CON ratios show a 20-30% change, which indicates the variation of this readout. Second, the two examples shown in Figure 8 for Rab3A+/+ are actually quite similar (culture #1 and #2, particularly when ignoring the leftmost section of the data, which is heavily affected by the raw values approaching zero.

We are no longer presenting ratio plots in the revised manuscript, so we do not base our conclusion that mEPSC amplitude data is not always corresponding to GluA2 receptor data on the difference in behavior of TTX/CON ratio values, but only on the difference in direction of the TTX effect in one out of three cultures. We agree with the reviewer that the ratio plots are much more sensitive to differences between control and treated values than the rank order plot, and we feel these differences are important, for example, there is still a homeostatic increase in the Rab3A-/- cultures, and the effect is still divergent rather than uniform. But the comparison of ratio data will be presented elsewhere.

(13) Figure 7A: TTX CDF was shifted to smaller mEPSC amplitude values in Rab3A-/- cultures. How can this be explained?

While this result is most obvious in CDF plots, we still observe a trend towards smaller mEPSC amplitudes after TTX treatment in two of three individual cultures prepared from Rab3A-/- mice when comparing means (new Fig. 7, Table 1) which did not reach statistical significance for the pooled data (new Fig. 5, new Table 1). There was not any evidence of this decrease in the larger data set (new Fig. 1) nor for Rab3A-/- neurons on Rab3A+/+ glia (new Fig. 8). Given that this effect is not consistent, we did not comment on it in the revised manuscript. It may be that there is a non-Rab3A-dependent mechanism that results in a decrease in mEPSC amplitude after activity blockade, which normally pulls down the magnitude of the activity-dependent increase typically observed. But studying this second component would be difficult given its magnitude and inconsistent presentation.

Reviewer #1 (Recommendations For the Authors):

(1) Abstract, last sentence: The conclusion of the present manuscript should be primarily based on the results presented. At present, it is mainly based on a previous publication by the authors.

We have revised the last sentence to reflect actual findings of the current study (Abstract, lines 47 to 49).

(2) Line 55: “neurodevelopmental”

This phrase has been removed.

(3) Line 56: “AMPAergic” should be replaced by AMPAR-mediated

This sentence was removed when all references to “scaling” were removed; no other instances of “AMPAergic” are present.

(4) Figure 9: The use of BioRender should be disclosed in the Figure Legend.

We used BioRender in new Figures 3, 7 and 8, and now acknowledge BioRender in those figure legends.

(5) Figure legends and results: The number of cultures should be indicated for each comparison.

Number of cultures has been added to the figure legends.

(6) Line 289: A comparison of p-values between conditions does not allow any meaningful conclusions.

Agreed, therefore we have removed CDFs and the KS test comparison p values. All comparisons in the revised manuscript are for cell means.

(7) Line 623ff: The argument referring to NMJ data is weak, given that different types of receptors are involved.

We still think it is valid to point out that Rab3A is required for the increase in mEPC at the NMJ but that ACh receptors do not increase (Discussion, lines 522 to 525). We are not saying that postsynaptic receptors do not contribute in cortical cultures, only that there could be another Rab3A-dependent mechanism that also affects mEPSC amplitude.

(8) Plotting data points outside of the ranges should be avoided (e.g., Fig. 2Giii, 7F).

These two figures are no longer present in the revised manuscript. In revising figures, we made sure no other plots have data points outside of the ranges.

(9) The rationale for investigating Rab3AEbd/Ebd remains elusive and should be described.

A rationale for investigating Rab3AEbd/Ebd is that if the results are similar to the KO, it strengthens the evidence for Rab3A being involved in homeostatic synaptic plasticity. In addition, since its phenotype of early awakening was stronger than that demonstrated in Rab3A KO mice (Kapfhamer et al., 2002), it was possible we would see a more robust effect. These points have been added to the Results, lines 118 to 126.

(10) Figures 3 and 4, as well as Figure 5 and 6 could be merged.

In the revised version, Figure 3 has been eliminated since its main point was a difference in scaling behavior. Figure 4 has been expanded to include a model of how NASPM could reduce frequency (new Fig. 3.) Images of the pyramidal cell body have been added to Figure 5 (new Fig. 4), and Figure 6 has been completely revised and now includes pooled data for both Rab3A+/+ and Rab3A-/- cultures, for mEPSC amplitude, GluA2 receptor cluster size, intensity and integral.

(11) Figure 5: The legend refers to MAP2, but this is not indicated in the figure.

MAP2 has now been added to the labels for each image and described in the figure legend (new Fig. 4).

Reviewer #2:

Technical concerns:

(1) The culture condition is questionable. The authors saw no NMDAR current present during spontaneous recordings, which is worrisome since NMDARs should be active in cultures with normal network activity (Watt et al., 2000; Sutton et al., 2006). It is important to ensure there is enough spiking activity before doing any activity manipulation. Similarly it is also unknown whether spiking activity is normal in Rab3AKO/Ebd neurons.

In the studies cited by the reviewer, NMDA currents were detected under experimental conditions in which magnesium was removed. In our recordings, we have normal magnesium (1.3 mM) and also TTX, which prevents the necessary depolarization to allow inward current through NMDA receptors. This point has been added to our Methods, lines 674 to 677. We acknowledge we do not know the level of spiking in cultures prepared from Rab3A+/+, Rab3A-/- or Rab3A_Ebd/Ebd_ mice. Given the similar mEPSC amplitude for untreated cultures from WT and KO studies, we think it unlikely that activity was low in the latter, but it remains a possibility for untreated cultures from Rab3A_Ebd/Ebd_ mice, where mEPSC amplitude was increased. These points are added to the Methods, lines 615 to 622.

(2) Selection of mEPSC events is not conducted in an unbiased manner. Manually selecting events is insufficient for cumulative distribution analysis, where small biases could skew the entire distribution. Since the authors claim their ratio plot is a better method to detect the uniformity of scaling than the well-established rank-order plot, it is important to use an unbiased population to substantiate this claim.

We no longer include any cumulative distributions or ratio plot analysis in the revised version. We have added the following text to Methods, lines 703 to 720:

“MiniAnalysis selects many false positives with the automated feature when a small threshold amplitude value is employed, due to random fluctuations in noise, so manual re-evaluation of the automated process is necessary to eliminate false positives. If the threshold value is set high, there are few false positives but small amplitude events that visually are clearly mEPSCs are missed, and manual re-evaluation is necessary to add back false negatives or the population ends up biased towards large mEPSC amplitudes. As soon as there is a manual step, bias is introduced. Interestingly, a manual reevaluation step was applied in a recent study that describes their process as ‘unbiased (Wu et al., 2020). In sum, we do not believe it is currently possible to perform a completely unbiased detection process. A fully manual detection process means that the same criterion (“does this look like an mEPSC?”) is applied to all events, not just the false positives, or the false negatives, which prevents the bias from being primarily at one end or the other of the range of mEPSC amplitudes. It is important to note that when performing the MiniAnalysis process, the researcher did not know whether a record was from an untreated cell or a TTX-treated cell.”

(3) Immunohistochemistry data analysis is problematic. The authors only labeled dendrites without doing cell-fills to look at morphology, so it is questionable how they differentiate branches from pyramidal neurons and interneurons. Since glutamatergic synapse on these two types of neuron scale in the opposite directions, it is crucial to show that only pyramidal neurons are included for analysis.

We identified neurons with a pyramidal shape and a prominent primary dendrite at 60x magnification without the zoom feature. This should have been made clear in the description of imaging. We have added an image of the two selected cells to our figure of dendrites (old Fig. 5, new Fig. 4), and described this process in the Methods, lines 736 to 739, and Results, lines 246 to 253. Given the morphology of the neurons selected it is highly unlikely that the dendrites we analyzed came from interneurons.

Conceptual Concerns

The only novel finding here is the implicated role for Rab3A in synaptic scaling, but insights into mechanisms behind this observation are lacking. The authors claim that Rab3A likely regulates scaling from the presynaptic side, yet there is no direct evidence from data presented. In its current form, this study’s contribution to the field is very limited.

We have demonstrated that loss of Rab3A and expression of a Rab3A point mutant disrupt homeostatic plasticity of mEPSC amplitudes, and that in the absence of Rab3A, the increase in GluA2 receptors at synaptic sites is abolished. Further, we show that this effect cannot be through release of a factor, like TNFα, from astrocytes. In the new version, we add the finding that VGLUT1 is not increased after activity blockade, ruling out this presynaptic factor as a contributor to homeostatic increases in mEPSC amplitude. We show for the first time by examining mEPSC amplitudes and GluA2 receptors in the same cultures that the increases in GluA2 receptors are not as consistent as the increases in mEPSC amplitude, suggesting the possibility of another contributor to homeostatic increases in mEPSC amplitude. We first proposed this idea in our previous study of Rab3A-dependent homeostatic increases in mEPC amplitudes at the mouse neuromuscular junction. In sum, we dispute that there is only one novel finding and that we have no insights into mechanism. We acknowledge that we have no direct evidence for regulation from the presynaptic side, and have removed this claim from the revised manuscript. We have retained the Discussion of potential mechanisms affecting the presynaptic quantum and evidence that Rab3A is implicated in these mechanisms (vesicle size, fusion pore kinetics; Discussion, lines 537 to 563). One way to directly show that the amount of transmitter released for an mEPSC has been modified after activity blockade is to demonstrate that a fast off-rate antagonist has become less effective at inhibiting mEPSCs (because the increased glutamate released out competes it; see (Liu et al., 1999) and (Wilson et al., 2005) for example experiments). This set of experiments is underway but will take more time than originally expected, because we are finding surprisingly large decreases in frequency, possibly the result of mEPSCs with very low glutamate concentration that are completely inhibited by the dose used. Once mEPSCs are lost, it is difficult to compare the mEPSC amplitude before and after application of the antagonist. Therefore we intend to include this experiment in a future report, once we determine the reason for the frequency reduction, or, can find a dose where this does not occur.

(1) Their major argument for this is that homeostatic effects on mEPSC amplitudes and GluA2 cluster sizes do not match. This is inconsistent with reports from multiple labs showing that upscaling of mEPSC amplitude and GluA2 accumulation occur side by side during scaling (Ibata et al., 2008; Pozo et al., 2012; Tan et al., 2015; Silva et al., 2019). Further, because the acquisition and quantification methods for mEPSC recordings and immunohistochemistry imaging are entirely different (each with its own limitations in signal detection), it is not convincing that the lack of proportional changes must signify a presynaptic component.

Within the analyses in the revised manuscript, which are now based only on comparison of cell/dendrite means, we find a very good match in the magnitude of increase for the pooled data of mEPSC amplitudes and GluA2 receptor cluster sizes (+19.7% and +20.0% respectively; new Table 1). However, when looking at individual cultures, we had one of three WT cultures in which mEPSC amplitude increased 17.2% but GluA2 cluster size decreased 9.5%. This result suggests that while activity blockade does lead to an increase in GluA2 receptors after activity blockade, the effect is more variable than that for mEPSC amplitude. We went back to published studies to see if this has been previously observed, but found that it was difficult to compare because the sample sizes were different for the two characteristics (see Author response table 1). We included these particular 5 studies because they use the same treatment (TTX), examine receptors using imaging of identified synaptic sites, and record mEPSCs in their cultures (although the authors do not indicate that imaging and recordings are done simultaneously on the same cultures.) Only one of the studies listed by the Reviewer is in our group (Ibata et al., 2008). The study by (Tan et al., 2015) uses western blots to measure receptors; the study by (Silva et al., 2019) blocks activity using a combination of AMPA and NMDA receptor blockers; the study by (Pozo et al., 2012) correlates mEPSC amplitude changes with imaging but not in response to activity blockade, instead for changing the expression of GluA2. While it may seem like splitting hairs to reject studies that use other treatment protocols, there is ample evidence that the mechanisms of homeostatic plasticity depend on how activity was altered, see the following studies for several examples of this (Sutton et al., 2006; Soden and Chen, 2010; Fong et al., 2015). A discussion of the 5 articles we selected is in the revised manuscript, Discussion, lines 456 to 474. In sum, we provide evidence that activity blockade is associated with an overall increase in GluA2 receptors; what we propose is that this increase, being more variable, does not fully explain the increase in mEPSC amplitude. However, we acknowledge that the disparity could be explained by the differences in limitations of the two methods (Discussion, lines 469-472).

(2) The authors also speculate in the discussion that presynaptic Rab3A could be interacting with retrograde BDNF signaling to regulate postsynaptic AMPARs. Without data showing Rab3A-dependent presynaptic changes after TTX treatment, this argument is not compelling. In this retrograde pathway, BDNF is synthesized in and released from dendrites (Jakawich et al., 2010b; Thapliyal et al., 2022), and it is entirely possible for postsynaptic Rab3A to interfere with this process cell-autonomously.

We have added the information that Rab3A could control BDNF from the postsynaptic cell and included the two references provided by the reviewer, Discussion, lines 517 to 518. We have added new evidence, recently published, that the Rab3 family has been shown to regulate targeting of EGF receptors to rafts (among other plasma membrane molecules), with Rab3A itself clearly present in nonneuronal cells (Diaz-Rohrer et al., 2023) (added to Discussion, lines 509 to 515).

(3) The authors propose that a change in AMPAR subunit composition from GluA2-containing ones to GluA1 homomers may account for the distinct changes in mEPSC amplitudes and GluA2 clusters. However, their data from the NASPM wash-in experiments clearly show that the GluA1 homomer contributions have not changed before and after TTX treatment.

We have revised this section in the Discussion, lines 534 to 536, to clarify that any change due to GluA1 homomers should have been detectable by a greater ability of NASPM to reverse the TTX-induced increase.

Reviewer #2 (Recommendations for the Authors):

For authors to have more convincing arguments in general, they will need to clarify/improve certain details in their data collection by addressing the above technical concerns. Additionally, the authors should design experiments to test whether Rab3A regulates scaling from pre- or post-synaptic site. For example, they could sparsely knock out Rab3A in WT neurons to test the postsynaptic possibility. On the other hand, their argument for a presynaptic role would be much more compelling if they could show whether there are clear functional changes such as in vesicle sizes and release probability in the presynaptic terminal of Rab3AKO neurons.

An important next step is to identify whether Rab3A is acting pre- or post-synaptically (Discussion, lines 572 to 573), but these experiments will be undertaken in the future. It would not add much to simply show vesicle size is altered in the KO (and we do not necessarily expect this since mEPSC amplitude is normal in the KO). It will be very difficult to establish that vesicle size is changing with activity blockade and that this change is prevented in the Rab3A KO, because we are looking for a ~25% increase in vesicle volume, which would correspond to a ~7.5% increase in diameter. Finally, we do not believe demonstrating changes in release probability tell us anything about a presynaptic role for Rab3A in regulating the size of the presynaptic quantum.

Reviewer #3 (Public Review)

Weaknesses: However, the rather strong conclusions on the dissociation of AMPAR trafficking and synaptic response are made from somewhat weaker data. The key issue is the GluA2 immunostaining in comparison with the mEPSC recordings. Their imaging method involves only assessing puncta clearly associated with a MAP2 labeled dendrite. This is a small subset of synapses, judging from the sample micrographs (Fig. 5). To my knowledge, this is a new and unvalidated approach that could represent a particular subset of synapses not representative of the synapses contributing to the mEPSC change (they are also sampling different neurons for the two measurements; an additional unknown detail is how far from the cell body were the analyzed dendrites for immunostaining.) While the authors acknowledge that a sampling issue could explain the data, they still use this data to draw strong conclusions about the lack of AMPAR trafficking contribution to the mEPSC amplitude change. This apparent difference may be a methodological issue rather than a biological one, and at this point it is impossible to differentiate these. It will unfortunately be difficult to validate their approach. Perhaps if they were to drive NMDAdependent LTD or chemLTP, and show alignment of the imaging and ephys, that would help. More helpful would be recordings and imaging from the same neurons but this is challenging. Sampling from identified synapses would of course be ideal, perhaps from 2P uncaging combined with SEP-labeled AMPARs, but this is more challenging still. But without data to validate the method, it seems unwarranted to make such strong conclusions such as that AMPAR trafficking does not underlie the increase in mEPSC amplitude, given the previous data supporting such a model.

In the new version, we soften our conclusion regarding the mismatch between GluA2 receptor levels and mEPSC amplitudes, now only stating that receptors may not be the sole contributor to the TTX effect on mEPSC amplitude (Discussion, lines 472 to 474). With our analysis in the new version focusing on comparisons of cell means, the GluA2 receptor cluster size and the mEPSC amplitude data match well in magnitude for the data pooled across the 3 matched cultures (20.0% and 19.7%, respectively, see new Table 1). However, in one of the three cultures the direction of change for GluA2 receptors is opposite that of mEPSC amplitudes (Table 1, Culture #3, -9.5% vs +17.2%, respectively).

It is unlikely that the lack of matching of homeostatic plasticity in one culture, but very good matching in two other cultures, can be explained by an unvalidated focus on puncta associated with MAP2 positive dendrites. We chose to restrict analysis of synaptic GluA2 receptors to the primary dendrite in order to reduce variability, reasoning that we are always measuring synapses for an excitatory pyramidal neuron, synapses that are relatively close to the cell body, on the consistently identifiable primary dendrite. We measured how far this was for the two cells depicted in old Figure 5 (new Fig. 4). Because we always used the 5X zoom window which is a set length, and positioned it within ~10 microns of the cell body, these cells give a ball park estimate for the usual distances. For the untreated cell, the average distance from the cell body was 38.5 ± 2.8 µm; for the TTX-treated cell, it was 42.4 ± 3.2 µm (p = 0.35, KruskalWallis test). We have added these values to the Results, lines 270 to 274.

We did not mean to propose that AMPA receptor levels do not contribute at all to mEPSC amplitude, and we acknowledge there are clear cases where the two characteristics change in parallel (for example, in the study cited by Reviewer #2, (Pozo et al., 2012), increases in GluA2 receptors due to exogenous expression are closely matched by increases in mEPSC amplitudes.) What our matched culture experiments demonstrate is that in the case of TTX treatment, both GluA2 receptors and mEPSC amplitudes increase on average, but sometimes mEPSC amplitudes can increase in the absence of an increase in GluA2 receptors (Culture #3, Rab3A+/+ cultures), and sometimes mEPSC amplitudes do not increase even though GluA2 receptor levels do increase (Culture #3, Rab3A-/- cultures). Therefore, it would not add anything to our argument to examine receptors and mEPSCs in NMDA-dependent LTP, a different plasticity paradigm in which changes in receptors and mEPSCs may more closely align. It has been demonstrated that mEPSCs of widely varying amplitude can be recorded from a single synaptic site (Liu and Tsien, 1995), so we would need to measure a large sample of individual synapse recordings to detect a modest shift in average values due to activity blockade. In addition, it would be essential to express fluorescent AMPA receptors in order to correlate receptor levels in the same cells we record from (or at the same synapses). And yet, even after these heroics, one is still left with the issue that the two methods, electrophysiology and fluorescent imaging, have distinct limitations and sources of variability that may obscure any true quantitative correlation.

Other questions arise from the NASPM experiments, used to justify looking at GluA2 (and not GluA1) in the immunostaining. First, there is a frequency effect that is quite unclear in origin. One would expect NASPM to merely block some fraction of the post-synaptic current, and not affect pre-synaptic release or block whole synapses. It is also unclear why the authors argue this proves that NASPM was at an effective concentration (lines 399-400). Further, the amplitude data show a strong trend towards smaller amplitude. The p value for both control and TTX neurons was 0.08 – it is very difficult to argue that there is no effect. And the decrease is larger in the TTX neurons. Considering the strong claims for a presynaptic locus and the use of this data to justify only looking at GluA2 by immunostaining, these data do not offer much support of the conclusions. Between the sampling issues and perhaps looking at the wrong GluA subunit, it seems premature to argue that trafficking is not a contributor to the mEPSC amplitude change, especially given the substantial support for that hypothesis. Further, even if trafficking is not the major contributor, there could be shifts in conductance (perhaps due to regulation of auxiliary subunits) that does not necessitate a pre-synaptic locus. While the authors are free to hypothesize such a mechanism, it would be prudent to acknowledge other options and explanations.

We have created a model cartoon to explain how NASPM could reduce mEPSC frequency (new Fig. 3D). mEPSCs that arise from a synaptic site that has only Ca2+-permeable AMPA receptors will be completely blocked by NASPM, if the NASPM concentration is maximal. The reason we conclude that we have sufficient NASPM reaching the cells is that the frequency is decreased, as expected if there are synaptic sites with only Ca2+-permeable AMPA receptors. We previously were not clear that there is an effect of NASPM on mEPSC amplitude, although it did not reach statistical significance (new Fig. 3B). Where there is no effect is on the TTX-induced increase in mEPSC amplitude, which remains after the acute NASPM application (new Fig. 3A). We have revised the description of these findings in Results, lines 220 to 241. In reviewing the literature further, we could find no previous studies demonstrating an increase in conductance in GluA2 or Ca2+-impermeable receptors, only in GluA1 homomers. In other words, any conductance change would have been due to a change in GluA1 homomers, and should have been visible as a disruption of the homeostatic plasticity by NASPM application. We have added text to Results, lines 211 to 217; 236-241; Discussion, lines 420 to 422; 526-536 and Methods, lines 685 to 695 regarding this point.

The frequency data are missing from the paper, with the exception of the NASPM dataset. The mEPSC frequencies should be reported for all experiments, particularly given that Rab3A is generally viewed as a pre-synaptic protein regulating release. Also, in the NASPM experiments, the average frequency is much higher in the TTX treated cultures. Is this statistically above control values?

This comment is addressed by the major change #3, above.

Unaddressed issues that would greatly increase the impact of the paper:

(1) Is Rab3A activity pre-synaptically, post-synaptically or both. The authors provide good evidence that Rab3A is acting within neurons and not astrocytes. But where is it acting (pre or post) would aid substantially in understanding its role (and particularly the hypothesized and somewhat novel idea that the amount of glutamate released per vesicle is altered in HSP). They could use sparse knockdown of Rab3A, or simply mix cultures from KO and WT mice (with appropriate tags/labels). The general view in the field has been that HSP is regulated post-synaptically via regulation of AMPAR trafficking, and considerable evidence supports this view. The more support for their suggestion of a pre-synaptic site of control, the better.

This is similar to the request of Reviewer #2, Recommendations to the Authors. An important next step is to identify whether Rab3A is working pre- or postsynaptically. However, it is possible that it is acting pre-synaptically to anterogradely regulate trafficking of AMPAR, as we have depicted in our model, new Fig. 9. To demonstrate that the presynaptic quantum is being altered, we would need to show that vesicle size is increased, or the amount of transmitter being released during an mEPSC is increased after activity blockade. To that end, we are currently performing experiments using a fast off-rate antagonist. As described above in response to Reviewer #2’s Conceptual Concerns, we find dramatic decreases in frequency not explained by the 30-60% inhibition observed for the largest amplitude mEPSCs, which suggests the possibility that small mEPSCs are more sensitive than large mEPSCs and therefore may have less transmitter. Due to these complexities and the delay while we test other antagonists to see if the effect is specific to fast-off rate antagonists, we are not including these results here.

(2) Rab3A is also found at inhibitory synapses. It would be very informative to know if HSP at inhibitory synapses is similarly affected. This is particularly relevant as at inhibitory synapses, one expects a removal of GABARs and/or a decrease of GABA-packaging in vesicles (ie the opposite of whatever is happening at excitatory synapses.). If both processes are regulated by Rab3A, this might suggest a role for this protein more upstream in the signaling, an effect only at excitatory synapses would argue for a more specific role just at these synapses.

It will be important to determine if homeostatic synaptic plasticity at inhibitory synapses on excitatory neurons is sensitive to Rab3A deletion, especially in light of the fact that unlike many of the other molecules implicated in homeostatic increases in mEPSCS, Rab3A is not a molecule known to be selective for glutamate receptor trafficking (in contrast to Arc/Arg3.1 or GRIP1, for example). Such a study would warrant its own publication.

Reviewer #3 (Recommendations for the Authors):

There are a number of minor points or suggestions for the authors:

Is RIM1 part of this pathway (or expected to be)? Some discussion of this would be nice.

RIM, Rab3-interacting molecule, has been implicated at the drosophila neuromuscular junction in a presynaptic form of homeostatic synaptic plasticity in which evoked release is increased after block of postsynaptic receptors (Muller et al., 2012), a plasticity that also requires Rab3-GAP (Muller et al., 2011). To our knowledge there is no evidence that RIM is involved in the homeostatic plasticity of mEPSC amplitude after activity blockade by TTX. The Rim1a KO does not have a change in mEPSC amplitude relative to WT (Calakos et al., 2004), but that is not unexpected given the normal mEPSC amplitude in neurons from cultures prepared from Rab3A-/- mice in the current study. It would be interesting to look at homeostatic plasticity in cortical cultures prepared from Rim1a or other RIM deletion mice, but we have not added these points to the revised manuscript since there are a number of directions one could go in attempting to define the molecular pathway and we feel it is more important to discuss the potential location of action and physiological mechanisms.

Is the Earlybird mutation a GOF? More information about this mutation would help.

We have added a description of how the Earlybird mutation was identified, in a screen for rest:activity mutants (Results, lines 118 to 123). Rab3A Earlybird mice have a shortened circadian period, shifting their wake cycle earlier and earlier. When Rab3A deletion mice were tested in the same activity raster plot measurements, the shift was smaller than that for the Earlybird mutant, suggesting the possibility that it is a dominant negative mutation.

The high K used in the NASPM experiments seems a bit unusual. Have the authors done high K/no drug controls to see if this affects the synapses in any way?

We used the high K based on previous studies that indicated the blocking effect of the Ca2+-permeable receptor blockers was use dependent (Herlitze et al., 1993; Iino et al., 1996; Koike et al., 1997). We reasoned that a modest depolarization would increase the frequency of AMPA receptor mEPSCs and allow access of the NASPM.  We have added this point to the Methods, lines 695 to 708. 

The NASPM experiments do not show that GluA1 does not contribute (line 401), only that GluA1 homomers are not contributing (much – see above). GluA1/A2 heteromers are quite likely involved. Also, the SEM is missing from the WT pre/post NASPM data.

Imaging of GluA2-positive sites will not distinguish between GluA2 homomers and GluA2-GluA1 heteromers, so we have added this clarification to Results, lines 242 to 246. We have remade the NASPM pre-post line plots so that the mean values and error bars are more visible (new Fig. 3B, C).

It seems odd to speculate based on non-significant findings (line 650-1), with lower significance (p = 0.11) than findings being dismissed in the paper (NASPM on mEPSC amplitude; p = 0.08).

We did not mean to dismiss the effect of NASPM on mEPSC amplitude (new Fig. 3B), rather, we dismiss the effect of NASPM on the homeostatic increase in mEPSC amplitude caused by TTX treatment (new Fig. 3A). We have emphasized this distinction in Results, lines 223 to 225, and Discussion, lines 420 to 422, as well as adding that the stronger effect of NASPM on frequency after TTX treatment suggests an activity-dependent increase in the number of synapses expressing only Ca2+ permeable homomers (Results, lines 236 to 241; Discussion, lines 431 to 435).

Fig. 4 could be labeled better (to make it clear that B is amplitude and C is freq from the same cells).

Fig. 4 has been revised—now the amplitude and frequency plots from the same condition (new Fig. 3, B, C; CON or TTX) are in a vertical line and the figure legend states that the frequency data are from the same cells as in Fig. 3A.

The raw amplitude data seems a bit hidden in the inset panels – I would suggest these data are at least as important as the cumulative distributions in the main panel. Maybe re-organizing the figures would help.

We have removed all cumulative distributions, rank order plots, and ratio plots. The box plots are now full size in new Figures 1, 2, 5, 6, 7 and 8.

I’m not sure I would argue in the paper that 12 cells a day is a limiting issue for experiments. It doesn’t add anything and doesn’t seem like that high a barrier. It is fine to just say it is difficult and therefore there is a limited amount of data meeting the criteria.

We have removed the comment regarding difficulty.

Calakos N, Schoch S, Sudhof TC, Malenka RC (2004) Multiple roles for the active zone protein RIM1alpha in late stages of neurotransmitter release. Neuron 42:889-896.

De Gois S, Schafer MK, Defamie N, Chen C, Ricci A, Weihe E, Varoqui H, Erickson JD (2005) Homeostatic scaling of vesicular glutamate and GABA transporter expression in rat neocortical circuits. J Neurosci 25:7121-7133.

Diaz-Rohrer B, Castello-Serrano I, Chan SH, Wang HY, Shurer CR, Levental KR, Levental I (2023) Rab3 mediates a pathway for endocytic sorting and plasma membrane recycling of ordered microdomains. Proc Natl Acad Sci U S A 120:e2207461120.

Dubes S, Soula A, Benquet S, Tessier B, Poujol C, Favereaux A, Thoumine O, Letellier M (2022) miR-124dependent tagging of synapses by synaptopodin enables input-specific homeostatic plasticity. EMBO J 41:e109012.

Fong MF, Newman JP, Potter SM, Wenner P (2015) Upward synaptic scaling is dependent on neurotransmission rather than spiking. Nat Commun 6:6339.

Herlitze S, Raditsch M, Ruppersberg JP, Jahn W, Monyer H, Schoepfer R, Witzemann V (1993) Argiotoxin detects molecular differences in AMPA receptor channels. Neuron 10:1131-1140.

Hou Q, Zhang D, Jarzylo L, Huganir RL, Man HY (2008) Homeostatic regulation of AMPA receptor expression at single hippocampal synapses. Proc Natl Acad Sci U S A 105:775-780.

Ibata K, Sun Q, Turrigiano GG (2008) Rapid synaptic scaling induced by changes in postsynaptic firing. Neuron 57:819-826.

Iino M, Koike M, Isa T, Ozawa S (1996) Voltage-dependent blockage of Ca(2+)-permeable AMPA receptors by joro spider toxin in cultured rat hippocampal neurones. J Physiol 496 ( Pt 2):431437.

Jakawich SK, Neely RM, Djakovic SN, Patrick GN, Sutton MA (2010a) An essential postsynaptic role for the ubiquitin proteasome system in slow homeostatic synaptic plasticity in cultured hippocampal neurons. Neuroscience 171:1016-1031.

Jakawich SK, Nasser HB, Strong MJ, McCartney AJ, Perez AS, Rakesh N, Carruthers CJ, Sutton MA (2010b) Local presynaptic activity gates homeostatic changes in presynaptic function driven by dendritic BDNF synthesis. Neuron 68:1143-1158.

Kapfhamer D, Valladares O, Sun Y, Nolan PM, Rux JJ, Arnold SE, Veasey SC, Bucan M (2002) Mutations in Rab3a alter circadian period and homeostatic response to sleep loss in the mouse. Nat Genet 32:290-295.

Koike M, Iino M, Ozawa S (1997) Blocking effect of 1-naphthyl acetyl spermine on Ca(2+)-permeable AMPA receptors in cultured rat hippocampal neurons. Neurosci Res 29:27-36.

Liu G, Tsien RW (1995) Properties of synaptic transmission at single hippocampal synaptic boutons. Nature 375:404-408.

Liu G, Choi S, Tsien RW (1999) Variability of neurotransmitter concentration and nonsaturation of postsynaptic AMPA receptors at synapses in hippocampal cultures and slices. Neuron 22:395409.

Muller M, Pym EC, Tong A, Davis GW (2011) Rab3-GAP controls the progression of synaptic homeostasis at a late stage of vesicle release. Neuron 69:749-762.

Muller M, Liu KS, Sigrist SJ, Davis GW (2012) RIM controls homeostatic plasticity through modulation of the readily-releasable vesicle pool. J Neurosci 32:16574-16585.

Pozo K, Cingolani LA, Bassani S, Laurent F, Passafaro M, Goda Y (2012) beta3 integrin interacts directly with GluA2 AMPA receptor subunit and regulates AMPA receptor expression in hippocampal neurons. Proc Natl Acad Sci U S A 109:1323-1328.

Silva MM, Rodrigues B, Fernandes J, Santos SD, Carreto L, Santos MAS, Pinheiro P, Carvalho AL (2019) MicroRNA-186-5p controls GluA2 surface expression and synaptic scaling in hippocampal neurons. Proc Natl Acad Sci U S A 116:5727-5736.

Soden ME, Chen L (2010) Fragile X protein FMRP is required for homeostatic plasticity and regulation of synaptic strength by retinoic acid. J Neurosci 30:16910-16921.

Sun HY, Bartley AF, Dobrunz LE (2009) Calcium-permeable presynaptic kainate receptors involved in excitatory short-term facilitation onto somatostatin interneurons during natural stimulus patterns. J Neurophysiol 101:1043-1055.

Sutton MA, Ito HT, Cressy P, Kempf C, Woo JC, Schuman EM (2006) Miniature neurotransmission stabilizes synaptic function via tonic suppression of local dendritic protein synthesis. Cell 125:785-799.

Tan HL, Queenan BN, Huganir RL (2015) GRIP1 is required for homeostatic regulation of AMPAR trafficking. Proc Natl Acad Sci U S A 112:10026-10031.

Thapliyal S, Arendt KL, Lau AG, Chen L (2022) Retinoic acid-gated BDNF synthesis in neuronal dendrites drives presynaptic homeostatic plasticity. Elife 11.

Wilson NR, Kang J, Hueske EV, Leung T, Varoqui H, Murnick JG, Erickson JD, Liu G (2005) Presynaptic regulation of quantal size by the vesicular glutamate transporter VGLUT1. J Neurosci 25:62216234.

Wu YK, Hengen KB, Turrigiano GG, Gjorgjieva J (2020) Homeostatic mechanisms regulate distinct aspects of cortical circuit dynamics. Proc Natl Acad Sci U S A 117:24514-24525.

Xu X, Pozzo-Miller L (2017) EEA1 restores homeostatic synaptic plasticity in hippocampal neurons from Rett syndrome mice. J Physiol 595:5699-5712.

Author response:

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

eLife assessment This valuable paper reports a theoretical framework and methodology for identifying Cancer Driving Nucleotides (CDNs), primarily based on single nucleotide variant (SNV) frequencies. A variety of solid approaches indicate that a mutation recurring three or more times is more likely to reflect selection rather than being the consequence of a mutation hotspot. The method is rigorously quantitative, though the requirement for larger datasets to fully identify all CDNs remains a noted limitation. The work will be of broad interest to cancer geneticists and evolutionary biologists. 

The key criticism “the requirement for larger datasets to fully identify all CDNs remains a noted limitation” that is also found in both reviews. We have clarified the issue in the main text, the relevant parts, from which are copied below. The response below also addresses many comments in the reviews. In addition, Discussion of eLife-RP-RA-2024-99341 has been substantially expanded to answer the questions of Reviewer 2.

We shall answer the boldface comment in three ways. First, it can be answered using GENIE data. Fig. 7 of the main text (eLife-RP-RA-2024-99340) shows that, when n increases from ~ 1000 to ~ 9,000, the numbers of discovered CDNs increase by 3 – 5 fold, most of which come from the two-hit class. Hence, the power of discovering more CDNs with larger datasets is evident. By extrapolation, a sample size of 100,000 should be able to yield 90% of all CDNs, as calculated here. (Fig. 7 also addresses the queries of whether we have used datasets other than TCGA. We indeed have used all public data, including GENIE and COSMIC.) 

Second, the power of discovering more cancer driver genes by our theory is evident even without using larger datasets. Table 3 of the companion study (eLife-RP-RA-2024-99341) shows that, averaged across cancer types, the conventional method would identify 45 CDGs while the CDN method tallies 258 CDGs. The power of the CDN method is demonstrated. This is because the conventional approach has to identify CDGs (cancer driver genes) in order to identify the CDNs they carry. However, many CDNs occur in non-CDGs and are thus missed by the conventional approach. In Supplementary File S2, we have included a full list of CDNs discovered in our study, along with population allele frequency annotations from gnomAD. The distribution patterns of these CDNs across different cancer types show their pan-cancer properties as further explored in the companion paper.

Third, while many, or even most CDNs occur in non-CDGs and are thus missed, the conventional approach also includes non-CDN mutations in CDGs. This is illustrated in Fig. 5 of the companion study (eLife-RP-RA-2024-99341) that shows the adverse effect of misidentifications of CDNs by the conventional approach. In that analysis, the gene-targeting therapy is effective if the patient has the CDN mutations on EGFR, but the effect is reversed if the EGFR mutations are non-CDN mutations.

Reviewer #1 (Public Review):

The authors developed a rigorous methodology for identifying all Cancer Driving Nucleotides (CDNs) by leveraging the concept of massively repeated evolution in cancer. By focusing on mutations that recur frequently in pan-cancer, they aimed to differentiate between true driver mutations and neutral mutations, ultimately enhancing the understanding of the mutational landscape that drives tumorigenesis. Their goal was to call a comprehensive catalogue of CDNs to inform more effective targeted therapies and address issues such as drug resistance.

Strengths

(1) The authors introduced a concept of using massively repeated evolution to identify CDNs. This approach recognizes that advantageous mutations recur frequently (at least 3 times) across cancer patients, providing a lens to identify true cancer drivers.

(2) The theory showed the feasibility of identifying almost all CDNs if the number of sequenced patients increases to 100,000 for each cancer type.

Weaknesses

(1) The methodology remains theoretical and no novel true driver mutations were identified in this study.

We now address the weakness criticism, which is gratefully received.

The second part of the criticism (no novel true driver mutations were identified in this study) has been answered in the long responses to eLife assessment above. The first part “The methodology remains theoretical” is somewhat unclear. It might be the lead to the second part. However, just in case, we interpret the word “theoretical” to mean “the lack of experimental proof” and answer below.

As Reviewer #1 noted, a common limitation of theoretical and statistical analyses of cancer drivers is the need to validate their selective advantage through in vitro or in vivo functional testing. This concern is echoed by both reviewers in the companion paper (eLife-RP-RA-2024-99341), prompting us to consider the methodology for functional testing of potential cancer drivers. An intuitive approach would involve introducing putative driver mutations into normal cells and observing phenotypic transformation in vitro and in vivo. In a recent stepwise-edited human melanoma model, Hodis et al. demonstrated that disease-relevant phenotypes depend on the “correct” combinations of multiple driver mutations (Hodis et al. 2022). Other high-throughput strategies can be broadly categorized into two approaches: (1) introducing candidate driver mutations into pre-malignant model systems that already harbor a canonical mutant driver (Drost and Clevers 2018; Grzeskowiak et al. 2018; Michels et al. 2020) and (2) introducing candidate driver mutations into growth factor-dependent cell models and assessing their impact on resulting fitness (Bailey et al. 2018; Ng et al. 2018). The underlying assumption of these strategies is that the fitness outcomes of candidate driver mutations are influenced by pre-existing driver mutations and the specific pathways or cancer hallmarks being investigated. This confines the functional test of potential cancer driver mutations to conventional cancer pathways. A comprehensive identification of CDNs is therefore crucial to overcome these limitations. In conjunction with other driver signal detection methods, our study aims to provide a more comprehensive profile of driver mutations, thereby enabling the functional testing of drivers involved in non-conventional cancer evolution pathways.

(2) Different cancer types have unique mutational landscapes. The methodology, while robust, might face challenges in uniformly identifying CDNs across various cancers with distinct genetic and epigenetic contexts.

We appreciate the comment. Indeed, different cancer types should have different genetic and epigenetic landscapes. In that case, one may have expected CDNs to be poorly shared among cancer types. However, as reported in Fig. 4 of the companion study, the sharing of CDNs across cancer types is far more common than the sharing of CDGs (Cancer Driving Genes). We suggest that CDNs have a much higher resolution than CDGs, whereby the signals are diluted by non-driver mutations. In other words, despite that the mutational landscape may be cancer-type specific, the pan-cancer selective pressure may be sufficiently high to permit the detection of CDN sharing among cancer types.

Below, we shall respond in greater details. Epigenetic factors, such as chromatin states, methylation/acetylation levels, and replication timing, can provide valuable insights when analyzing mutational landscapes at a regional scale (Stamatoyannopoulos et al. 2009; Lawrence et al. 2013; Makova and Hardison 2015; Baylin and Jones 2016; Alexandrov et al. 2020; Abascal et al. 2021; Sherman et al. 2022). However, at the site-specific level, the effectiveness of these covariates in predicting mutational landscapes depends on their integration into a detailed model. Overemphasizing these covariates could lead to false negatives for known driver mutations (Hess et al. 2019; Elliott and Larsson 2021). In figure 3B of the main text, we illustrate the discrepancy between the mutation rate predictions from Dig and empirical observation. Ideally, no covariates would be needed under extensive sample sizes, where each mutable genomic sites would have sufficient mutations to yield a statistic significance and consequently, synonymous mutations would be sufficient for the characterization of mutational landscape. In this sense, the integration of mutational covariates represents a compromise under current sample size. In our study, the effect of unique mutational landscapes is captured by E(u), the mean mutation rate for each cancer type. We further accounted for the variability of site-level mutability using a gamma distribution. The primary goal of our study is to determine the upper limit of mutation recurrences under mutational mechanisms only. While selection force acts blindly to genomic features, mutational hotspots should exhibit common characteristics determined by their underlying mechanisms. In the main text, we attempted to identify such shared features among CDNs. Until these mutational mechanisms are fully understood, CDNs should be considered as potential driver mutations.

(3) L223, the statement "In other words, the sequences surrounding the high-recurrence sites appear rather random.". Since it was a pan-cancer analysis, the unique patterns of each cancer type could be strongly diluted in the pan-cancer data.

We now state that the analyses of mutation characteristic have been applied to the individual cancer types and did not find any pattern that deviates from randomness. Nevertheless, it may be argued that, with the exception of those with sufficiently large sample sizes such as lung and breast cancers, most datasets do not have the power to reject the null hypothesis. To alleviate this concern, we applied the ResNet and LSTM/GRU methods for the discovery of potential mutation motifs within each cancer type. All methods are more powerful than the one used but the results are the same – no cancer type yields a mutation pattern that can reject the null hypothesis of randomness (see below).

As a positive control, we used these methods for the discovery of splicing sites of human exons. When aligned up with splicing site situated in the center (position 51 in the following plot), the sequence motif would look like:

Author response image 1.

5-prime

Author response image 2.

3-prime

However, To account for the potential influence of distance from the mutant site in motif analysis, we randomly shuffled the splicing sites within a specified window around the alignment center, and their sequence logo now looks like:

Author response image 3.

5-prime shuffled

Author response image 4.

3-prime shuffled

Author response image 5.

random sequences from coding regions

The classification results of the shuffled 5-prime (donner), 3-prime (acceptor) and random sequences from coding regions (Random CDS) are presented in the Author response table 1 (The accuracy for the aligned results, which is approximately 99%, is not shown here).

Author response table 1.

With the positive results from these positive controls (splicing site motifs) validating our methodology, we applied the same model structure to the train and test of potential mutational motifs of CDN sites. All models achieved approximately 50% accuracy in CDN motif analysis, suggesting that the sequence contexts surrounding CDN sites are not significantly different from other coding regions of the genome. This further implies that the recurrence of mutations at CDN sites is more likely driven by selection rather than mutational mechanisms.

Note that this preliminary analysis may be limited by insufficient training data for CDN sites. Future studies will require larger sample sizes and more sophisticated models to address these limitations.

(4) To solidify the findings, the results need to be replicated in an independent dataset.

Figure 7 validates our CDN findings using the GENIE dataset, which primarily consists of targeted sequencing data from various panels. By focusing on the same genomic regions sequenced by GENIE, we observed a 3-5 fold increase in the number of discovered CDNs as sample size increased from approximately 1000 to 9000. Moreover, the majority of CDNs identified in TCGA were confirmed as CDNs in GENIE.

(5) The key scripts and the list of key results (i.e., CDN sites with i{greater than or equal to}3) need to be shared to enable replication, validation, and further research. So far, only CDN sites with i{greater than or equal to}20 have been shared.

We have now updated the “Data Availability” section in the main text, the corresponding scripts for key results are available on Gitlab at: https://gitlab.com/ultramicroevo/cdn_v1.

(6) The versions of data used in this study are not clearly detailed, such as the specific version of gnomAD and the version and date of TCGA data downloaded from the GDC Data Portal.

The versions of data sources have now been updated in the revised manuscript.

Recommendations For The Authors:

(1) L119, states "22.7 million nonsynonymous sites," but Table 1 lists the number as 22,540,623 (22.5 million). This discrepancy needs to be addressed for consistency.<br /> (2) Figure 2B, there is an unexplained drop in the line at i = 6 and 7 (from 83 to 45). Clarification is needed on why this drop occurs.<br /> (3) Figure 3A, for the CNS type, data for recurrence at 8 and 9 are missing. An explanation should be provided for this absence.<br /> (4) L201, the title refers to "100-mers," but L218 mentions "101-mers." This inconsistency needs to be corrected to ensure clarity and accuracy.<br /> (5) Figures 6 and 7 currently lack titles. Titles should be added to these figures to improve readability.

Thanks. All corrections have been incorporated into the revised manuscript.

Reviewer #2 (Public Review):<br /> Summary:<br /> The authors propose that cancer-driver mutations can be identified by Cancer Driving Nucleotides (CDNs). CDNs are defined as SNVs that occur frequently in genes. There are many ways to define cancer driver mutations, and the strengths and weaknesses are the reliance on statistics to define them.<br /> Strengths:<br /> There are many well-known approaches and studies that have already identified many canonical driver mutations. A potential strength is that mutation frequencies may be able to identify as yet unrecognized driver mutations. They use a previously developed method to estimate mutation hotspots across the genome (Dig, Sherman et al 2022). This publication has already used cancer sequence data to infer driver mutations based on higher-than-expected mutation frequencies. The advance here is to further illustrate that recurrent mutations (estimated at 3 or more mutations (CDNs) at the same base) are more likely to be the result of selection for a driver mutation (Figure 3). Further analysis indicates that mutation sequence context (Figure 4) or mutation mechanisms (Figure 5) are unlikely to be major causes for recurrent point mutations. Finally, they calculate (Figure 6) that most driver mutations identifiable by the CDN approach could be identified with about 100,000 to one million tumor coding genomes.<br /> Weaknesses:<br /> The manuscript does provide specific examples where recurrent mutations identify known driver mutations but do not identify "new" candidate driver mutations. Driver mutation validation is difficult and at least clinically, frequency (ie observed in multiple other cancer samples) is indeed commonly used to judge if an SNV has driver potential. The method would miss alternative ways to trigger driver alterations (translocations, indels, epigenetic, CNVs). Nevertheless, the value of the manuscript is its quantitative analysis of why mutation frequencies can identify cancer driver mutations.

Recommendations For The Authors<br /> Whereas the analysis of driver mutations in WES has been extensive, the application of the method to WGS data (ie the noncoding regions) would provide new information.

We appreciate that Reviewer #2 has suggested the potential application of our method to noncoding regions. Currently, the background mutation model is based on the site level mutations in coding regions, which hinders its direct applications in other mutation types such as CNVs, translocations and indels. We acknowledge that the proportion of patients with driver event involving CNV (73%) is comparable to that of coding point mutations (76%) as reported in the PCAWG analysis (Fig. 2A from Campbell et al., 2020). In future studies, we will attempt to establish a CNV-based background mutation rate model to identify positive selection signals driving tumorigenesis.

References

Abascal F, Harvey LMR, Mitchell E, Lawson ARJ, Lensing SV, Ellis P, Russell AJC, Alcantara RE, Baez-Ortega A, Wang Y, et al. 2021. Somatic mutation landscapes at single-molecule resolution. Nature:1–6.

Alexandrov LB, Kim J, Haradhvala NJ, Huang MN, Tian Ng AW, Wu Y, Boot A, Covington KR, Gordenin DA, Bergstrom EN, et al. 2020. The repertoire of mutational signatures in human cancer. Nature 578:94–101.

Bailey MH, Tokheim C, Porta-Pardo E, Sengupta S, Bertrand D, Weerasinghe A, Colaprico A, Wendl MC, Kim J, Reardon B, et al. 2018. Comprehensive Characterization of Cancer Driver Genes and Mutations. Cell 173:371-385.e18.

Baylin SB, Jones PA. 2016. Epigenetic Determinants of Cancer. Cold Spring Harb Perspect Biol 8:a019505.

Campbell PJ, Getz G, Korbel JO, Stuart JM, Jennings JL, Stein LD, Perry MD, Nahal-Bose HK, Ouellette BFF, Li CH, et al. 2020. Pan-cancer analysis of whole genomes. Nature 578:82–93.

Drost J, Clevers H. 2018. Organoids in cancer research. Nat Rev Cancer 18:407–418.

Elliott K, Larsson E. 2021. Non-coding driver mutations in human cancer. Nat Rev Cancer 21:500–509.

Grzeskowiak CL, Kundu ST, Mo X, Ivanov AA, Zagorodna O, Lu H, Chapple RH, Tsang YH, Moreno D, Mosqueda M, et al. 2018. In vivo screening identifies GATAD2B as a metastasis driver in KRAS-driven lung cancer. Nat Commun 9:2732.

Hess JM, Bernards A, Kim J, Miller M, Taylor-Weiner A, Haradhvala NJ, Lawrence MS, Getz G. 2019. Passenger Hotspot Mutations in Cancer. Cancer Cell 36:288-301.e14.

Hodis E, Triglia ET, Kwon JYH, Biancalani T, Zakka LR, Parkar S, Hütter J-C, Buffoni L, Delorey TM, Phillips D, et al. 2022. Stepwise-edited, human melanoma models reveal mutations’ effect on tumor and microenvironment. Science 376:eabi8175.

Lawrence MS, Stojanov P, Polak P, Kryukov GV, Cibulskis K, Sivachenko A, Carter SL, Stewart C, Mermel CH, Roberts SA, et al. 2013. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 499:214–218.

Makova KD, Hardison RC. 2015. The effects of chromatin organization on variation in mutation rates in the genome. Nat Rev Genet 16:213–223.

Michels BE, Mosa MH, Streibl BI, Zhan T, Menche C, Abou-El-Ardat K, Darvishi T, Członka E, Wagner S, Winter J, et al. 2020. Pooled In Vitro and In Vivo CRISPR-Cas9 Screening Identifies Tumor Suppressors in Human Colon Organoids. Cell Stem Cell 26:782-792.e7.

Ng PK-S, Li J, Jeong KJ, Shao S, Chen H, Tsang YH, Sengupta S, Wang Z, Bhavana VH, Tran R, et al. 2018. Systematic Functional Annotation of Somatic Mutations in Cancer. Cancer Cell 33:450-462.e10.

Sherman MA, Yaari AU, Priebe O, Dietlein F, Loh P-R, Berger B. 2022. Genome-wide mapping of somatic mutation rates uncovers drivers of cancer. Nat Biotechnol 40:1634–1643.

Stamatoyannopoulos JA, Adzhubei I, Thurman RE, Kryukov GV, Mirkin SM, Sunyaev SR. 2009. Human mutation rate associated with DNA replication timing. Nat Genet 41:393–395.

Author response:

The following is the authors’ response to the original reviews

Public Reviews: 

Reviewer #1 (Public review): 

Summary:  

Wang et al. investigate sexual dimorphic changes in the transcriptome of aged humans. This study relies upon analysis of the Genotype-Tissue Expression dataset that includes 54 tissues from human donors. The authors investigate 17,000 transcriptomes from 35 tissues to investigate the effect of age and sex on transcriptomic variation, including the analysis of alternative splicing. Alternative splicing is becoming more appreciated as an influence in the aging process, but how it is affected by sexual dimorphism is still largely unclear. The authors investigated multiple tissues but ended up distilling brain tissue down to four separate regions: decision, hormone, memory, and movement. Building upon prior work, the authors used an analysis method called principal component-based signal-to-variation ratio (pcSVR) to quantify differences between sex or age by considering data dispersion. This method also considers differentially expressed genes and alternative splicing events. 

Strengths:  

(1) The authors investigate sexual dimorphism on gene expression and alternative splicing events with age in multiple tissues from a large publicly available data set that allows for reanalysis. 

(2) Furthermore, the authors take into account the ethnic background of donors. Identification of agingmodulating genes could be useful for the reanalysis of prior data sets. 

Weaknesses:  

The models built off of the GTEx dataset should be tested in another data set (ex. Alzheimer's disease) where there are functional changes that can be correlated. Gene-length-dependent transcription decline, which occurs with age and disease, should also be investigated in this data set for potential sexual dimorphism. 

We appreciate the reviewer’s constructive feedback and acknowledgment of the strengths of our study. The detailed results are included in the ‘Recommendations for the authors’ from the editorial office. Below we summarize our feedback that address the concerns of this reviewer:

(1) Independent Alzheimer’s disease (AD) datasets:

We acknowledge the importance of validating our models beyond GTEx to assess their generalizability aging to Alzheimer’s disease. While GTEx provides valuable transcriptomic data across multiple tissues, it lacks direct functional assessments linked to disease states. We have already analyzed RNA-seq data from ROSMAP and GEO in Figure 4, focusing on sex-biased gene expression and splicing changes between aging and AD.  The results showed a male-biased association with Alzheimer’s disease at AS resolution, indicating that the AS changes during aging could contribute more to AD in males than females. We added a highlight to this analysis in the manuscript (Pages 6-7).

(2) Sexual dimorphism in Gene-Length-Dependent Transcription Decline (GLTD) 

We appreciate the reviewer’s suggestion to explore gene-length-dependent transcription decline (GLTD), which has been implicated in both aging and disease. As the reviewer suggested, our analysis revealed that GLTD exhibits sex-biased patterns in different tissues, aligning with recent literature on sex-dimorphic transcriptional aging. Our findings also revealed that longer genes with greater transcriptional decline are enriched in AD-related pathways. We have incorporated this new analysis in the ‘Recommendations for the authors’ in Author response image 5-6 and expanded the discussion of the biological relevance. 

Reviewer #2 (Public review): 

Summary: 

In this manuscript, Wang et al analyze ~17,000 transcriptomes from 35 human tissues from the GTEx database and address transcriptomic variations due to age and sex. They identified both gene expression changes as well as alternative splicing events that differ among sexes. Using breakpoint analysis, the authors find sex dimorphic shifts begin with declining sex hormone levels with males being affected more than females. This is an important pan-tissue transcriptomic study exploring age and sex-dependent changes although not the first one. 

Strengths:  

(1) The authors use sophisticated modeling and statistics for differential, correlational, and predictive analysis. 

(2) The authors consider important variables such as genetic background, ethnicity, sampling bias, sample sizes, detected genes, etc. 

(3) This is likely the first study to evaluate alternative splicing changes with age and sex at a pan-tissue scale. 

(4) Sex dimorphism with age is an important topic and is thoroughly analyzed in this study.  Weaknesses:  

(1) The findings have not been independently validated in a separate cohort or through experiments. Only selective splicing factor regulation has been verified in other studies. 

(2) It seems the authors have not considered PMI or manner of death as a variable in their analysis. 

(3) The manuscript is very dense and sometimes difficult to follow due to many different types of analyses and correlations. 

(4) Short-read data can detect and quantify alternative splicing events with only moderate confidence and therefore the generalizability of these findings remains to be experimentally validated. 

We appreciate the thorough review and thoughtful feedback. We have addressed the reviewer’s concerns and added clarification. The detailed results are included in Recommendations for the authors. Here are the summaries.

(1) Challenge of independent validation in separate cohorts

• The GTEx dataset includes the most comprehensive transcriptome resource for studying population-level differences in age and sex across tissues, particularly including large-scale brain samples. This provides a unique opportunity to analyze sex-dimorphic aging and the relevance of age-associated diseases.  Several technical issues, including cell type heterogeneity, postmortem artifacts, as well as sequencing biases, lead to technical challenges in different cohorts.

• As the reviewer mentioned, we analyzed transcriptomic data from Shen et al. (2024) and compared them with GTEx results (Author response image 2). Limited overlap in differentially expressed genes again highlighted the challenges in cross-dataset validation due to the differences in cell composition and data processing (peripheral blood mononuclear cells (PBMCs) vs whole blood). 

• Due to the limited human brain transcriptome data covering different age and sex groups, we found mouse hippocampus datasets from Mass spectrometry (MS), including young and old, as well as female and male groups.  The results validated the expression of splicing factors in brain (Author response image 9). This cross-species consistency supports the robustness of our findings in human brain aging.

(2) Effects of Postmortem Interval, Manner of Death, and Time of Death

• We agree that the sample collections could introduce confounding effects. To address this, we calculated the correlations between the confounding factors with Postmortem Interval (PMI), Manner of Death (DTHMNNR), or Time of Death (DTHTIME and DTHSEASON). We observed strong correlations in some surrogate variables in most tissues, indicating that those factors could be well-regressed during our analysis (Recommendations for the authors, Figure S4 and R8). 

• In addition, we re-evaluated our analyses while incorporating PMI as a covariate in our models. Our results align with our initial findings (Author response image 1), suggesting that age- and sex-dependent transcriptomic changes are not strongly confounded by PMI and confirming that our model has controlled PMI. These results are detailed in ‘Recommendations for the authors’ and included in Figure S4C-E with the description in text, Page 5. 

(3) Readability of manuscript and flow of analyses

• In summary, our study first examined global alternative splicing (AS) and gene expression (GE) across all tissues before focusing on specific regions for deeper insights. To improve clarity, we have made the following revisions:

• Add clearer statements when transitioning between all-tissue and brain-specific analyses (Page 6-7).

• Modify the subtitle of Results to highlight all-tissue vs. brain analyses (Page 6).

• These refinements could enhance the manuscript’s structure, making the flow of analysis and conclusions more intuitive for readers.

(4) Limitations of short-read RNA-seq for splicing analysis

• Short-read RNA-seq provides only moderate confidence in detecting and quantifying full-length isoforms. However, its higher sequencing depth makes it more suitable for quantifying changes in alternative splicing (AS) events.

• Our analysis focused on splicing event-level quantification, applying stringent filters and using our GPU-based tool, which showed strong concordance with RT-PCR and other pipelines. Therefore, we also cited and included the updated Paean manuscript that benchmarks its performance in AS analysis.

Reviewer #3 (Public review): 

Summary:  

In this study, Wang et al utilized the available GTEx data to compile a comprehensive analysis that attempt to reveal aging-related sex-dimorphic gene expression as well as alternative splicing changes in humans. 

The key conclusions based on their analysis are that. 

(1) extensive sex-dimorphisms during aging with distinct patterns of change in gene expression and alternative splicing (AS), and 

(2) the male-biased age-associated AS events have a stronger association with Alzheimer's disease, and  (3) the female-biased events are often regulated by several sex-biased splicing factors that may be controlled by estrogen receptors. They further performed break-point analysis and revealed that in males there are two main breakpoints around ages 35 and 50, while in females, there is only one breakpoint at 45. 

Strengths:  

This study sets an ambitious goal, leveraging the extensive GTEx dataset to investigate aging-related, sexdimorphic gene expression and alternative splicing changes in humans. The research addresses a significant question, as our understanding of sex-dimorphic gene expression in the context of human aging is still in its early stages. Advancing our knowledge of these molecular changes is vital for identifying therapeutic targets for age-related diseases and extending the human health span. The study is highly comprehensive, and the authors are commendable for their attempted thorough analysis of both gene expression and alternative splicing - an area often overlooked in similar studies. 

We thank this reviewer for the insightful review and recognition of our study's significance.  We agree with the reviewer on how to examine sex-dimorphic gene expression and alternative splicing in aging by using the GTEx dataset.  This is indeed an essential aspect of developing potential therapeutic targets for agerelated diseases to promote human health span.

Weaknesses:  

Due to the inherent noise within the GTEx dataset - which includes numerous variables beyond aging and sex - there are significant technical concerns surrounding this study. Additionally, the lack of crossvalidation with independent, existing data raises questions about whether the observed gene expression changes genuinely reflect those associated with human aging. For instance, the break-point analysis in this study identifies two major breakpoints in males around ages 35 and 50, and one breakpoint in females at age 45; however, these findings contradict a recent multi-omics longitudinal study involving 108 participants aged 25 to 75 years, where breakpoint at 44 and 60 years was observed in both male and females (Shen et al, 2024). These issues cast doubt on the robustness of the study's conclusions. Specific concerns are outlined below: 

References: 

Ferreira PG, Muñoz-Aguirre M, Reverter F, Sá Godinho CP, Sousa A, Amadoz A, Sodaei R, Hidalgo MR, Pervouchine D, Carbonell-Caballero J et al (2018) The effects of death and post-mortem cold ischemia on human tissue transcriptomes. Nature Communications 9: 490. 

Shen X, Wang C, Zhou X, Zhou W, Hornburg D, Wu S, Snyder MP (2024) Nonlinear dynamics of multiomics profiles during human aging. Nature Aging. 

Wucher V, Sodaei R, Amador R, Irimia M, Guigó R (2023) Day-night and seasonal variation of human gene expression across tissues. PLOS Biology 21: e3001986. 

(1) The primary method used in this study is linear regression, incorporating age, sex, and age-by-sex interactions as covariates, alongside other confounding factors (such as ethnicity) as unknown variables. However, the analysis overlooks two critical known variables in the GTEx dataset: time of death (TOD) and postmortem interval (PMI). Both TOD and PMI are recorded for each sample and account for substantial variance in gene expression profiles. A recent study by Wucher et al.(Wucher et al, 2023) demonstrated the powerful impact of TOD on gene expression by using it to reconstruct human circadian and even circannual datasets. Similarly, Ferreira et al. (Ferreira et al, 2018) highlighted PMI's influence on gene expression patterns. Without properly adjusting for these two variables, confidence in the study's conclusions remains limited at best. 

We appreciate the reviewer for raising this important point regarding the impact of post-mortem interval (PMI) and time of death (TOD) on gene expression, including the death seasons (DTHSEASON) and daytime (DTHTIME). To address this point, we carefully evaluated whether our linear model controlled for these factors as potential confounders. 

Our results showed that PMI and TOD significantly correlated with the estimated covariates in most tissues, suggesting that their effects could be effectively regressed out using our model (Figure S4).  As the reviewers and editors suggested, we have now included this correlation analysis in the updated Figure S4C-E and the text in the Results section, citing relevant literature [1,2] (Page 5). 

Author response image 1.

The results of differential gene expression analysis with vs without the inclusion of PMI correction as a known covariate. The scatter plots show the correlations of significance levels (pvalues, left panel) and effect sizes (coefficients, right panel) of sex (A) and age (B). Whole-blood tissue is used as an example.

In addition, we did the differential analysis that incorporated PMI as a covariate in the regression models and re-evaluated the age- and sex-related transcriptomic changes. Using WholeBlood gene expression as an example, our revised analysis shows that the inclusion of PMI in the covariates has minimal impact on the significance levels and effects of sex and age (i.e., p-values and coefficients, respectively), indicating that our findings are robust using confounding factors (Author response image 1). 

(2) To demonstrate that their analysis is robust and that the covariates TOD and PMI are otherwise negligible - the authors should cross-validate their findings with independent datasets to confirm that the identified gene expression changes are reproducible for some tissues. For instance, the recent study by Shen et al. (Shen et al., 2024) in Nature Aging offers an excellent dataset for cross-validation, particularly for blood samples. Comparing the GTEx-derived results with this longitudinal transcriptome dataset would enable verification of gene expression changes at both the individual gene and pathway levels. Without such validation, confidence in the study's conclusions remains limited. 

We thank the reviewer for the insightful suggestion regarding cross-validation with independent datasets. We understand that validating findings across datasets is crucial for ensuring robustness. As the reviewers suggested, we see whether there are some shared findings in the GTEx data with the study by Shen et al. (2024) in Nature Aging. However, after performing comparisons with our GTEx results in whole blood tissue, we found that the overlaps of differentially expressed genes are limited (Fig. 3). In our results, we found a large proportion of age-associated genes in the GTEx data, whereas just 54 genes are age-associated from Shen et al.’s PBMC data. 3 in 7 genes are differentially expressed in both datasets (Fig. 3A). Additionally, we performed the functional enrichment analysis on the GTEx-specific age-associated genes.

We observed a strong enrichment in the biological pathways related to neutrophil functions and innate immune responses, which are specific to the cell compositions in whole blood rather than PBMC (Fig. 3B).

Author response image 2.

The comparison between the gene expression of whole blood tissue from GTEx and PBMCs from Shen et al. (A) The bar plot shows the number of age (left panel) or sex-associated  (right panel) genes in the two datasets. The grey bars highlight the proportion of overlapped genes in both datasets. (B) The top 10 significantly enriched biological processes in the GTEx-specific age-associated genes. The color bar shows the number of age-associated genes in specific pathways.

These discrepancies highlighted the crucial factors in cross-dataset comparison:

• Cell compositions: GTEx used whole blood, which contains all blood components, including neutrophils and erythrocytes, whereas PBMCs contain lymphocytes and monocytes. Under the influence of granulocytes and red blood cells in whole blood, the gene expression profiles between these two datasets are different.

• Biological functions: Whole blood includes both innate and adaptive immune components; thus, aging-related gene expression changes in whole blood may include a broader systemic response than those in PBMCs. This difference in biological context contributes to the observed variation in the differentially expressed genes, as demonstrated by our functional enrichment analysis (Fig. 3B). 

• Sequencing biases and data processing: The two datasets were generated using different RNAseq processing pipelines, including distinct normalization, batch correction, and quantification methodologies. These technical differences may introduce systematic variations that complicate direct cross-validation.

Due to these fundamental problems, a direct one-to-one validation between the two datasets is challenging. We understand the importance of independent dataset validation and appreciate the reviewer’s suggestion. However, future studies could be performed more precisely if comparable whole-blood-based datasets are available. In addition, GTEx data provides nearly thousands of samples in whole blood, which is a largescale, comprehensive, and clinically relevant dataset for studying aging-related changes, particularly in innate immunity and inflammation, which are not well captured in PBMCs.

(3) As a demonstration of the lack of such validation, in the Shen et al. study (Shen et al., 2024), breakpoints at 44 and 60 years were observed in both males and females, while this study identifies two major breakpoints in males around ages 35 and 50, and one breakpoint in females at age 45. What caused this discrepancy? 

We thank the reviewer and the editors for both coming up with the non-linear multi-omic aging patterns observed by Shen et al.  They observed two prominent crests around the ages of 45 and 60 from omics data.

Similarly, we also identified two breakpoints in our analysis, with some differences in specific age breakpoints. These could be the result of sample preparation methods and breakpoint definition. These responses are also included in the editor’s recommendations.

Definition of breakpoints vs crests:

• Crests represent age-related molecular changes at each time point across the human lifespan. They indicate the number of molecules that are differentially expressed during aging (q < 0.05), without considering individual expression levels.

• Our breakpoints, in contrast, are identified after filtering the chronological trends using the Autoregressive Integrated Moving Average (ARIMA) model. We calculated the rate of change at each age point using the smooth approach and sliding windows. Breakpoints are defined as local maxima where the distance to the nearest minimum, relative to the global maximum. We indeed found some local wide peaks around 60 in some tissues, shown in Figure S10, however, we excluded these due to our strict cutoffs to remove noise.

Differences and similarities between sequenced tissues: 

• Whole-blood vs PBMC: In the GTEx RNA-seq data used in our study, whole blood samples from donors were sequenced, whereas their study used PBMCs. Whole blood contains all blood components, including red blood cells, platelets, granulocytes (e.g., neutrophils), lymphocytes, and monocytes, while PBMCs represent a subset of white blood cells, primarily consisting of lymphocytes (T cells, B cells, NK cells) and monocytes, excluding granulocytes and erythrocytes. As we mentioned in the previous responses, the gene expression changes observed in whole blood capture the contributions of neutrophils and other granulocytes, which are neglected in the PBMC profile (also shown in Figure S11C). 

• For the shared tissues in two studies – skin, we looked at the non-linear changes during aging and found the same two breakpoints: 43 and 58. 

Novelties in our study:

• Whole blood can serve as a readily accessible resource for testing age-related disease biomarkers without cell separation, making it more practical for clinical applications.

• Our analysis was performed on females and males, respectively. The main object of our analysis is to compare the differences in aging rates between sexes. Our results reveal clear sex-specific differences across multiple human tissues. Therefore, the identified breakpoints may differ when sex effects are not taken into account, highlighting the specificity of our analysis. 

• Additionally, our breakpoints are integrated across multiple tissues. Our results showed that there is a large diversity of aging patterns in different tissues.

As the reviewers and editors suggested, we have added the following statements to clarify this distinction in the Discussion section: ‘Our analysis observed the non-linear aging patterns with two breakpoints, which is consistent with recent findings, with differences in specific age points due to sex differences as well as tissue diversities 3.’ (Page 14), and ‘These breakpoints could represent key junctures in the aging process that align with the non-linear patterns of aging and disease progression.’ (Page 15)

(4) Although the alternative splicing analysis is intriguing, the authors did not differentiate between splicing events that alter the protein-coding sequence and those that do not. Many splicing changes occurring in the 5' UTR and 3' UTR regions do not impact protein coding, so it is essential to filter these out and focus specifically on alternative splicing events that can modify protein-coding sequences. 

The reviewer raises an important point. In our study, we included the AS events in protein-coding genes to gain a comprehensive understanding of sex-biased age-associated splicing. As the reviewer suggested, focusing on coding-sequence-altering events is particularly relevant to protein function. To address this, we performed an additional analysis to specifically annotate sBASEs occurring within the coding sequence (defeined as CDS-altering sBASEs) and reanalyzed their functional pathways and AD-associations (Author response image 3).  

Our analysis revealed that most of the sBASEs are relevant to protein-coding sequences (CDS) across multiple tissues (Author response image 3A).  We then confirmed our findings using CDS-altering sBASEs. We found that those sBASEs in brain regions were significantly enriched in pathways related to amyloid-beta formation and actin filament organization (Author response image 3B). Notably, male-biased sBASEs in decision-related brain regions were particularly associated with dendrite development and regulation of cell morphogenesis, highlighting the sex-specific roles of sBASEs in brain functions. Additionally, we performed a random forest classification using only CDS-altering sBASEs in AD datasets (Author response image 3C-D), again confirming the malebiased association between aging and AD.

Overall, we found that most of the identified sBASEs could modify protein-coding sequences, and our main conclusions remain consistent even after filtering out non-coding events. 

Nevertheless, in addition to AS events that impact protein sequences, alternative splicing in untranslated regions (UTRs) also plays a critical regulatory role. Splicing events in the 5′ UTR can influence translation efficiency by modifying upstream open reading frames (uORFs) or RNA secondary structures, while splicing in the 3′UTR can affect mRNA stability, localization, and translation by altering microRNA binding sites and RNA-binding protein interactions. Given these functional implications, we believe that UTR-targeted AS events should also be considered to supplement the understanding of post-transcriptional gene regulation in future research.

Author response image 3.

The distribution and functional relevance of sBASEs with coding effects. (A) The number of sBASEs and CDS-altering sBASEs across multiple tissues. The deeper bars show the number of sBASEs whose alternative splice sites are located at protein-coding regions. (B) GO biological pathways in each sex and brain region. Heatmap shows the sex-specific pathways that are significantly enriched by CDS-altering sBASEs in more than 2 brain regions and sex. (C) Correlation between ADassociated and age-associated AS changes across the CDS-altering sBASEs that alter protein-coding sequences in females and males. (D) Performances of sex-stratified models predicted by CDS-altering sBASEs in 100 iterations using the random forest approach

(5) One of the study's main conclusions - that "male-biased age-associated AS events have a stronger association with Alzheimer's disease" - is not supported by the data presented in Figure 4A, which shows an association with "regulation of amyloid precursor formation" only in female, not male, alternative splicing genes. Additionally, the gene ontology term "Alzheimer's disease" is absent from the unbiased GO analysis in Figure S6. These discrepancies suggest that the focus on Alzheimer's disease may reflect selective data interpretation rather than results driven by an unbiased analysis. 

We thank the reviewer for this point. In our functional analysis, we identified distinct biological processes enriched in female- and male-biased AS genes, such as the regulation of amyloid precursor formation in females and structural constituents of the cytoskeleton in males. However, Alzheimer’s disease (AD) is a complex neurodegenerative disorder with multiple pathological mechanisms beyond amyloid-beta (Aβ) formation, many of which are strongly age-related in both sexes. This complexity motivates us to explore novel relationships between splicing and AD in distinct sexes.

Although Figure 4A shows the enrichment of “regulation of amyloid precursor formation” in female-biased AS events, this does not contradict the broader enrichment of AD-related processes in male-biased AS events. Our disease ontology analysis supports this finding, as male-biased age-associated AS events are enriched in neurodegenerative diseases, including cognitive disorders. Additionally, we considered not only individual GO terms but also the disease-associated transcriptomic signatures from AD-related datasets, which collectively indicate a stronger association in males. 

Regarding Figure S6 mentioned by the reviewer, the GO term “Alzheimer’s disease” is not explicitly listed in the heatmap because we filtered the pathways that are consistently enriched in multiple tissues. As noted in the figure legend, we only displayed sex-specific GO terms that were significant in at least 15 tissues. Then, since the brain is highly affected by age-related processes and neurological conditions show sex differences, the sex-biased AS events could help explain differential susceptibility to age-related cognitive decline and neurodegeneration. That’s why we chose the brain data for detailed analysis.

To improve clarity, we have revised the text to describe the purpose of our analysis in brain rather than other tissues (Page 6-7). We appreciate the reviewer’s feedback, and we will consider additional analyses to further explore the sex-biased AS as well as disease risk in other tissues.

(6) The experimental data presented in Figures 5E - I merely demonstrate that estrogen receptor regulates the expression of two splicing factors, SRSF1 and SRSF7, in an estradiol-dependent manner. However, this finding does not support the notion that this regulation actually contributes to sex-dimorphic alternative splicing changes during human aging. Notably, the authors do not provide evidence that SRSF1 and SRSF7 expression changes actually occur in a sex-dependent manner with human aging (in a manner similar to TIA1). As such, this experimental dataset is disconnected from the main focus of the study and does not substantiate the conclusions on sex-dimorphic splicing during human aging. The authors performed RNAseq in wild-type and ER mutant cells, and they should perform a comprehensive analysis of ER-dependent alternative splicing and compare the results with the GTEx data. It should be straightforward. 

Thanks for the reviewer’s feedback. The main purpose of the analyses in Figures 5E-I was to explore which factors affect the sex-biased expression of splicing factors during aging and substantially regulate alternative splicing (AS). To address the reviewer’s concerns, we have included additional analysis and explained the challenge of linking estrogen receptor (ER)-regulated splicing factors to sex-dimorphic AS changes during human aging in specific human cell types. 

• As suggested by the reviewer, we first examined the expression changes of SRSF1 and SRSF7 during aging in males and females, like TIA1 in decision-related brain regions (Fig. 5I).

• Secondly, the regulation is based on a highly complex regulatory network involving multiple splicing factors and cell heterogeneity. Due to these complexities, we did not overlap ER-dependent AS changes with sBASEs from GTEx datasets directly. As far as the reviewer is concerned, we supplemented the AS analysis in the GSE89888 dataset (Fig. 5H) and identified the estrogenregulated AS events mediated by ESR1. We found that ~6% (26/396) of female-specific ageassociated AS events were regulated by ESR1, of which 6 sBASEs can be regulated by femalebiased splicing factors. The low overlaps could be represented by the limited coverage of different RNA-seq datasets and cell types used across these analyses. Notably, the results indicated that only a fraction of AS could be directly accounted for by estrogen via ESR1, suggesting the complexity of transcriptional and splicing regulatory networks during aging. 

• Meanwhile, we downloaded independent experimental datasets to discover the regulation by our candidate splicing factors. Due to SRSF1 is identified as a potential regulator of sex-biased splicing, we analyzed RNA-seq data with SRSF1 knock-down (KD) glioblastoma cell lines (U87MG and U251), a type of brain cancer formed from astrocytes that support nerve cells 4.  As a result, we indeed found that some sBASEs are regulated by SRSF1 during aging through this experiment using brain cell lines (Author response image 4). Together, these results suggested that some of the SF-RNA regulatory relationships can be observed in another cellular system, further supporting our findings. 

Due to the limitations of cell-based models and the complexity in the splicing regulatory network, it is challenging to directly validate aging regulation, particularly between different sexes, based on ER treatments in vivo. However, our findings still provide valuable mechanistic insights into ER-regulated splicing factors, implying their potential role in sex-biased aging.

Author response image 4.

SRSF1 regulations on specific sBASEs using SRSF1 knock-down RNA-seq data in GBM cells. Three examples are shown to be regulated during aging with significant changes between SRSF1 KD vs control in U251 and U87MG cell lines. The splicing diagrams are shown below.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors): 

The authors found that alternative splicing was affected by both sex and age across many tissues, with gene expression differences affected by both parameters only present in some tissues. This trend was consistent when the effects of sex chromosomes were subtracted from the analysis. The effect of aging on differential gene expression and alternative splicing was more prevalent in male than female samples. For analysis purposes, young subjects were deemed to be anyone under 40, and old subjects were over 60 years old. The authors then investigated if specific genes or alternative splicing events were responsible for these effects. Some candidate genes or splicing events were identified but there was little overlap between tissues, suggesting no universal gene or event as a driver of aging. Surrogate variables like the ethnic backgrounds of donors were also investigated. Ultimately the authors found that alternative splicing events showed a stronger sexual dimorphic effect with age than did differential gene expression and that at least for the brain, alternative splicing changes showed a bias for Alzheimer's disease in male samples. This was highlighted by examples of exon skipping in SCL43A2 and FAM107A in males that were associated respectively with plaques and tangles. 

The authors go on to identify sexual dimorphic differences in splicing factors in particular brain regions during age. Finally, the authors performed analysis for aging-modulated genes, identifying nearly 1000 across the tissues, nearly 70% of which are sex-specific. Their work suggests that further analysis of these aging-modulated genes could be differentially modulating the transcriptome based on sex. The work is novel and interesting, especially investigating sexual dimorphism in alternative splicing. However, the work is still preliminary, and these assumptions need to be applied to other data sets beyond GTEx for validation as well as some other phenomena that need to be considered. I recommend major revisions to address the points below. 

(1) At the beginning of the results section, the authors state that the brain is stratified into four functional regions. It would be useful to explicitly state those four regions in the text at that point. 

We agree that specifying these regions early in the text will improve clarity and provide the reader with a clear understanding of the analysis. As the reviewer’s suggestion, we revised the Results section (Page 3) to explicitly state the four functional brain regions as follows: ‘Due to data sparseness, the brain tissues were recombined into four functional regions (table S1), including hormone- or emotion-related region, movement-related region, memory-related region, and decision-related region (See Methods).’. This ensures that the regions are clearly defined before the subsequent analysis is presented. 

(2) The manuscript becomes a bit confusing when the authors shift from all the tissues as a whole specifically to the brain and then back to the larger tissue set to make assumptions. This can be a bit confusing and should be better delineated.

We thank the reviewer and editor for the feedback regarding the transitions between the analysis of all tissues and the brain-specific analysis. In our study, we first conducted a broad analysis of alternative splicing (AS) and gene expression (GE) across all tissues. For the AS analyses, we did sBASEs analysis in all tissues and then focused on specific tissue (i.e., brain) whose splicing changes are functionally enriched with age-related diseases.  For the GE analyses, we also analyzed the aging rate across tissues and identified the tissue-specific/shared patterns. 

We agree that the shifts of the tissues for AS and GE may cause some confusion, and have made the following revisions to delineate why we focused on different tissues for distinct analyses:

• We have added clear statements to better delineate when we shift focus from the analysis of all tissues to the region-specific analysis and vice versa. For instance, in the Results section (Page 67), we include a transitional phrase: ‘Having established patterns across all tissues, we now turn to a more focused analysis to investigate tissue-specific alternative splicing changes.’

• To improve the overall structure, we have reorganized the Results section, adding distinct subheadings for the analysis of all tissues and the brain (Page 6), which should make the transition between these sections smoother and more intuitive for the reader.

We believe that these revisions will make the manuscript’s structure clearer and allow the reader to better follow the flow of the analysis and the subsequent conclusions.

(3) Gene-length-dependent transcription decline (GLTD) is another phenomenon that occurs with aging and is known to be associated with Alzheimer's disease [PMID38519330]. The authors should make some statement if this is present in their dataset and if any sexual dimorphism in tissues is present. 

We thank the editors and reviewers for bringing up the possible connection of gene-length-dependent transcription decline (GLTD), which was reported to be associated with both aging and Alzheimer’s disease (AD). We appreciate the reviewer’s suggestion and have addressed whether GLTD is present in our dataset and whether any sex differences are observed in this context.

We evaluated GLTD using the correlation between gene length with age-associated changes (i.e., the coefficients of the ‘age’ term in the linear regression model) in GTEx data. We did observe strong evidence of GLTD, particularly in the brain, heart, muscle, pancreas, spleen, skin, muscle, etc (Author response image 5A). In brain, we performed the functional enrichment analysis on the genes with Foldchange > 2 and length > 10⁵ bp (Author response image 5B). We found that these extremely long genes are significantly relevant to synapse and neuron functions. These findings align with previous studies showing that GLTD can occur with aging in the tissues that are relevant to Alzheimer’s disease, cardiovascular diseases, and common failures of metabolism (e.g., diabetes) [5,6]. Additionally, it was not a ubiquitous phenomenon across all tissues. The correlations could be positive in tissues like adipose and artery.  These findings suggested the GLTD could be varied and tissuespecific in its manifestation during aging. 

Author response image 5.

(A) The correlation between gene length and age-associated changes across GTEx tissues in human samples. The correlation tests are evaluated using Spearman’s approach. The color bar indicates the -log10 transformed p-values in the correlation test. (B) The results of GO enrichment analysis using the genes with Foldchange > 2 and length > 10⁵ bp. The parent terms calculated by ‘rrvgo’ with a similarity threshold of 0.9 are shown.

Regarding sexual dimorphism, we conducted this analysis in females and males, respectively (Author response image 6). We found GLTD exists in both females and males in most tissues, such as brain, whole blood, muscle, etc, consistent with the previous results without considering the sex groups. Interestingly, we observed sexbiased patterns in certain tissues. In particular, the left ventricle, pancreas, and hippocampus showed notable male-biased patterns in the degree of transcriptional decline with gene length, whereas skin, liver, small intestine, and esophagus showed that in females. These findings suggest that GLTD could be relevant to aging and age-related diseases; the levels of expression and sexual dimorphism may vary depending on the tissue type. We hope this clarification addresses the reviewer’s concern and provides a more comprehensive understanding of the GLTD and sex differences observed in our dataset. 

Author response image 6.

The correlation between gene length and age-associated changes across tissues in females and males, respectively. The correlation tests are evaluated using the Spearman’s approach. The red dots indicate the significant correlations in females, while the navy dots show those in males.

(4) Because the majority of this work has been performed in the GTEx dataset, applying this analysis to another publicly available dataset would be useful validation. For instance, the authors have interesting findings in the brain and correlations to Alzheimer's disease. Analysis of an existing RNAseq dataset from Alzheimer's disease patients and controls (with functional outcomes) would provide more evidence beyond the preliminary findings from GTEx. 

We appreciate the reviewer’s suggestion on the validation of our findings by applying our analysis to independent RNA-seq datasets from Alzheimer’s disease patients. 

• We have used two Alzheimer’s disease datasets, GEO and ROSMAP, to investigate the correlation between aging and Alzheimer’s disease (AD) and included these analyses in our study (Fig. 4B-C and Figure S8C).

• In the Results section (Page 7), we have presented the results of this validation, where we identified correlations between sex-biased aging-related splicing changes and AD-related changes. These findings support the conclusions from the GTEx dataset and further strengthen the relevance of our results to AD.

As suggested, we have updated the manuscript to more explicitly highlight this validation in the Discussion section (Page 12), noting: ‘We further validated our findings using Alzheimer’s disease dataset, ROSMAP, where we observed consistent correlations between aging-related splicing changes and Alzheimer’s disease-related changes, providing additional evidence for the robustness of our results.’ 

Reviewer #2 (Recommendations for the authors): 

(1) In the text (Introduction and Discussion), the authors mention analyzing 54 tissues, the abstract states 35 tissues, Table S1 lists 48, and Figure 2A-B shows 33. Could the authors please clarify exactly how many tissues they used? I am also confused by the sample numbers in Table S1. For example: for adiposesubcutaneous tissue, the total number of females is listed as 218 but the sum of young and old females is only 110. Does this mean some samples were excluded? What is the exclusion criterion? 

We thank the reviewers and editors for pointing out the discrepancies regarding the number of tissues analyzed and the sample numbers in Table S1. We appreciate the opportunity to clarify these points:

Number of tissues analyzed:

• We downloaded and analyzed 17,382 samples in 54 tissues from GTEx in total (31 tissues and 13 brain regions), as mentioned in the Results, Methods, and Discussion sections. Table S1 lists 48 tissues (31 tissues, 13 brain regions, and 4 merged brain regions), which include a refined classification of the tissues we analyzed, accounting for the variations in brain region categorization in the dataset.

• The discrepancy also arises from the different sample size cutoffs in specific analyses. For pcSVR analysis (Figure 2A-B), we did the subsampling for the permutation analysis for certain key findings, so we filtered a subset of 33 tissues (29 tissues and 4 merged brain regions), which included at least 3 samples in each age group in females or males. 

• To resolve this, we have clarified the total number of tissues analyzed and aligned the numbers across the manuscript. In the revised manuscript, we now explicitly state in both the Abstract and Methods sections that 54 tissues were analyzed in the context of this study. We added a note in Methods to clarify that 35 tissues are 31 tissues and 4 merged brain regions (Page 16). In Figure 2A-B, we clarified that the 33 tissues are filtered due to the usage in this analysis (Page 17).

Sample numbers in Table S1:

• Regarding the sample sizes of age groups, the discrepancy occurred due to the classification of the age groups. We classify the samples into three: Young, Middle, and Old, as mentioned in the Results section (Page 4). 

• Additionally, we excluded the sample sizes in 13 single brain regions. We aligned the total tissue number to 35 with our texts.

We hope this resolves the confusion regarding the number of tissues and the sample sizes used in the analysis. These clarifications have been incorporated into the revised manuscript to ensure consistency.

(2) Was post-mortem interval (PMI) or manner of death considered in the model? For example, traumatic death may have major consequences on gene expression. Similarly, a few tissues have low sample numbers, for example, kidney cortex and brain. The pooling of brain samples is explained and the kidney cortex is excluded, so why is it listed in Table S1? 

Thank you for raising this important point regarding the potential impact of post-mortem interval (PMI) and manner of death (DTHMNNR) on gene expression. We carefully considered both factors as potential confounders in our analysis. 

Specifically, to evaluate their impacts, we calculated the correlations between the coefficients of PMI or manner of death, with the confounding factors. Our results showed that PMI and DTHMNNR are significantly correlated with the covariates in most tissues, suggesting that their effects could be effectively regressed in our model (Figure S4). As we have mentioned in Figure S4 and Author response image 1, we conducted a differential analysis that incorporated PMI as a covariate in the regression models and re-evaluated the age- and sex-related transcriptomic changes to address this concern. The high correlations showed the minor effect size of PMI when including the covariates in the model. As suggested by the reviewers and editors, we have now included this correlation analysis in Figure S4C-E and updated the text in the results section (Page 5).

Additionally, as the responses above, Table S1 provides the general sample sizes of all GTEx tissues without filtering. We have modified the table to include a total of 35 tissues, including 31 non-brain tissues and 4 brain regions.

(3) It might be important to show a simple visual of cohort details such as age ranges, sexes, ethnicities, PMIs, etc. 

To address this, we added summary figures to illustrate the distributions of key demographic variables, including age, sex, BMI, ethnicity, post-mortem intervals (PMIs), and manner of death (DTHMNNR) (Author response image 7 and Author response image 8). This will provide readers with a clearer overview of the dataset composition and potential covariates affecting the analysis. 

Author response image 7.

Age (left panel), BMI (Body Mass Index) (middle panel), and PMI (Post-Mortem Interval) (right panel) distribution in GTEx v8 cohort.

Author response image 8.

Sex (left panel), ethnicity (middle panel), and manner of death (DTHMNNR) (right panel) distribution in GTEx v8 cohort.

(4) Since this study is highly correlative, it is impossible to determine if the findings hold true without an independent cohort validation or experimental validation. They used the ROSMAP cohort for AD samples, and some splicing factors regulation but the generalizability to the age and sex effects have not been independently tested.

The reviewer raises an important point regarding the independent validation of sex- and age-associated splicing changes associated with AD. We used GTEx primarily because it includes approximately 17,000 RNA-seq samples across multiple human tissues, making it the most comprehensive public resource for studying population-level differences in age and sex. In particular, its large-scale brain samples provide a unique opportunity to analyze transcriptomic changes in sex-dimorphic aging.

We understand the reviewer’s concern that our findings are mainly supported by correlative evidence, which could be affected by dataset-specific biases. However, there are several technical issues in crossvalidation with transcriptomes across different datasets, including limited comparability due to cell type heterogeneity, postmortem artifacts, and sequencing biases.

Specifically, GTEx data is bulk RNA-seq that does not capture cell-type-specific transcriptomic changes. Given the cellular complexity of the brain and other tissues, observed differences in gene expression and splicing may be influenced by shifts in cellular composition rather than intrinsic transcriptional regulation. For example, we compared our results from GTEx whole blood with the analysis using an external dataset from Peripheral Blood Mononuclear Cells (PBMCs) provided by Shen et al. (2024) [3] (Author response image 2).  We observed limited overlap in differentially expressed genes between these datasets (probably because the whole blood contains diverse immune cell populations), highlighting the challenges in cross-dataset validation due to differences in tissue composition and sample processing.

Therefore, we applied surrogate variable analysis (SVA) to minimize technical and biological confounders. This approach helped reduce biases from genetic background to hidden batch effects, including postmortem artifacts, sequencing biases (Figure S4), and other covariates. This approach could help us identify whether sex-biased splicing events are biologically meaningful rather than technical artifacts.  

In addition, to address the reviewer’s concern on the splicing factor regulation, we managed to find a dataset in decision-related brain regions. Due to the limitation of human brain data covering different age and sex groups, we used mouse hippocampus datasets, including young and old, as well as female and male groups [7].  The analysis of protein levels from MS data identified sex-biased age-associated splicing factors, including Srsf1 and Srsf7.  We found that the changes are consistent with the findings from GTEx (Author response image 9), aligning with our sex-biased splicing factor expression during aging in the same region of the human brain. This cross-species consistency supports the robustness of our findings in human brain aging.

Author response image 9.

Protein levels of some male-specific splicing factors in human hippocampus quantified using MS data. The Y-axis shows the protein intensity. Different facets mean different sample batch sets. The yellow boxes indicate the protein levels in the young group, while the brown boxes indicate those in the old group.

In summary, despite the inherent limitations of RNA-seq studies in sex- and age-related transcriptomics, we have made our best efforts to address these concerns through comparisons with external datasets, statistical corrections, and validation using proteomic data. We appreciate the reviewer’s feedback and include additional discussion on these points (Page 13). 

(5) Are AS predictions from short-read data accurate enough to make the predictions the authors report? 

The reviewer is correct that the short-read sequencing has inherent limitations in reconstructing full-length isoforms.  However, the higher sequencing depth for short reads makes it a better choice in quantifying the relative change of each AS event across different conditions.  As a result, short-read data are extensively used in the splicing field to quantitatively measure the AS changes.  For this reason, we focused on the levels of alternative splicing events, rather than the quantification of full-length isoforms.  We used a series of stringent filters in our analyses to increase the reliability of our results.

Specifically, we filtered the read counts of the junction read counts (JC) of most differential AS events that were higher than 10, as mentioned in the Methods section. Also, we used our GPU-based gene expression quantification tool, Paean, which performed better in cross-validation with quantitative RT-PCR results. The results of Paean are consistent with other pipelines. We cited an updated version of Paean that included the comparison with other tools in analyzing AS for consistency.  The manuscript on the new Paean version is being reviewed in another journal, and we included the PDF of that manuscript (Fig. 3 in the Paean manuscript) in the revised documents. 

(6) Along the same lines, the finding that male age-related AS events are linked to Alzheimer's disease somewhat contradicts epidemiological studies that show that even after adjusting for age, women still have a greater risk of developing Alzheimer's than men. The authors show a significant overlap with AD GE events in females but don't explain the discrepancy. 

We appreciate the editor’s comment regarding these discrepancies with the epidemiological studies. Previous studies suggested that the disease manifestations of Alzheimer’s Disease (AD) showed sex differences in AD phenotypes, including cognitive decline and brain atrophy [8].  The analyses on the sex/age effect of AD are indeed pretty complex, depending on the molecular criteria (GE or AS vs epidemiological data) in distinct studies, probably due to the difficulty in capturing how environmental exposures interact with biological pathways.  We hope to bring up three related points regarding this concern, which were also discussed in the revised manuscript. 

• As we have mentioned in the Discussion section, an early study investigated the relationship between age, sex, and cognitive function in a large cohort of 17,127 UK Biobank participants [9]. Their study highlighted more apparent age-related changes in cognitive function among men, suggesting a potential vulnerability of men to cognitive decline with age.  Their main conclusion is consistent with our findings. 

• While men and women can both suffer from Alzheimer's disease, women are more likely to be diagnosed, possibly due to longer lifespans and potential differences in brain structure or other factors. Although women exhibit a higher overall risk of AD, they may also have distinct molecular compensatory mechanisms that influence disease progression. 

• To avoid the age effect, in our AD datasets, including ROSMAP, we filtered the samples over 90 years old to match the number of both sexes and the age distribution between the AD and control groups. Our analysis avoided the age biases in comparing AD and control, suggesting the crucial roles of sBASEs in AD during male aging.

Moreover, for gene expression (GE), we showed distinct patterns of AD-related genes in females with AS. These two molecular processes do not necessarily have the same functional impact. AS changes may precede or contribute to disease onset in different ways compared to GE alterations. Our study came up with the underlying mechanisms linking cognitive disorders and alternative splicing (AS) at a higher molecular resolution.   

(7) Could the authors explain which sBASE subset they used for their random forest prediction model and what was the rationale? 

We are sorry for missing the details in selecting sBASEs (sex-biased age-associated splicing events) for the random forest prediction model. We specifically used sBASEs that exhibited specific sex-biased changes in splicing associated with aging. This subset of sBASEs was chosen in terms of those that could also be detected in the ROSMAP AD dataset due to different sequencing depths or technical biases across datasets. These sBASEs were further input to a prediction model with the feature selection algorithm RFE, and then evaluated their contributions. In the revised manuscript, we added the details of this selection in the Methods (Page 7).

(8) The breakpoint analysis is particularly interesting. Can this be speculated to correlate with the recent non-linear multi-omic aging patterns observed by Shen et al in Nature Aging? 

Thank you for highlighting the interesting aspects of our breakpoint analysis and suggesting its potential correlation with the non-linear aging patterns observed by Shen et al. 

Shen et al. observed two prominent crests around the ages of 45 and 60 using omics data. Similarly, we also identified the non-linear aging patterns with two breakpoints in our analysis. However, there are some notable differences in specific breakpoints between these two studies, resulting from the breakpoint definition, as well as the sample preparations. According to the response in Author response image 2, the differences come from the following aspects:

The definition of breakpoints vs crests:

• Crests represent age-related molecular changes at each time point across the human lifespan. They indicate the number of molecules that are differentially expressed during aging (q < 0.05), without considering individual expression levels.

• Our breakpoints, in contrast, are identified after filtering the chronological trends based on the expression levels and calculating the rate of change at each age point using sliding windows. Breakpoints are defined as local maxima where the distance to the nearest minimum, relative to the global maximum, exceeds 10%. We indeed found some local wide peaks around 60 in some tissues, shown in Figure S10, however, we excluded these due to our strict cutoffs.

The sequenced biosamples: 

• Whole-blood vs Peripheral Blood Mononuclear Cells (PBMC): As mentioned in previous responses, in GTEx, whole blood samples from donors were sequenced, whereas their study used PBMCs. Whole blood contains all blood components, including red blood cells, platelets, granulocytes (e.g., neutrophils), lymphocytes, and monocytes, while PBMCs only represent a subset of white blood cells, primarily consisting of lymphocytes (T cells, B cells, NK cells) and monocytes, excluding granulocytes and erythrocytes. Gene expression changes observed in whole blood capture the contributions from neutrophils and other granulocytes, which are absent in PBMC analyses (as shown in Figure S11C and Author response image 2). Additionally, whole blood can serve as a readily accessible biomarker source for testing age-related diseases without the need for cell separation, making it a more practical option for clinical applications.

• For both studies, we share a tissue, which is skin, we looked at the non-linear changes during aging and found the same two breakpoints: 43 and 58. 

Sex-specific analysis in females and males:

• The main object of our analysis is to compare the differences in aging rates between sexes. Notably, the identified breakpoints may differ when sex effects are not taken into account, highlighting the importance of analyzing males and females separately.

We have added the following statements to further clarify this connection: ‘Our analysis observed the nonlinear aging patterns with two breakpoints, which is consistent with recent findings (Nature Aging, 2024), with differences in specific age points due to the sex differences as well as tissue diversities.’ (Page 14), and ‘These breakpoints could represent key junctures in the aging process that align with the non-linear patterns of aging and disease progression.’ (Page 15)

(9) Minor - the authors should refer to figures in the Discussion. They do so in some cases but this needs to be more extensive. 

Thank you for pointing this out. In response, we have reviewed the Discussion section and added references to relevant figures where appropriate. In the section discussing the discrepancies between the profiles of GE vs. AS, we now refer to Figure 3 to highlight the earlier onset of different transcriptomic resolutions (Page 12); When describing the sex-specific age-associated AS changes and their associations with Alzheimer’s disease, we have added references to Figure 4 (Page 12); In the discussion of estrogen-mediated regulation of splicing factors, we have referred to Figure 5A, which detail the construction of RBP-RNA regulatory network integrating muti-dimensional data obtained through several orthogonal state-of-the-art approaches (Page 14).

Reference:

(1) Ferreira, P.G. et al. The effects of death and post-mortem cold ischemia on human tissue transcriptomes. Nature communications 9, 490 (2018).

(2) Wucher, V., Sodaei, R., Amador, R., Irimia, M. & Guigó, R. Day-night and seasonal variation of human gene expression across tissues. PLoS Biology 21, e3001986 (2023).

(3) Shen, X. et al. Nonlinear dynamics of multi-omics profiles during human aging. Nature aging, 116 (2024).

(4) Zhou, X. et al. Splicing factor SRSF1 promotes gliomagenesis via oncogenic splice-switching of MYO1B. The Journal of clinical investigation 129, 676-693 (2019).

(5) Soheili-Nezhad, S., Ibáñez-Solé, O., Izeta, A., Hoeijmakers, J.H. & Stoeger, T. Time is ticking faster for long genes in aging. Trends in Genetics 40, 299-312 (2024).

(6) Brouillette, M. Gene length could be a critical factor in the aging of the genome. Proceedings of the National Academy of Sciences 121, e2416630121 (2024).

(7) Keele, G.R. et al. Global and tissue-specific aging effects on murine proteomes. Cell reports 42(2023).

(8) Ferretti, M.T. et al. Sex differences in Alzheimer disease—the gateway to precision medicine. Nature Reviews Neurology 14, 457-469 (2018).

(9) Foo, H. et al. Age-and sex-related topological organization of human brain functional networks and their relationship to cognition. Frontiers in aging neuroscience 13, 758817 (2021).

Author Response

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

eLife assessment

The work is a useful contribution towards understanding the role of archaeal and plant D-aminoacyl-tRNA deacylase 2 (DTD2) in deacylation and detoxification of D-Tyr-tRNATyr modified by various aldehydes produced as metabolic byproducts in plants. It integrates convincing results from both in vitro and in vivo experiments to address the long-standing puzzle of why plants outperform bacteria in handling reactive aldehydes and suggests a new strategy for stress-tolerant crops. The impact of the paper is limited by the fact that only one modified D-aminoacyl tRNA was examined, in lack of evidence that plant eEF1A mimics EF-Tu in protecting L-aminoacyl tRNAs from modification, and in failure to measure accumulation of toxic D-aminoacyl tRNAs or impairment of translation in plant cells lacking DTD2.

We have now addressed all the drawbacks as follows:

‘only one modified D-aminoacyl tRNA was examined’

We wish to clarify that only D-Leu (Yeast), D-Asp (Bacteria, Yeast), D-Tyr (Bacteria, Cyanobacteria, Yeast) and D-Trp (Bacteria) show toxicity in vivo in the absence of known DTD (Soutourina J. et al., JBC, 2000; Soutourina O. et al., JBC, 2004; Wydau S. et al., JBC, 2009) and D-Tyr-tRNATyr is used as a model substrate to test the DTD activity in the field because of the conserved toxicity of D-Tyr in various organisms. DTD2 has been shown to recycle D-Asp-tRNAAsp and D-Tyr-tRNATyr with the same efficiency both in vitro and in vivo (Wydau S. et al., NAR, 2007) and it also recycles acetaldehyde-modified D-Phe-tRNAPhe and D-Tyr-tRNATyr in vitro as shown in our earlier work (Mazeed M. et al., Science Advances, 2021). We have earlier shown that DTD1, another conserved chiral proofreader across bacteria and eukaryotes, acts via a side chain independent mechanism (Ahmad S. et al., eLife, 2013). To check the biochemical activity of DTD2 on D-Trp-tRNATrp, we have now done the D-Trp, D-Tyr and D-Asp toxicity rescue experiments by expressing the archaeal DTD2 in dtd null E. coli cells. We found that DTD2 could rescue the D-Trp toxicity with equal efficiency like D-Tyr and D-Asp (Figure: 1). Considering the action on multiple side chains with different chemistry and size, it can be proposed with reasonable confidence that DTD2 also operates based on a side chain independent manner.

Author response image 1.

DTD2 recycles multiple D-aa-tRNAs with different side chain chemistry and size. Growth of wildtype (WT), dtd null strain (∆dtd), and Pyrococcus horikoshii DTD2 (PhoDTD2) complemented ∆dtd strains of E. coli K12 cells with 500 µM IPTG along with A) no D-amino acids, B) 2.5 mM D-tyrosine, C) 30 mM D-aspartate and D) 5 mM D-tryptophan.

‘lack of evidence that plant eEF1A mimics EF-Tu in protecting L-aminoacyl tRNAs from modification’

To understand the role of plant eEF1A in protecting L-aa-tRNAs from aldehyde modification, we have done a thorough sequence and structural analysis. We analysed the aa-tRNA bound elongation factor structure from bacteria (PDB ids: 1TTT) and found that the side chain of amino acid in the amino acid binding site of EF-Tu is projected outside (Figure: 2A; 3A). In addition, the amino group of amino acid is tightly selected by the main chain atoms of elongation factor thereby lacking a space for aldehydes to enter and then modify the L-aa-tRNAs and Gly-tRNAs (Figure: 2B; 3B). Modelling of D-amino acid (D-phenylalanine and smallest chiral amino acid, D-alanine) in the same site shows serious clashes with main chain atoms of EF-Tu, indicating D-chiral rejection during aa-tRNA binding by elongation factor (Figure: 2C-E). Next, we superimposed the tRNA bound mammalian eEF-1A cryoEM structure (PDB id: 5LZS) with bacterial structure to understand the structural differences in terms of tRNA binding and found that elongation factor binds tRNA in a similar way (Figure: 3C-D). Modelling of D-alanine in the amino acid binding site of eEF-1A shows serious clashes with main chain atoms, indicating a general theme of D-chiral rejection during aa-tRNA binding by elongation factor (Figure: 2F; 3E). Structure-based sequence alignment of elongation factor from bacteria, archaea and eukaryotes (both plants and mammals) shows a strict conservation of amino acid binding site (Figure: 2G). This suggests that eEF-1A will mimic EF-Tu in protecting L-aa-tRNAs from reactive aldehydes. Minor differences near the amino acid side chain binding site (as indicated in Wolfson and Knight, FEBS Letters, 2005) might induce the amino acid specific binding differences (Figure: 3F). However, those changes will have no influence when the D-chiral amino acid enters the pocket, as the whole side chain would clash with the active site. We have now included this sequence and structural conservation analysis in our revised manuscript (in text: line no 107-129; Figure: 2 and S2). Overall, our structural analysis suggests a conserved mode of aa-tRNA selection by elongation factor across life forms and therefore, our biochemical results with bacterial elongation factor Tu (EF-Tu) reflect the protective role of elongation factor in general across species.

Author response image 2.

Elongation factor enantio-selects L-aa-tRNAs through D-chiral rejection mechanism. A) Surface representation showing the cocrystal structure of EF-Tu with L-Phe-tRNAPhe. Zoomed-in image showing the binding of L-phenylalanine with side chain projected outside of binding site of EF-Tu (PDB id: 1TTT). B) Zoomed-in image of amino acid binding site of EF-Tu bound with L-phenylalanine showing the selection of amino group of amino acid through main chain atoms (PDB id: 1TTT). C) Modelling of D-phenylalanine in the amino acid binding site of EF-Tu shows severe clashes with main chain atoms of EF-Tu. Modelling of smallest chiral amino acid, alanine, in the amino acid binding site of EF-Tu shows D) no clashes with L-alanine and E) clashes with D-alanine. F) Modelling of D-alanine in the amino acid binding site of eEF-1A shows clashes with main chain atoms. (*Represents modelled molecule). G) Structure-based sequence alignment of elongation factor from bacteria, archaea and eukaryotes (both plants and animals) showing conserved amino acid binding site residues. (Key residues are marked with red star).

Author response image 3.

Elongation factor protects L-aa-tRNAs from aldehyde modification. A) Cartoon representation showing the cocrystal structure of EF-Tu with L-Phe-tRNAPhe (PDB id: 1TTT). B) Zoomed-in image of amino acid binding site of EF-Tu bound with L-phenylalanine (PDB id: 1TTT). C) Cartoon representation showing the cryoEM structure of eEF-1A with tRNAPhe (PDB id: 5LZS). D) Image showing the overlap of EF-Tu:L-Phe-tRNAPhe crystal structure and eEF-1A:tRNAPhe cryoEM structure (r.m.s.d. of 1.44 Å over 292 Cα atoms). E) Zoomed-in image of amino acid binding site of eEF-1A with modelled L-alanine (PDB id: 5ZLS). (*Modelled) F) Overlap showing the amino acid binding site residues of EF-Tu and eEF-1A. (EF-Tu residues are marked in black and eEF-1A residues are marked in red).

‘failure to measure accumulation of toxic D-aminoacyl tRNAs or impairment of translation in plant cells lacking DTD2’

We agree that measuring the accumulation of D-aa-tRNA adducts from plant cells lacking DTD2 is important. We tried to characterise the same with dtd2 mutant plants extensively through Northern blotting as well as mass spectrometry. However, due to the lack of information about the tissue getting affected (root or shoot), identity of aa-tRNA as well as location of aa-tRNA (cytosol or organellar), we are so far unsuccessful in identifying them from plants. Efforts are still underway to identify them from plant system lacking DTD2. However, we have used a bacterial surrogate system, E. coli, as used earlier in Mazeed M. et al., Science Advances, 2021 to show the accumulation of D-aa-tRNA adducts in the absence of dtd. We could identify the accumulation of both formaldehyde and MG modified D-aa-tRNA adducts via mass spectrometry (Figure: 4). These results are now included in the revised manuscript (in line no: 190-197 and Figure: S5).

Author response image 4.

Loss of DTD results in accumulation of modified D-aminoacyl adducts on tRNAs in E. coli. Mass spectrometry analysis showing the accumulation of aldehyde modified D-Tyr-tRNATyr in A) Δdtd E. coli, B) formaldehyde and D-tyrosine treated Δdtd E. coli, and C) MG and D-tyrosine treated Δdtd E. coli. ESI-MS based tandem fragmentation analysis for unmodified and aldehyde modified D-Tyr-tRNATyr in D) Δdtd E. coli, E) and F) formaldehyde and D-tyrosine treated Δdtd E. coli, G) and H) MG and D-tyrosine treated Δdtd E. coli.

Response to Public Reviews:

We are grateful for the reviewers’ positive feedback and their comments and suggestions on this manuscript. Reviewer 1 has indicated two weaknesses and Reviewer 2 has none. We have now addressed all the concerns of the Reviewers.

Reviewer #1 (Public Review):

Summary:

This work is an extension of the authors' earlier work published in Sci Adv in 2001, wherein the authors showed that DTD2 deacylates N-ethyl-D-aminoacyl-tRNAs arising from acetaldehyde toxicity. The authors in this study, investigate the role of archaeal/plant DTD2 in the deacylation/detoxification of D-Tyr-tRNATyr modified by multiple other aldehydes and methylglyoxal (produced by plants). Importantly, the authors take their biochemical observations to plants, to show that deletion of DTD2 gene from a model plant (Arabidopsis thaliana) makes them sensitive to the aldehyde supplementation in the media especially in the presence of D-Tyr. These conclusions are further supported by the observation that the model plant shows increased tolerance to the aldehyde stress when DTD2 is overproduced from the CaMV 35S promoter. The authors propose a model for the role of DTD2 in the evolution of land plants. Finally, the authors suggest that the transgenic crops carrying DTD2 may offer a strategy for stress-tolerant crop development. Overall, the authors present a convincing story, and the data are supportive of the central theme of the story.

We are happy that reviewer found our work convincing and would like to thank the reviewer for finding our data supportive to the central theme of the manuscript.

Strengths:

Data are novel and they provide a new perspective on the role of DTD2, and propose possible use of the DTD2 lines in crop improvement.

We are happy for this positive comment on the manuscript.

Weaknesses:

(a) Data obtained from a single aminoacyl-tRNA (D-Tyr-tRNATyr) have been generalized to imply that what is relevant to this model substrate is true for all other D-aa-tRNAs (term modified aa-tRNAs has been used synonymously with the modified Tyr-tRNATyr). This is not a risk-free extrapolation. For example, the authors see that DTD2 removes modified D-Tyr from tRNATyr in a chain-length dependent manner of the modifier. Why do the authors believe that the length of the amino acid side chain will not matter in the activity of DTD2?

We thank the reviewer for bringing up this important point. As mentioned above, we wish to clarify that only half of the aminoacyl-tRNA synthetases are known to charge D-amino acids and only D-Leu (Yeast), D-Asp (Bacteria, Yeast), D-Tyr (Bacteria, Cyanobacteria, Yeast) and D-Trp (Bacteria) show toxicity in vivo in the absence of known DTD (Soutourina J. et al., JBC, 2000; Soutourina O. et al., JBC, 2004; Wydau S. et al., JBC, 2009). D-Tyr-tRNATyr is used as a model substrate to test the DTD activity in the field because of the conserved toxicity of D-Tyr in various organisms. DTD2 has been shown to recycle D-Asp-tRNAAsp and D-Tyr-tRNATyr with the same efficiency both in vitro and in vivo (Wydau S. et al., NAR, 2007). Moreover, we have previously shown that it recycles acetaldehyde-modified D-Phe-tRNAPhe and D-Tyr-tRNATyr in vitro as shown in our earlier work (Mazeed M. et al., Science Advances, 2021). We have earlier shown that DTD1, another conserved chiral proofreader across bacteria and eukaryotes, acts via a side chain independent mechanism (Ahmad S. et al., eLife, 2013). To check the biochemical activity of DTD2 on D-Trp-tRNATrp, we have now done the D-Trp, D-Tyr and D-Asp toxicity rescue experiments by expressing the archaeal DTD2 in dtd null E. coli cells. We found that DTD2 could rescue the D-Trp toxicity with equal efficiency like D-Tyr and D-Asp (Figure 1). Considering the action on multiple side chains with different chemistry and size, it can be proposed with reasonable confidence that DTD2 also operates based on a side chain independent manner.

(b) While the use of EFTu supports that the ternary complex formation by the elongation factor can resist modifications of L-Tyr-tRNATyr by the aldehydes or other agents, in the context of the present work on the role of DTD2 in plants, one would want to see the data using eEF1alpha. This is particularly relevant because there are likely to be differences in the way EFTu and eEF1alpha may protect aminoacyl-tRNAs (for example see description in the latter half of the article by Wolfson and Knight 2005, FEBS Letters 579, 3467-3472).

We thank the reviewer for bringing up this important point. As mentioned above, to understand the role of plant eEF1A in protecting L-aa-tRNAs from aldehyde modification, we have done a thorough sequence and structural analysis. We analysed the aa-tRNA bound elongation factor structure from bacteria (PDB ids: 1TTT) and found that the side chain of amino acid in the amino acid binding site of EF-Tu is projected outside (Figure: 2A; 3A). In addition, the amino group of amino acid is tightly selected by the main chain atoms of elongation factor thereby lacking a space for aldehydes to enter and then modify the L-aa-tRNAs and Gly-tRNAs (Figure: 2B; 3B). Modelling of D-amino acid (D-phenylalanine and smallest chiral amino acid, D-alanine) in the same site shows serious clashes with main chain atoms of EF-Tu, indicating D-chiral rejection during aa-tRNA binding by elongation factor (Figure: 2C-E). Next, we superimposed the tRNA bound mammalian eEF-1A cryoEM structure (PDB id: 5LZS) with bacterial structure to understand the structural differences in terms of tRNA binding and found that elongation factor binds tRNA in a similar way (Figure: 3C-D). Modelling of D-alanine in the amino acid binding site of eEF-1A shows serious clashes with main chain atoms, indicating a general theme of D-chiral rejection during aa-tRNA binding by elongation factor (Figure: 2F; 3E). Structure-based sequence alignment of elongation factor from bacteria, archaea and eukaryotes (both plants and mammals) shows a strict conservation of amino acid binding site (Figure: 2G). Minor differences near the amino acid side chain binding site (as indicated in Wolfson and Knight, FEBS Letters, 2005) might induce the amino acid specific binding differences (Figure: 3F). However, those changes will have no influence when the D-chiral amino acid enters the pocket, as the whole side chain would clash with the active site. We have now included this sequence and structural conservation analysis in our revised manuscript (in text: line no 107-129; Figure: 2 and S2). Overall, our structural analysis suggests a conserved mode of aa-tRNA selection by elongation factor across life forms and therefore, our biochemical results with bacterial elongation factor Tu (EF-Tu) reflect the protective role of elongation factor in general across species.

Reviewer #2 (Public Review):

In bacteria and mammals, metabolically generated aldehydes become toxic at high concentrations because they irreversibly modify the free amino group of various essential biological macromolecules. However, these aldehydes can be present in extremely high amounts in archaea and plants without causing major toxic side effects. This fact suggests that archaea and plants have evolved specialized mechanisms to prevent the harmful effects of aldehyde accumulation.

In this study, the authors show that the plant enzyme DTD2, originating from archaea, functions as a D-aminoacyl-tRNA deacylase. This enzyme effectively removes stable D-aminoacyl adducts from tRNAs, enabling these molecules to be recycled for translation. Furthermore, they demonstrate that DTD2 serves as a broad detoxifier for various aldehydes in vivo, extending its function beyond acetaldehyde, as previously believed. Notably, the absence of DTD2 makes plants more susceptible to reactive aldehydes, while its overexpression offers protection against them. These findings underscore the physiological significance of this enzyme.

We thank the reviewer for the positive comments the manuscript.

Response to recommendation to authors:

Reviewer #1 (Recommendations For The Authors):

I enjoyed reading the manuscript entitled, "Archaeal origin translation proofreader imparts multi aldehyde stress tolerance to land plants" from the Sankaranarayanan lab. This work is an extension of their earlier work published in Sci Adv in 2001, wherein they showed that DTD2 deacylates N-ethyl-D-aminoacyl-tRNAs arising from acetaldehyde toxicity. Now, the authors of this study (Kumar et al.) investigate the role of archaeal/plant DTD2 in the deacylation/detoxification of D-Tyr-tRNATyr modified by multiple other aldehydes and methylglyoxal (which are produced during metabolic reactions in plants). Importantly, the authors take their biochemical observations to plants, to show that deletion of DTD2 gene from a model plant (Arabidopsis thaliana) makes them sensitive to the aldehyde supplementation in the media especially in the presence of D-Tyr. These conclusions are further supported by the observation that the model plant shows increased tolerance to the aldehyde stress when DTD2 is overproduced from the CaMV 35S promoter. The authors propose a model for the role of DTD2 in the evolution of land plants. Finally, the authors suggest that the transgenic crops carrying DTD2 may offer a strategy for stress-tolerant crop development. Overall, the authors present a convincing story, and the data are supportive of the central theme of the story.

We are happy that reviewer enjoyed our manuscript and found our work convincing. We would also like to thank reviewer for finding our data supportive to the central theme of the manuscript.

I have the following observations that require the authors' attention.

1) The title of the manuscript will be more appropriate if revised to, "Archaeal origin translation proofreader, DTD2, imparts multialdehyde stress tolerance to land plants".

Both the reviewer’s suggested to change the title. We have now changed the title based on reviewer 2 suggestion.

2) Abstract (line 19): change, "physiologically abundantly produced" to "physiologically produced".

As per the reviewer’s suggestion, we have now changed it to "physiologically produced".

3) Introduction (line 50): delete, 'extremely'.

We have removed the word 'extremely' from the Introduction.

4) Line 79: change, "can be utilized" to "may be explored".

We have changed "can be utilized" to "may be explored" as suggested by the reviewers.

5) Results in general:

(a) Data obtained from a single aminoacyl-tRNA (D-Tyr-tRNATyr) have been generalized to imply that what is relevant to this model substrate is true for all other D-aa-tRNAs (term modified aa-tRNAs has been used synonymously with the modified D-Tyr-tRNATyr). This is a risky extrapolation. For example, the authors see that DTD2 removes modified D-Tyr from tRNATyr in a chain-length dependent manner of the modifier. Why do the authors believe that the length of the amino acid side chain will not matter in the activity of DTD2?

We thank the reviewer for bringing up this important point. As mentioned above, we wish to clarify that only half of the aminoacyl-tRNA synthetases are known to charge D-amino acids and only D-Leu (Yeast), D-Asp (Bacteria, Yeast), D-Tyr (Bacteria, Cyanobacteria, Yeast) and D-Trp (Bacteria) show toxicity in vivo in the absence of known DTD (Soutourina J. et al., JBC, 2000; Soutourina O. et al., JBC, 2004; Wydau S. et al., JBC, 2009). D-Tyr-tRNATyr is used as a model substrate to test the DTD activity in the field because of the conserved toxicity of D-Tyr in various organisms. DTD2 has been shown to recycle D-Asp-tRNAAsp and D-Tyr-tRNATyr with the same efficiency both in vitro and in vivo (Wydau S. et al., NAR, 2007). Moreover, we have previously shown that it recycles acetaldehyde-modified D-Phe-tRNAPhe and D-Tyr-tRNATyr in vitro as shown in our earlier work (Mazeed M. et al., Science Advances, 2021). We have earlier shown that DTD1, another conserved chiral proofreader across bacteria and eukaryotes, acts via a side chain independent mechanism (Ahmad S. et al., eLife, 2013). To check the biochemical activity of DTD2 on D-Trp-tRNATrp, we have now done the D-Trp, D-Tyr and D-Asp toxicity rescue experiments by expressing the archaeal DTD2 in dtd null E. coli cells. We found that DTD2 could rescue the D-Trp toxicity with equal efficiency like D-Tyr and D-Asp (Figure 1). Considering the action on multiple side chains with different chemistry and size, it can be proposed with reasonable confidence that DTD2 also operates based on a side chain independent manner.

(b) Interestingly, the authors do suggest (in the Materials and Methods section) that the experiments were performed with Phe-tRNAPhe as well as Ala-tRNAAla. If what is stated in Materials and Methods is correct, these data should be included to generalize the observations.

We regret for the confusing statement. We wish to clarify that L- and D-Tyr-tRNATyr were used for checking the TLC-based aldehyde modification, EF-Tu based protection assays and deacylation assays, D-Phe-tRNAPhe was used to characterise aldehyde-based modification by mass spectrometry and L-Ala-tRNAAla was used to check the modification propensity of multiple aldehydes. We used multiple aa-tRNAs to emphasize that aldehyde-based modifications are aspecific towards the identity of aa-tRNAs. All the data obtained with respective aa-tRNAs are included in manuscript.

(c) While the use of EFTu supports that the ternary complex formation by the elongation factor can resist modifications of L-Tyr-tRNATyr by the aldehydes or other agents, in the context of the present work on the role of DTD2 in plants, one would want to see the data using eEF1alpha. This is particularly relevant because there are likely to be differences in the way EFTu and eEF1alpha may protect aminoacyl-tRNAs (for example see description in the latter half of the article by Wolfson and Knight 2005, FEBS Letters 579, 3467-3472).

We thank the reviewer for bringing up this important point. As mentioned above, to understand the role of plant eEF1A in protecting L-aa-tRNAs from aldehyde modification, we have done a thorough sequence and structural analysis. We analysed the aa-tRNA bound elongation factor structure from bacteria (PDB ids: 1TTT) and found that the side chain of amino acid in the amino acid binding site of EF-Tu is projected outside (Figure: 2A; 3A). In addition, the amino group of amino acid is tightly selected by the main chain atoms of elongation factor thereby lacking a space for aldehydes to enter and then modify the L-aa-tRNAs and Gly-tRNAs (Figure: 2B; 3B). Modelling of D-amino acid (D-phenylalanine and smallest chiral amino acid, D-alanine) in the same site shows serious clashes with main chain atoms of EF-Tu, indicating D-chiral rejection during aa-tRNA binding by elongation factor (Figure: 2C-E). Next, we superimposed the tRNA bound mammalian eEF-1A cryoEM structure (PDB id: 5LZS) with bacterial structure to understand the structural differences in terms of tRNA binding and found that elongation factor binds tRNA in a similar way (Figure: 3C-D). Modelling of D-alanine in the amino acid binding site of eEF-1A shows serious clashes with main chain atoms, indicating a general theme of D-chiral rejection during aa-tRNA binding by elongation factor (Figure: 2F; 3E). Structure-based sequence alignment of elongation factor from bacteria, archaea and eukaryotes (both plants and mammals) shows a strict conservation of amino acid binding site (Figure: 2G). Minor differences near the amino acid side chain binding site (as indicated in Wolfson and Knight, FEBS Letters, 2005) might induce the amino acid specific binding differences (Figure: 3F). However, those changes will have no influence when the D-chiral amino acid enters the pocket, as the whole side chain would clash with the active site. We have now included this sequence and structural conservation analysis in our revised manuscript (in text: line no 107-129; Figure: 2 and S2). Overall, our structural analysis suggests a conserved mode of aa-tRNA selection by elongation factor across life forms and therefore, our biochemical results with bacterial elongation factor Tu (EF-Tu) reflect the protective role of elongation factor in general across species.

6) Results (line 89): Figure: 1C-G (not B-G).

As correctly pointed out by the reviewer(s), we have changed it to Figure: 1C-G.

7) Results (line 91): Figure: S1B-G (not C-G).

We wish to clarify that this is correct.

8) Line 97: change, "propionaldehyde" to "propionaldehyde (Figure: 1H)".

As per the reviewer’s suggestion, we have now changed, "propionaldehyde" to "propionaldehyde (Figure: 1H)".

9) Line 124: The statement, "DTD2 cleaved all modified D-aa-tRNAs at 50 pM to 500 nM range (Figure: 2A_D)" is not consistent with the data presented. For example, Figure 2D does not show any significant cleavage. Figure S2A-B also does not show cleavage.

We thank the reviewers for pointing this out. We have changed the sentence to “DTD2 cleaved majority of aldehyde modified D-aa-tRNAs at 50 pM to 500 nM range".

10) Line 131: Cleavage observed in Fig. S2E is inconsistent with the generalized statement on DTD1.

We wish to clarify that the minimal activity seen in Fig. S2E is inconsistent with the general trend of DTD1’s biochemical activity seen on modified D-aa-tRNAs. In addition, we have earlier shown that D-aa-tRNA fits snugly in the active site of DTD1 (Ahmad S. et al., eLife, 2013) whereas the modified D-aa-tRNA cannot bind due to the space constrains in the active site of DTD1 (Mazeed M. et al., Science Advances, 2021). Therefore, this minimal activity could be a result of technical error during this biochemical experiment and could be considered as no activity.

11) Lines 129-133: Citations of many figure panels particularly in the supplementary figures are inconsistent with generalized statements. This section requires a major rewrite or rearrangement of the figure panels (in case the statements are correct).

We thank the reviewers for bringing forth this point and we have accordingly modified the statement into “DTD2 from archaea recycled short chain aldehyde-modified D-aa-tRNA adducts as expected (Figure: 3E-G) and, like DTD2 from plants, it did not act on aldehyde-modified D-aa-tRNAs longer than three chains (Figure: 3H; S3C-D; S4G-L)”.

12) Line 142: I don't believe one can call PTH a proofreader. Its job is to recycle tRNAs from peptidyl-tRNAs.

We thank the reviewers for pointing out this very important point. This is now corrected.

13). Line 145: change, "DTD2 can exert its protection for" to "DTD2 may exert protection from".

As per the reviewer’s suggestion, we have now changed"DTD2 can exert its protection for" to "DTD2 may exert protection from".

14) Line 148: change, "a homozygous line (Figure: 3A) and checked for" to "homozygous lines (Figure: 3A) and checked them for".

As per the reviewer’s suggestion, we have now changed, "a homozygous line (Figure: 3A) and checked for" to "homozygous lines (Figure: 3A) and checked them for".

15) Line 148: Change, the sentence beginning with dtd2 as follows. Similar to earlier results30-32, dtd2-/- (dtd2 hereafter) plants were susceptible to ethanol (Figure: S4A) confirming the non-functionality DTD2 gene in dtd2 plants.

As per the reviewer’s suggestion, we have now changed the sentence accordingly.

16) Line 161: change, "linked" to "associated".

As per the reviewer’s suggestion, we have now changed "linked" to "associated".

17) Lines 173-176: It would be interesting to know how well the DTD2 OE lines do in comparison to the other known transgenic lines developed with, for example, ADH, ALDH, or AOX lines. Any ideas would help appreciate the observation with DTD2 OE lines!

We greatly appreciate the reviewer’s suggestion. We have not done any comparison experiment with any transgenic lines so far. However, it can be potentially done in further studies with DTD2 OE lines.

18) Line 194: change, "necessary" with "present".

As per the reviewer’s suggestion, we have now changed "necessary" with "present".

19) Line 210: what is meant by 'huge'? Would 'significant' sound better?

As per the reviewer’s suggestion, we have now changed "huge" with "significant".

20) Lines 239-243: This needs to be rephrased. Isn't alpha carbonyl of the carboxyl group that makes ester bond with the -CCA end of the tRNA required for DTD2 activity as well? Are you referring to the carbonyl group in the moiety that modifies the alpha-amino group? Please clarify. The cited reference (no. 64) of Atherly does not talk about it.

We regret for the confusing statement. To clarify, we were referencing to the carbonyl carbon of the modification post amino group of the amino acid in aa-tRNAs (Figure: 5). We have now included a figure (Figure: S4Q of revised manuscript) to show the comparison of the carbonyl group for the better clarity. The cited reference Atherly A. G., Nature, 1978 shows the activity of PTH on peptidyl-tRNAs and peptidyl-tRNAs possess carbonyl carbon at alpha position post amino group of amino acid in L-aa-tRNAs.

Author response image 5.

Figure showing the difference in the position of carbonyl carbon in acetonyl and acetyl modification on aa-tRNAs.

21) Line 261: thrive (not thrives).

As per the reviewer’s suggestion, we have now changed it to thrive.

22) In Fig3A: second last lane, it should be dtd-/-:: AtDTDH150A (not dtd-/-:: AtDTDH150A).

We thank the reviewers for pointing out this, we have corrected it.

23). Materials and methods: Please clarify which experiments used tRNAPhe, tRNAAla, PheRS, etc. Also, please carefully check all other details provided in this section.

As per the reviewer’s suggestion, we would like to provide a table below explaining the use of different substrates as well as enzymes in our experiments.

Author response table 1.

24) Figure legends (many places): p values higher than 0.05 (not less than) are denoted as ns.

We thank the reviewers for pointing out this. We have corrected it.

Reviewer #2 (Recommendations For The Authors):

I have only minor comments for the authors:

Title: I would replace "Archeal origin translation proofreader" with " A translation proofreader of archeal origin"

As per the reviewer’s suggestion, we have now changed the title.

Abstract: This section could benefit from some rewriting. For instance, at the outset, the initial logical connection between the first and second sentences of the abstract is somewhat unclear. At the very least, I would suggest swapping their order to enhance the narrative flow. Later in the text, the term "chiral proofreading systems" is introduced; however, it is only in a subsequent sentence that these systems are explained to be responsible for removing stable D-aminoacyl adducts from tRNA. Providing an immediate explanation of these systems would enhance the reader's comprehension. The authors switch from the past participle tense to the present tense towards the end of the text. I would recommend that they choose one tense for consistency. In the final sentence, I would suggest toning down the statement and replacing "can be used" with "could be explored." (https://www.nature.com/articles/d41586-023-02895-w). The same comment applies to the introduction, line 79.

As per the reviewer’s suggestion, we have now changed the abstract appropriately.

General note: Conventionally, the use of italics is reserved for the specific species "Arabidopsis thaliana," while the broader genus "Arabidopsis" is not italicized.

We acknowledge the reviewer for this pertinent suggestion. This is now corrected in revised version of our manuscript.

General note: I would advise the authors against employing bold characters in conjunction with colors in the figures.

We thank the reviewer for this suggestion. We have now changed it appropriately in revised version of our manuscript.

Figure 1A: I recommend including the concentrations of the various aldehydes used in the experiment within the figure legend. While this information is available in the materials and methods section, it would be beneficial to have it readily accessible when analyzing the figure.

As per the reviewer’s suggestion, we have now included the concentrations in figure legend.

Figure 1I, J: some error bars are invisible.

We thank the reviewers for pointing out this, we have corrected it.

Figure 2M: The table could be simplified by removing aldehydes for which it was not feasible to demonstrate activity. The letter "M" within the cell labeled "aldehydes" appears to be a typographical error, presumably indicating the figure panel.

As per the reviewer’s suggestion, we have now changed this appropriately.

Figure 3: For consistency with the other panels in the figure, I recommend including an additional panel to display the graph depicting the impact of MG on germination.

As per the reviewer’s suggestion, we have now changed this appropriately.

Figure 4: Considering that only one plant is presented, it would be beneficial to visualize the data distribution for the other plants used in this experiment, similar to what the authors have done in panel A of the same figure.

We thank the reviewer for bringing up this point. We wish to clarify that we have done experiment with multiple plants. However, for the sake of clarity, we have included the representative images. Moreover, we have included the quantitative data for multiple plants in Figure 3C-G.

Figure 5E: The authors may consider presenting a chronological order of events as they believe they occurred during evolution.

We thank the reviewer for the suggestion. However, it is very difficult to pinpoint the chronology of the events. Aldehydes are lethal for systems due to their hyper reactivity and systems would require immediate solutions to survive. Therefore, we think that both problem (toxic aldehyde production) and its solution (expansion of aldehyde metabolising repertoire and recruitment of archaeal DTD2) might have appeared simultaneously.

Figure 6: The model appears somewhat crowded, which may affect its clarity and ease of interpretation. The authors might also consider dividing the legend sentence into two separate sentences for better readability.

As per the reviewer’s suggestion, we have now changed this appropriately.

Line 149: I recommend explicitly stating that ethanol metabolism produces acetaldehyde. This clarification will help the general reader immediately understand why DTD2 mutant plants are sensitive to ethanol.

As per the reviewer’s suggestion, we have now changed this appropriately.

Line 289: there is a typographical error, "promotor" instead of the correct term "promoter.".

We thank the referee for pointing out this, we have now corrected it.

Figure S5: The root morphology of DTD2 OE plants appears to exhibit some differences compared to the WT, even in the absence of a high concentration of aldehydes. It would be valuable if the authors could comment on these observed differences unless they have already done so, and I may have overlooked it.

We thank the referee for pointing out this. We do see minor differences in root morphology, but they are more pronounced with aldehyde treatments. The reason for this phenotype remains elusive and we are trying to understand the role of DTD2 in root development in detail in further studies.

Some Curiosity Questions (not mandatory for manuscript acceptance):

1) Do DTD2 OE plants display an earlier flowering phenotype than wild-type Col-0?

We have not done detailed phenotyping of DTD2 OE plants. However, our preliminary observations suggest no differences in flowering pattern as compared to wild-type Col-0.

2) What is the current understanding of the endogenous regulation of DTD2?

We have not done detailed analysis to understand the endogenous regulation of DTD2.

3) Could the protective phenotype of DTD2 OE plants in the presence of aldehydes be attributed to additional functions of this enzyme beyond the removal of stable D-aminoacyl adducts from tRNAs?

Based on the available evidence regarding the biochemical activity and in vivo phenotypes of DTD2, it appears that removal of stable D-aminoacyl adducts from tRNA is key for the protective phenotype of DTD2 OE.

A Suggestion for Future Research (not required for manuscript acceptance):

The authors could explore the possibility of overexpressing DTD2 in pyruvate decarboxylase transgenic plants and assess whether this strategy enhances flood tolerance without incurring a growth penalty under normal growth conditions.

We thank the referee for this interesting suggestion for future research. We will surely keep this in mind while exploring the flood tolerance potential of DTD2 OE plants.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

In this study, Yue et al. re-processed publicly available DNA methylation data (published in 2012 and 2017 from the Meissner lab) from pre- and post-implantation mouse embryos. Against the global wave of genome-wide reduction of DNA methylation occurring during pre-implantation development, they detected a slight increase (~1% on average) of DNA methylation at gene promoter regions during the transition from 8-cell to blastocyst stage. They claim that many such promoters are located in the X chromosome. Subsequently, they knocked down Dnmt3b (presumably because of its upregulation during the transition from the 8-cell to blastocyst stage) and detected the aberrant patterning of H3K27me3 in the mutant female embryos. Based on this observation, they claim that imprinted X-chromosome inactivation is impaired in the Dnmt3b-Kd pre-implantation embryos. Finally, they propose a model where such an increase of DNA methylation together with H3K27me3 regulates imprinted X-chromosome inactivation in the pre-implantation embryos. While their observation is of potential interest, the current version of the work fails to provide enough evidence to support their conclusions. Below are suggestions and comments on the manuscript.

Major issues:

(1) Sex of the embryos of the genome-wide bisulfite-sequencing data

The authors re-analyzed publicly available genome-wide DNA methylation data from the Meissner lab published in 2012 and 2017. The former used reduced representation bisulfite sequencing (RRBS) and the latter used whole-genome bisulfite sequencing (WGBS). Based mainly on the RRBS data, Yue et al. detected de novo DNA methylated promoters during the transition from 8-cell to blastocyst against the global wave of genome-wide DNA demethylation. They claim that such promoter regions are enriched at the "inactive" X chromosome. However, it would be difficult to discuss DNA methylation at inactive X-chromosomes as the RRBS data were derived from a mixture of male and female embryos. It would also be notable that the increase of DNA methylation at these promoter regions is ~1% on average. Such a slight increase in DNA methylation during pre-implantation development could also be due to the developmental variations between the embryos or between the sexes of embryos.

Thanks so much for your insightful comments. Whether de novo DNA methylation occurs in a sex-dimorphic manner would be of significance for our study. Based on your comments, we have added a reanalysis based on a publicly available single cell multi-omics sequencing (COOL-seq) data of mouse early embryos (Guo et al., 2017). The results showed that both male and female embryonic cells gain DNA methylation during the transition from the 8-cell to ICM (Figure 1—figure supplement 1C-D; Lines 112-115 in the revised manuscript).

With regards to the increase in the promoter region, many previous studies have revealed that promoter and overlapping CGI regions, especially high CpG promoters, always showed low levels of DNA methylation (Auclair et al., 2014; Borgel et al., 2010; Dahlet et al., 2020). The relatively lower basal levels make the increase seem relatively slight. Thus, we added relevant statements to clarify this information and rewritten the sentences in the revised manuscript (Lines 116-118, 125-127 in the revised manuscript).

In addition, using the single cell COOL-seq data, we also specifically reanalyzed the DNA methylation changes on the X chromosome in female embryos. The X chromosome showed a more notable increase than that on autosomes, and the female X chromosome showed a higher DNA methylation level than that of the male (Figure 3—figure supplement 2A-B; Lines 203-206 in the revised manuscript).

Thanks again for your insightful and constructive comments that significantly strengthen our evidence. We have added these results in the revised manuscript.

(2) Imprinted X-chromosome inactivation and evaluation of H3K27me3 (related to Figures 2C, D; 3F; Figure2-supplement 2 F, G; Figure3-supplement 3G)

Based on the slight change in the H3K27me3 signals in the Dnmt3b-Kd blastocysts, the authors claim that imprinted X-chromosome inactivation is impaired in the mutant embryo. It would be not easy to reach this conclusion from such a rough analysis of H3K27me3 presented in Figure 2C, D. Rigorous quantification/evaluation of the H3K27me3 signals in the Dnmt3b-Kd embryos should be considered. Additional evidence for the impairment of H3K27me3 in the mutant embryos should also be provided (expression of a subset of X-linked genes by RNA-FISH or RT-PCR etc.). Though technically challenging, high-resolution genome-wide approach such as ChIP-seq of H3K27me3 in the Dnmt3b-kd female embryos (with traceable SNPs between maternal and paternal X chromosome to distinguish inactive and active X-chromosome) could more precisely evaluate regions that lose H3K27me3 in the X-chromosome (de novo DNA methylated promoters from 8-cell to blastocyst, for example).

Thanks so much for your insightful comments that make our results more convincing. The H3K27me3 domain is a classic marker for establishment of XCI by achieving X chromosome wide heterochromatinization of transcriptional depression (Chow and Heard, 2009; Heard et al., 2004; Huynh and Lee, 2005). Thus, in the present study, we have performed immunostaining for H3K27me3 domains to evaluate the iXCI status in the blastocysts, as previously reported (Fukuda et al., 2014; Gontan et al., 2018; Inoue et al., 2010; Tan et al., 2016). Base on your comments, we have added another statistical method to quantify the establishment of iXCI, i.e. the percentage of H3K27me3-positive and -negative cells to total trophoblast cells in female blastocysts subject to Dnmt3b knockdown or not. The result also indicated that Dnmt3b knockdown led to a significant loss of H3K27me3 domains from total trophoblast cells. Similarly, new data based on statistical analyses of total trophoblast cells, has also been added in the results of Dnmt3b knockout and 5-aza-dC (Figure 3F; Figure 3—figure supplement 3D, H in the revised manuscript).

To clarify the significance and reliability of detecting H3K27me3 domains, we have added a schematic diagram depicting the process of iXCI initiation and establishment, as well as the experimental design and work flows, to make our results easier to be understood (Figure 3C in the revised manuscript).

In addition, we agree with your comments that additional evidence will benefit the conclusion. Thus, we have reanalyzed the RNA-seq and H3K27me3 CHIP-seq data in extraembryonic ectoderm (ExE) of E6.5 single embryos that underwent Dnm3a/3b knockout because preimplantation iXCI status maintains extraembryonic cells (Chen et al., 2019; Galupa and Heard, 2015; Schulz and Heard, 2013). The results showed that Dnmt knockout-induced chromosome-wide loss of DNA methylation led to a nearly complete loss of H3k27me3 on paternal X chromosome (specifically inactivated in iXCI), along with a notable transcriptional upregulation cross the chromosome. By contrast, these changes cannot be not observed on maternal X chromosome.

We have added this result in the revised manuscript (Lines 253-261; Figure 3—figure supplement 4A in the revised manuscript).

(3) Analysis of the developmental potential of Dnmt3b-kd embryos

While the authors claim that Dnmt3b-mediated de novo DNA methylation plays an important role in imprinted X-chromosome inactivation, it remains unclear whether the analysis presented in Figure 4 is derived from "female" embryos. This analysis seemed confusing as the authors claim that de novo DNA methylation in the promoter regions during the transition from 8-cell to blastocyst regulates imprinted X-chromosome inactivation, but this should not happen in the male embryos. Was the impairment of embryonic proliferation and differentiation observed in both male and female embryos? Or is this specific to the female embryos? We think that the sex of the embryos would be critical for the analysis presented in Figure 4.

Thanks so much for your constructive comments to make our results smoother and clearer. The Figure 4 mainly presents the developmental role of minor de novo methylation based on the integrated analysis of DNA methylation and gene expression dynamics from the 8-cell to ICM. Because our data indicated that both male and female embryos undergo minor de novo methylation (Figure 1—figure supplement 1C-D in the revised manuscript). This section mainly focused on genome wide and general changes, but not on sex dimorphic consequence.

To avoid the possible confusion, we have reorganized the RESULTS AND DISCUSSION section and presented this section as Figure 2 in the revised manuscript, before the chromosomal distribution analysis and subsequent detection relevant to iXCI.

Reviewer #2 (Public Review):

Summary:

Here, Yue et al. set out to determine if the low DNMT3B expression that is observed prior to de novo DNA methylation (before the blastocyst stage) has a function. Re-analyzing existing DNA methylation data from Smith et al. (2012) they find a small DNA methylation gain over a subset of promoters and gene bodies, occurring between the 8-cell and blastocyst stages, and refer to this as "minor de novo DNA methylation". They attempt to assess the relevance/functionality of this minor DNA methylation gain, and report reduced H3K27me3 in Dnmt3b knockdown (KD) trophoblast cells that normally undergo imprinted X-chromosome inactivation (iXCI) before the blastocyst stage. In addition, they assess the proliferation, differentiation, metabolic function, implantation rate, and live birth rate of Dnmt3b KD blastocysts.

Strengths:

Working with early embryos is technically demanding, making the well-designed experiments from this manuscript useful to the epigenetics community. Particularly, the DNMT3B expression and 5-mC staining at different embryonic stages.

Thanks for your positive evaluation, we have revised manuscript based on your comments, and the items need to be addressed in detail are explained in the point-by-point response to each comment.

Weaknesses:

- Throughout the manuscript, please represent DNA methylation changes as delta DNA methylation instead of fold change.

Thanks so much for your constructive comments. We have represented DNA methylation changes as “ΔDNA methylation” (Figure 2—figure supplement 1A; Figure 3—figure supplement 1A; Figure 3—figure supplement 3I in the revised manuscript).

- Detailed methods on the re-analysis of the DNA methylation data from Smith et al. 2012 are missing from the materials and methods section. Was a minimum coverage threshold used?

Thanks so much for your reminder. We have added relevant statements and provided the detail of the coverage criteria in the subsection of Bioinformatics analysis in the Materials and methods section as follows: RRBS data of mouse embryos (2-cell embryos, 4-cell embryos, 8-cell embryos, ICM, and E6.5 embryos) were downloaded from the published article by Smith et al (Smith et al., 2012) (accession number: GSE34864). The methylation level was calculated as the number of “methylated” reads (reporting as C), divided by the total number of “methylated” and “unmethylated” read, which reporting as C or T. The genomic region information was downloaded from the mm9 Repeat Masker. As described in the published article, promoters were defined as 1 kb up- and downstream of the TSS and classified into high-density CpG promoter (HCP), intermediate-density CpG promoter (ICP) and low-density CpG promoter (LCP). Only CpG sites with at least fivefold coverage were included in the methylation analysis. We have added relevant information in the revised manuscript (Lines 462-470 in the revised manuscript).

- Detailed methods on the establishment and validation of Dnmt3b KO blastocysts and 5-aza-dC treated blastocysts are missing (related to Figure 2).

Thanks so much for your detailed reminder. In the present study, we used a well-established Dnmt3b-deficient mouse model (Okano et al., 1999) to validate the role of minor de novo DNA methylation in iXCI establishment. Heterozygous Dnmt3b^+/- mice that carry one mutant locus of Dnmt3b, were obtained from the Mutant Mouse Resource & Research Centers (MMRRC, NIH). Homozygous embryos were obtained by intercrossing Dnmt3b^+/- male and female mice. Genotyping assays of collected embryos was performed by PCR using primers that were designed based on the gene targeting strategy following the MMRRC genotyping protocol (https://www.med.unc.edu/mmrrc/genotyping-protocols/mmrrc-center-protocol-29886/). We have provided the detailed methods in the revised manuscript (Lines 350-354; 391-393 in the revised manuscript). In addition, we added a schematic diagram depicting the processes of embryo collection and detection (Figure 3—figure supplement 3A in the revised manuscript).

Similarly, we have provided relevant details of 5-aza-dC supplementation in the revised manuscript (Lines 412-415 in the revised manuscript) and added a schematic diagram depicting the details of experimental design and processes (Figure 3—figure supplement 3E in the revised manuscript).

- Detailed methods on the re-analysis of the ChIPseq data from Liu et al. 2016 are missing from the materials and methods section.

Thank you for pointing this out. The bigwig files of H3K27me3 ChIP-seq data were downloaded from the published article by Liu et al (Liu et al., 2016)(accession number: GSE73952). These signal tracks were generated using the MACS2 (v2.0.10.20131216) pileup function and normalized to 1 million reads for visualization, as described in the original publication. We have added relevant information to the MATERIALS AND METHODS section in the revised manuscript (Lines 474-479 in the revised manuscript).

- Some of the data represented in bar graphs does not look convincing/significant. Maybe this data can be better represented differently, such as in box plots or violin plots, which would better represent the data.

Thanks so much for your comments that improve our result presentation, relevant results have been changed into box plots in the revised manuscript (Figure 3E; Figure 3—figure supplement 3C; Figure 3—figure supplement 3G in the revised manuscript). In addition, to strengthen our evidence, we have added alternative statistical method to quantify the establishment of iXCI, i.e. the percentage of H3K27me3-positive and -negative cells to total trophoblast cells in female blastocysts subject to Dnmt3b knockdown or not. (Figure 3F; Figure 3—figure supplement 3D, H in the revised manuscript).

- The relevance and rationale for experiments using 5-aza-dC treatment is unclear.

Thanks so much for reminding us to make our results more informative and convincing. 5-aza-dC is a well-established global DNA hypomethylating agent that efficiently inhibit the activity of all DNMTs, and thus has been frequently used to study the maintenance of DNA methylation and de novo DNA methylation (Maslov et al., 2012; Oka et al., 2005).

In our study, to validate the function of minor de novo DNA methylation in iXCI, we take advantage of 5-aza-dC-induced DNMT inhibition, which allows us, despite its inhibitory effect common to various DNMTs, to transiently treat embryos specifically during the window of minor de novo DNA methylation (from the 8-cell to blastocyst stage). We have added these statements, as well as a schematic diagram depicting the experimental design, in the revised manuscript to make our experiments more rational and easier to be understood (Lines 183-188; Figure 3—figure supplement 3E in the revised manuscript).

References

Auclair, G., Guibert, S., Bender, A. and Weber, M. (2014). Ontogeny of CpG island methylation and specificity of DNMT3 methyltransferases during embryonic development in the mouse. Genome Biol. 15, 545.

Borgel, J., Guibert, S., Li, Y., Chiba, H., Schubeler, D., Sasaki, H., Forne, T. and Weber, M. (2010). Targets and dynamics of promoter DNA methylation during early mouse development. Nat. Genet. 42, 1093-1100.

Chen, Z., Yin, Q., Inoue, A., Zhang, C. and Zhang, Y. (2019). Allelic H3K27me3 to allelic DNA methylation switch maintains noncanonical imprinting in extraembryonic cells. Sci Adv 5, eaay7246.

Chow, J. and Heard, E. (2009). X inactivation and the complexities of silencing a sex chromosome. Curr. Opin. Cell Biol. 21, 359-366.

Dahlet, T., Argueso Lleida, A., Al Adhami, H., Dumas, M., Bender, A., Ngondo, R. P., Tanguy, M., Vallet, J., Auclair, G., Bardet, A. F., et al. (2020). Genome-wide analysis in the mouse embryo reveals the importance of DNA methylation for transcription integrity. Nat Commun 11, 3153.

Fukuda, A., Tomikawa, J., Miura, T., Hata, K., Nakabayashi, K., Eggan, K., Akutsu, H. and Umezawa, A. (2014). The role of maternal-specific H3K9me3 modification in establishing imprinted X-chromosome inactivation and embryogenesis in mice. Nat Commun 5, 5464.

Galupa, R. and Heard, E. (2015). X-chromosome inactivation: new insights into cis and trans regulation. Curr. Opin. Genet. Dev. 31, 57-66.

Gontan, C., Mira-Bontenbal, H., Magaraki, A., Dupont, C., Barakat, T. S., Rentmeester, E., Demmers, J. and Gribnau, J. (2018). REX1 is the critical target of RNF12 in imprinted X chromosome inactivation in mice. Nat Commun 9, 4752.

Guo, F., Li, L., Li, J., Wu, X., Hu, B., Zhu, P., Wen, L. and Tang, F. (2017). Single-cell multi-omics sequencing of mouse early embryos and embryonic stem cells. Cell Res. 27, 967-988.

Heard, E., Chaumeil, J., Masui, O. and Okamoto, I. (2004). Mammalian X-chromosome inactivation: an epigenetics paradigm. Cold Spring Harb. Symp. Quant. Biol. 69, 89-102.

Huynh, K. D. and Lee, J. T. (2005). X-chromosome inactivation: a hypothesis linking ontogeny and phylogeny. Nat. Rev. Genet. 6, 410-418.

Inoue, K., Kohda, T., Sugimoto, M., Sado, T., Ogonuki, N., Matoba, S., Shiura, H., Ikeda, R., Mochida, K., Fujii, T., et al. (2010). Impeding Xist expression from the active X chromosome improves mouse somatic cell nuclear transfer. Science 330, 496-499.

Liu, X. Y., Wang, C. F., Liu, W. Q., Li, J. Y., Li, C., Kou, X. C., Chen, J. Y., Zhao, Y. H., Gao, H. B., Wang, H., et al. (2016). Distinct features of H3K4me3 and H3K27me3 chromatin domains in pre-implantation embryos. Nature 537, 558-562.

Maslov, A. Y., Lee, M., Gundry, M., Gravina, S., Strogonova, N., Tazearslan, C., Bendebury, A., Suh, Y. and Vijg, J. (2012). 5-aza-2'-deoxycytidine-induced genome rearrangements are mediated by DNMT1. Oncogene 31, 5172-5179.

Oka, M., Meacham, A. M., Hamazaki, T., Rodic, N., Chang, L. J. and Terada, N. (2005). De novo DNA methyltransferases Dnmt3a and Dnmt3b primarily mediate the cytotoxic effect of 5-aza-2'-deoxycytidine. Oncogene 24, 3091-3099.

Okano, M., Bell, D. W., Haber, D. A. and Li, E. (1999). DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99, 247-257.

Schulz, E. G. and Heard, E. (2013). Role and control of X chromosome dosage in mammalian development. Curr. Opin. Genet. Dev. 23, 109-115.

Smith, Z. D., Chan, M. M., Mikkelsen, T. S., Gu, H. C., Gnirke, A., Regev, A. and Meissner, A. (2012). A unique regulatory phase of DNA methylation in the early mammalian embryo. Nature 484, 339-344.

Tan, K., An, L., Miao, K., Ren, L., Hou, Z., Tao, L., Zhang, Z., Wang, X., Xia, W., Liu, J., et al. (2016). Impaired imprinted X chromosome inactivation is responsible for the skewed sex ratio following in vitro fertilization. Proc. Natl. Acad. Sci. U. S. A. 113, 3197-3202.

Reviewer #1 (Recommendations For The Authors):

Title

It would be hard to understand what "co"-regulates means. Does this mean DNA methylation and H3K27me3 co-regulate imprinted X- X-chromosome inactivation? If so, the title can be reworded.

Thanks for your insightful comments, the title has been corrected into “A wave of minor de novo DNA methylation initiates in mouse 8-cell embryos and co-regulates imprinted X- chromosome inactivation with H3K27me3” (Line 2 in the revised manuscript).

Text

(1) As DNA methylation analysis is a primary part of this study, how they processed DNA methylation data can be added to the "Bioinformatics analysis" in the MATERIALS AND METHODS section.

Thanks for your kind reminder. We have added relevant information in the Materials and methods section in the revised manuscript (Lines 462-474 in the revised manuscript).

(2) It seems that recent literature has not been cited in the manuscript. Specifically, none of the papers after 2018 were cited. Recent relevant papers should also be cited throughout the manuscript.

Thanks so much for your reminder. We have added more recent literature to update the relevant information, such as the evidence supporting the causal role between DNA methylation and XCI (Lines 225-228, 264-265 in the revised manuscript); the concurrent enrichment of DNA methylation and H3K27me3 in genes subject to XCI (Lines 301-303 in the revised manuscript); the dominant role of de novo methylation in X chromosome (Lines 253-256 in the revised manuscript), etc.

(3) Line 56: The first report that describes the dynamics of DNMT3B expression in pre-implantation embryonic development (Hirasawa et al., 2007) is missing. This paper should be cited.

Sorry for our carelessness, we have added relevant references and rewritten the sentence in the revised manuscript (Lines 56-57 in the revised manuscript). I think you meant the report by Hirasawa et al in 2008, in which presented expression and subcellular localization of Dnmt3a and Dnmt3b in mouse oocytes and preimplantation embryos.

(4) Line 98: It would be good to mention that the data were derived from reduced representation bisulfite sequencing as the authors used whole-genome bisulfite sequencing data from the same research group as well.

Thanks for your kind reminder. As you have suggested, we have added the description in the revised manuscript to emphasize that these data were derived from reduced representation bisulfite sequencing, while another data were derived from whole-genome bisulfite sequencing, respectively. (Lines 98-99, 111 in the revised manuscript).

(5) Line 101: We first... "the preferential target of DNMT3B (Auclair et al., 2014; Borgel et al., 2010)". More recent literature (Baubec et al., 2016, Duymich et al., 2016, for example) showed that the preferential target of DNMT3B is not a promoter but a gene body. This sentence should be reworded.

Thanks so much for your detailed reminder. As you have pointed out, “preferential target” seems to be an inaccurate statement. Besides of promoters, gene bodies and other elements also undergo de novo DNA methylation (Auclair et al., 2014; Dahlet et al., 2020; Duymich et al., 2016).

We have rewritten the sentence as follows in the revised manuscript: “Promoter regions are important target sites of DNMT3B (Choi et al., 2011). The acquisition of DNA methylation in promoters, especially in intermediate and low CpG promoters, during implantation is largely dependent on DNMT3B and plays an important role in regulating developmental genes (Auclair et al., 2014; Borgel et al., 2010; Dahlet et al., 2020). Thus, among genomic regions that may undergo de novo DNA methylation, we initially focused our analysis on DNA methylation dynamics of promoters...” (Lines 100-106 in the revised manuscript)

(6) Lines 108-109: It would be good to mention that these data were derived from whole-genome bisulfite sequencing.

Thanks for your kind reminder. As aforementioned, we have added a description in the revised manuscript to distinguish between data derived from reduced representation bisulfite sequencing and whole-genome bisulfite sequencing (Lines 98-99, 111 in the revised manuscript).

(7) Line 141: rXCI should be defined.

Thanks for your kind reminder. We have added full descriptions and more necessary information about iXCI and rXCI, to make our statements clearer and easier to be understood (Lines 210-213 in the revised manuscript). In addition, we carefully checked the relevant descriptions throughout the manuscript, and each abbreviation (such as “ICM”) has been defined at its first occurrence. Additionally, we have replaced abbreviations that appears only once in the manuscript with their full terms (Lines 122, 212 in the revised manuscript).

(8) Lines 145-149: The role of DNA methylation for imprinted X-inactivation has already been reported (Chiba et al., 2008). The relevant sentences should be reworded.

Thanks so much for reminding us the important earlier literature that explores the relationship between DNA methylation and XCI. However, the primary aim and hypothesis of the study by Chiba et al. are different from those of our study. Chiba et al focused on whether DNA methylation is the imprinting mark responsible for monoallelic expression of Xist (the initiation event of iXCI), while our study focused on the role of DNA methylation in achieving X chromosomal heterochromatinization (the late event of iXCI).

In detail, the study by Chiba et al. mainly focused on exploring why Xist is specifically expressed from paternal allele and iXCI occurs specifically on the paternal X chromosome in mouse preimplantation embryos. Because Previous studies have suggested that genomic imprinting of Xist is established during oogenesis (Oikawa et al., 2014; Tada et al., 2000), Chiba et al. wanted to test whether the DNA methylation imprinting established during oogenesis is responsible for the monoallelic expression of Xist in preimpantaiton embryos. Analyses of DNA methyltransferase maternal knockout embryos revealed that oocyte DNA methylation is dispensable for Xist imprinting (Chiba et al., 2008). Follow-up study by Inoue et al. identified a broad H3K27me3 enrichment within the Xist 5’region established during oocyte growth and persists through preimplantation development, as the imprinting mark of Xist (Inoue et al., 2017). These series of studies are very important and allows us to understand the mechanism underlying paternal allele-specific iXCI in mouse preimplantation embryos and extraembryonic tissues.

However, the hypothesis is different in our study. Based on the finding of minor de novo DNA methylation and its preferential distribution on the X chromosome, we have speculated that the minor de novo methylation, which occurs from the 8-cell to blastocyst stage, may participate in achieving X chromosomal heterochromatinization. Although DNA methylation is essential for maintaining X chromosome-wide transcriptional silence of rXCI, its role in iXCI remains controversial and it is even plausibly thought that DNA methylation is not required for achieving iXCI because preimplantation embryos undergo global and massive DNA demethylation.

We have reorganized this paragraph, relevant statements have been added to make the background and discussion clearer and easier to be understood. (Lines 217-234 in the revised manuscript)

(9) Lines 164-165: Information regarding Dnmt3b KO is missing. Did the authors generate an original KO line or use an already published one? It should be explicitly stated.

Thank you so much for your kind reminder. The Dnmt3b heterozygous mice were obtained from the Mutant Mouse Resource & Research Centers (MMRRC), and Dnmt3b knockout (KO) embryos were generated by mating Dnmt3b heterozygous females with heterozygous males. The genotyping of Dnmt3b KO embryos was performed by PCR following the MMRRC genotyping protocol (https://www.med.unc.edu/mmrrc/genotyping-protocols/mmrrc-center-protocol-29886/). The relevant information has been added to the MATERIALS AND METHODS section in the revised manuscript (Lines 350-354; 391-393 in the revised manuscript).

(10) Line 165: chemical-induced inhibition of DNMT3B. As 5-aza-dC also blocks DNMT3A and DNMT1, this sentence should be reworded.

Thank you for your valuable comments. 5-aza-dC is a well-established global DNA hypomethylating agent that efficiently inhibit the activity of all DNMTs, and has been frequently used to study the maintenance of DNA methylation and de novo DNA methylation (Maslov et al., 2012; Oka et al., 2005). Thus, despite its inhibitory effect common to various DNMTs, chemical-induced inhibition of DNMTs has the advantage of allowing us to transiently treated embryos specifically during the window of minor de novo DNA methylation (the 8-cell to blastocyst stage). We have rewritten the relevant sentences in the revised manuscript (Lines 183-188 in the revised manuscript).

(11) Lines 171-174: "The role of de novo methylation in iXCI...". This possibility was already tested in the previous study from the Sasaki lab (Chiba et al., 2008).

As mentioned above, the primary aim and hypothesis of the study by Chiba et al. are different from those of our study. Chiba et al. mainly focused on exploring why Xist is specifically expressed from paternal allele and iXCI occurs specifically on the paternal X chromosome in mouse preimplantation embryos, so they tested whether the DNA methylation imprinting established during oogenesis is responsible for this monoallelic expression of Xist in preimplantation embryos (the initiation event of iXCI).

By contrast, based on the finding of minor de novo DNA methylation and its preferential distribution on X chromosome, our study has speculated that the minor de novo DNA methylation, which occurs from the 8-cell to blastocyst stage, may participate in achieving X chromosomal heterochromatinization (the late event of iXCI).

Thanks so much for reminding us this important literature, to make our discussion more informative. We have reorganized this paragraph by rewriting or adding relevant statements to make the background and discussion clearer and easier to be understood (Lines 217-231 in the revised manuscript). In addition, to avoid repeated statement and make our discussion more concise, we have removed the similar sentences at the end of this paragraph.

(12) Lines 198-200: "Given DNA methylation...". These citations mention a general relationship between DNA methylation and H3K27me3 in cells in culture. As I believe the authors focus on X-chromosome inactivation in the female embryos, more relevant papers that discuss the order of the events for the establishment of H3K27me3 and DNA methylation in the inactive X-chromosome can be cited.

Thanks so much for your comment to improve our discussion. It has been thought that during the late phase of rXCI in fully differentiated cells, gene silencing is achieved by PRC2 complex-induced H3K27me3, and then is further stably maintained by the redundant action of multiple layers of epigenetic modifications, including DNA methylation, to reach the maximum level of chromatin compaction (Chow and Heard, 2009; Heard et al., 2004; Pintacuda and Cerase, 2015). In line with this, a recent multifaceted analysis showed that DNA methylation and H3K27me3 are concurrently enriched in genes subject to XCI (Balaton and Brown, 2021). We have added these statements in the revised manuscript (Lines 295-303 in the revised manuscript).

(13) Line 241: As 5-aza-dC blocks both de novo and maintenance DNA methylation, this sentence should be reworded.

Thank you for your kind reminder. As you have mentioned above, 5-aza-dC is a well-established global DNA hypomethylating agent that efficiently inhibit the activity of all DNMTs, and has been frequently used to study the maintenance of DNA methylation and de novo DNA methylation (Maslov et al., 2012; Oka et al., 2005). Thus, despite its inhibitory effect common to various DNMTs, chemical-induced inhibition of DNMTs has the advantage of allowing us to transiently treated embryos specifically during the window of minor de novo DNA methylation (the 8-cell to blastocyst stage). We have rewritten the relevant sentences in the revised manuscript (Lines 183-188 in the revised manuscript).

Figures

(1) Figure 1C, D: Do the rows in C and D show the corresponding genes?

Figure 1C and D represent the DNA methylation changes of promoters (C) and gene bodies (D) respectively, during the transition from the 8-cell to blastocyst stage. Two data were analyzed independently, and rows did not show the corresponding genes. Since we have focused on the minor de novo methylation in promoter regions, to avoid confusion, the results of the gene body have been removed from the revised manuscript.

(2) Figure 1G: Yy2 promoter gained DNA methylation during the transition from 8-cell to the blastocyst stage. Is this a representative locus for the de novo methylated promoters that are shown in Figure 1F where an increase of DNA methylation is about ~1% on average? Another representative locus could be shown instead of this gene promoter.

Thanks so much for you detailed reminder. The inconsistency between the global methylation change and bisulfite sequencing analysis of Yy2, may be due to the details of methodologies, such C-T conversion efficiency, the number of picked colonies, etc. Since we have confirmed the presence of minor de novo DNA methylation using different publicly available data, to avoid ambiguity, we have removed this result in revised manuscript.

(3) Figures 2C and 3A: It would be helpful to mention what the arrowheads mean.

Thanks so much for you detailed reminder. In Figure 2C, the arrowhead indicates the H3k27me3 domain and the blank arrowhead indicates the blastomere without the H3k27me3 domain. In Figure 3A, the arrowhead indicates Xist RNA domain and the blank arrowhead indicates the blastomere without Xist RNA domain. We have added the information in the revised manuscript (Lines 736-738, 747-749 in the revised manuscript).

(4) Figure 3-figure supplement 2B: It would be hard to see whether H3K27me3 is enriched at the promoter regions of presented genes. It would be helpful to show the values for the Y-axis as in panel A.

Thanks for your helpful reminder. We have added the scales to the figure to improve the result presentation (Figure 4—figure supplement 2B in the revised manuscript).

(5) Figure 4-figure supplement 2: 5-aza-dC blocks not only the activity of DNMT3B but also DNMT1, and DNMT3A (all these DNMTs are expressed during pre-implantation embryos, see Hirasawa et al., 2007). This part can be omitted from the manuscript.

Thanks for your insightful comments. As you have mentioned above, the relevance and rationale for experiments using 5-aza-dC treatment should be clarified. 5-aza-dC is a well-established global DNA hypomethylating agent that efficiently inhibit the activity of all DNMTs, and thus has been frequently used to study the maintenance of DNA methylation and de novo DNA methylation (Maslov et al., 2012; Oka et al., 2005).

In our study, to validate the function of minor de novo DNA methylation in iXCI and blastocyst development, we take advantage of 5-aza-dC-induced DNMT inhibition, which allows us to transiently treated embryos specifically during the window of minor de novo DNA methylation (the 8-cell to blastocyst stage), despite its non-specificity to various DNMTs.

Based on these considerations, we hope to retain this result, and wish to get your understanding.

We have added these statements in the revised manuscript to make our experiments more rational and easier to be understood (Lines 183-188 in the revised manuscript) and added a schematic diagram depicting the experimental design (Figure 3—figure supplement 3E in the revised manuscript).

Reviewer #2 (Recommendations For The Authors):

Recommendations/concerns in the text:

- Line 106, it is unclear what is meant by "in line with this"? Gene body DNA methylation is a characteristic of active transcription, so why would a gain in DNA methylation at promoters be in line with a gain in DNA methylation over gene bodies?

Thank you so much for your comments that pointed out our ambiguous statement. We meant both the promoter and gene body regions, albeit accounting for small proportions, gain DNA methylation during the transition from the 8-cell to blastocyst stage. Based on the comment by Reviewer#1, since we have focused on the minor de novo methylation in promoter regions, to avoid confusion, the results of the gene body have been removed from the revised manuscript.

- Line 111 & 114, can 6% DNA methylation really be considered "relatively hypermethylated" compared to 3% DNA methylation that is referred to as "more hypomethylated"?

We apologize for our unclear and ambiguous statements. Here we focused on the promoter regions. Many previous studies have revealed that compared with gene bodies and other genome elements, promoter and overlapping CGI regions, especially high CpG promoters, always showed low levels of DNA methylation. We have added relevant statements to clarify this information, and rewritten the sentences in the revised manuscript (Lines 100-106, 116-118, 121, 124 in the revised manuscript).

- Line 124, there are a number of processes identified, why only mention one in the text? Suggest changing writing to be more accurate, indicating what was included for the GO analysis and using the words "enriched for ... processes". Saying it may be linked to a process is an overstatement and not supported by further experiments/data.

Thank you so much for your detailed comments that make our results more informative. We have checked the relevant description and addressed your suggestions as follows: By performing gene ontology enrichment analysis of genes that undergo minor or major de novo DNA methylation respectively, we noticed that besides of many important basic processes common to two waves of de novo DNA methylation, genes subject to minor de novo DNA methylation were enriched in processes such as organic substance transport, chromosome organization, and cell fate specification (Lines 129-134 in the revised manuscript).

- Lines 149 - 152: sentence/message unclear.

We apologize for the ambiguous description. We have corrected the relevant descriptions as follows: To identify the biological function of minor de novo DNA methylation in iXCI, we knocked down Dnmt3b in preimplantation embryos by microinjecting Dnmt3b siRNA into zygotes (Lines 234-236 in the revised manuscript).

- Lines 162-164: the data in Figure 2C/D does not support this statement, as it does not show H3K27me3 loss specifically at the inactive X-chromosome.

Thanks so much for your insightful comments. Despite the global enrichment of H3K27me3, the H3K27me3 domain detected by immunostaining is a classic marker for establishment of XCI by achieving X chromosome wide heterochromatinization of transcriptional depression (Chow and Heard, 2009; Heard et al., 2004; Huynh and Lee, 2005). Thus, we have used immunostaining for H3K27me3 domains to evaluate the iXCI establishment in the blastocysts, as previously reported (Fukuda et al., 2014; Gontan et al., 2018; Inoue et al., 2010; Tan et al., 2016). To make our results more convincing, we have added another statistical method to quantify the establishment of iXCI, i.e., the percentage of H3K27me3-positive and -negative trophoblast cells to total trophoblast cells in female blastocysts subject to Dnmt3b knockdown or not.

In addition, we have added a schematic diagram depicting the process of iXCI initiation and establishment, as well as the experimental design and work flows, to make the result easier to be understood.

In addition, we agree with your comments that additional evidence will benefit the conclusion. To strengthen the evidence, and test whether DNA methylation loss leads to a prolonged effect on iXCI, we have reanalyzed the RNA-seq and H3K27me3 CHIP-seq data in extraembryonic ectoderm (ExE) of E6.5 single embryos that underwent Dnm3a/3b knockout because preimplantation iXCI status maintains extraembryonic cells (Chen et al., 2019; Galupa and Heard, 2015; Schulz and Heard, 2013). The results showed that chromosome-wide loss of DNA methylation led to a nearly complete loss of H3k27me3 on paternal (specifically inactivated in iXCI), along with a notable transcriptional upregulation cross the chromosome. By contrast, these changes cannot be not observed on maternal X chromosome. (Lines 253-261; Figure 3—figure supplement 4A in the revised manuscript)

- Lines 169-174: sentence/message unclear.

As aforementioned, we have reorganized this paragraph by rewriting or adding relevant statements relevant to the DNA methylation and XCI, to make the background and discussion clearer and easier to be understood (Lines 217-234 in the revised manuscript). In addition, to avoid repeated statement and make our discussion more concise, we have removed the similar sentences at the end of this paragraph.

- Lines 177-179: this statement is too bold. The data does not support "direct evidence".

Thank you for your detailed reminder. We have rewritten the sentence to avoid confusion and overstatement (Lines 262-268 in the revised manuscript).

- Line 198: these are not all enzymes, but could be referred to as chromatin modifiers.

We apologize for the ambiguous description. As you suggested, we have corrected “enzymes” to “chromatin modifiers” (Lines 284, 287 in the revised manuscript).

- Line 199: this statement is not correct in all contexts. There are many studies showing antagonism between DNA methylation and H3K27me3.

Thanks so much for you careful reviewing. As you have pointed out, the relationship of DNA methylation and H3K27me3 are divergent and largely controversial among studies. Under certain circumstances, DNA methylation shows antagonistic effect to H3K27me3 at promoters, via excluding the binding of PRC2 (the main complex responsible for H3K27me3 deposition) components to their targets (Bartke et al., 2010; Jermann et al., 2014), while other studies have presented alternative evidence that PRC2 (the main complex responsible for H3K27me3 deposition) and DNA methylation cooperate to achieve silencing (Hagarman et al., 2013; Vire et al., 2006). Thus, it has been thought that the relationship between DNA and methylation and histone modifications is complex, possibly in a cell-type and/or genomic region-specific manner. Both antagonism and coordination can be observed in different regulatory elements in mouse ES cells (King et al., 2016).

We apologize our incomplete statement because we mainly focused on their synergistic relationship. We have refined this section by rewriting relevant sentences and adding necessary statements (Lines 288-303 in the revised manuscript).

- Lines 228-230: the developmental significance of DNA methylation homeostasis is already well-established. Please reference relevant papers showing this here.

Thank you for this helpful suggestion. We have reorganized this section. Relevant references that highlight the developmental significance of DNA methylation homeostasis have added. The sentence has been rewritten and moved to the end of this paragraph, in the revised manuscript (Lines 159-161 in the revised manuscript).

- Line 238: an explanation/rationale for looking at energy metabolism is lacking.

Thank you for your comments to make our results earlier to be understood. The detection of energy metabolism is mainly based on the integrated analysis of DNA methylation and gene expression from the 8-cell embryos to ICM, to test the potential short-and long-term developmental consequences of minor de novo DNA methylation. Bioinformatic analysis suggested that many basic processes, such as cell differentiation, cell cycle and metabolic regulation, may be regulated by minor de novo DNA methylation. Among the enriched genes, several are related energy metabolism. In addition, because energy metabolism is crucial for supporting embryo differentiation and development, and oxidative phosphorylation (OXPHOS) metabolism is highly activated during the blastocyst stage (Zhao et al., 2021), we next examined the energy metabolism, particularly OXPHOS activity, of Dnmt3b-KD embryos. We have refined the section by rewritten relevant sentence and added necessary statements (Lines 175-179 in the revised manuscript).

- Lines 246-248: Looking at the data in Figure 2 figure supplement 2, this statement is simply not true with regards to DNMT3B protein, and also global DNA methylation level is reduced in the Dnmt3b KD blastocyst, which could lead to defective major de novo DNA methylation.

Thanks for your careful reviewing, we have rewritten the sentence to make our statement more accurate and avoid overstatement (Lines 188-190 in the revised manuscript).

Recommendations/concerns relating to figures:

Figure 1:

- Of all genic promoters, how many were included in the analysis (contained sufficient coverage)? What cut-off/thresholds were used to consider DNA methylation gain at a promoter?

Thanks for your comments. In total, 11662 promoters were analyzed. Given that promoter methylation is generally at low level, particularly at the 8-cell stage at which minor de novo methylation is just initiated. The relatively lower basal levels make the increase before the blastocyst, seem considerably slight. To capture the slight changes, we have used the relaxed threshold based on ΔDNA methylation. Only CpG sites with at least fivefold coverage were included in the methylation analysis based on data from Smith et al. (Smith et al., 2012)., ΔDNA methylation greater or less than 0 was defined as gain or loss of DNA methylation. We have added this information in the revised manuscript (Lines 462-470 in the revised manuscript).

- Does an average methylation level of 0.02 represent 2% DNA methylation? Presuming yes, is the average 1.5% DNA methylation gain at promoters real? And meaningful? Especially compared to the gain in DNA methylation that takes place between ICM and E6.5 (Figure 1 Figure Supplement 1 D)

As you have pointed out, an average methylation level of 0.02 represent 2% DNA methylation. As aforementioned, promoters exhibited an average of 1.5% DNA methylation gain during the transition from 8-cell stage to ICM. The slight increase may be mainly due to the relatively lower basal levels. As you expected, compared with the comprehensive de novo DNA methylation during implantation, preimplantation de novo methylation occurs more slightly, at a small proportion of promoter regions, so designated it as minor de novo DNA methylation. It should be also mentioned that a proportion of these promoters continue to gain massive DNA methylation during implantation. We have refined the relevant sentences to provide more detailed information of our results (Lines 125-127 in the revised manuscript).

- Why is there a focus on promoters (which are not the preferential target of DNMT3B)?

Thanks so much for your detailed reminder. As you have pointed out, “preferential target” seems to be an inaccurate statement. besides of promoters, gene bodies and other elements also undergo de novo DNA methylation (Auclair et al., 2014; Dahlet et al., 2020; Duymich et al., 2016). We have focused on the promoter regions based on the following considerations: (1) Promoter regions are important target sites of DNMT3B (Choi et al., 2011); (2) The acquisition of DNA methylation in promoters, especially in intermediate and low CpG promoters, during implantation is largely dependent on DNMT3B and plays an important role in regulating developmental genes (Auclair et al., 2014; Borgel et al., 2010; Dahlet et al., 2020). We have rewritten the relevant sentence in the revised manuscript (Lines 100-106 in the revised manuscript).

- Figure 1H shows that promoters that gain DNA methylation during the "minor de novo DNA methylation" continue to gain DNA methylation during "de novo DNA methylation". Is the ~1.5% DNA methylation gain just the slow start of the main de novo DNA methylation wave?

Your comments is very helpful to improve the description of our results. In the present study, our analysis indicated that a small proportion of promoters initially gain methylation during the transition from the 8-cell to ICM. The finding challenges current knowledge: (1) de novo DNA methylation occurs during implantation, by which globally hypomethylated blastocysts acquire genome-wide DNA methylation (Borgel et al., 2010; Dahlet et al., 2020; Smith et al., 2012); (2) during preimplantation development, embryos undergo massive and global DNA demethylation.

To distinguish the current knowledge of the timing and dynamics of DNA methylation during the early development, we have designated our finding during the transition from the 8-cell to blastocyst stage, as minor de novo DNA methylation.

We agree with your notion that among the promoters undergoing minor de novo methylation, most of them continue to gain DNA methylation during implantation, as revealed in Fig. 1F. We have added refine the relevant statement in revised manuscript (Lines 125-127 in the revised manuscript).

- The GO analysis performed for Figure 1H, what was used as input? Promoters of genes that gain DNA methylation as identified in 1C?

Thank you for your comments. For the GO analysis shown in Figure 1H, we used genes with promoter regions that gained or lost DNA methylation during the transition from the 8-cell to ICM respectively (identified in Figure 1C, as input), respectively. This information has been clarified in the revised manuscript to ensure accuracy (Lines 129-134 in the revised manuscript).

- Figure 1 figure supplement 1, is there only a fold change as threshold or also a calculated significance (eg. p-value/FDR)?

Thanks for your valuable comments. Considering the relatively low DNA methylation levels at promoter regions, and the slightly changes occurring during the preimplantation embryo development, we used the relaxed threshold based on ΔDNA methylation. Only CpG sites with at least fivefold coverage were included in the methylation analysis based on data from Smith et al. (Smith et al., 2012), ΔDNA methylation greater or less than 0 was defined as gain or loss of DNA methylation. We have replaced relevant figures and added this information in the revised manuscript (Figure 1—figure supplement 1D-E; Lines 125-127 in the revised manuscript).

- To confirm DNMT3B is responsible for the DNA methylation gain: DNMT3B KD/KO followed by promoter DNA methylation analysis to confirm the promoters that gain DNA methylation between 8 cell and ICM don't gain DNA methylation in the absence of DNMT3B.

We agree with your comments that additional evidence will benefit the conclusion. To strengthen the evidence, we have reanalyzed the RNA-seq and H3K27me3 CHIP-seq data in extraembryonic ectoderm (ExE) of E6.5 single embryos that underwent Dnm3a/3b knockout because preimplantation iXCI status maintains extraembryonic cells (Chen et al., 2019; Galupa and Heard, 2015; Schulz and Heard, 2013). The results showed that chromosome-wide loss of DNA methylation led to a nearly complete loss of H3k27me3 on paternal (specifically inactivated in iXCI), which showed a notable transcriptional upregulation cross the chromosome. By contrast, these changes cannot be not observed on maternal X chromosome. We have added this result in the revised manuscript (Lines 253-261; Figure 3—figure supplement 4A in the revised manuscript).

Figure 2:

- Figure 2A: label missing for what the numbers on the y-axis represent.

Thank you for pointing this out. We apologize for the oversight. We have added the label of y-axis in Figure 2A to clarify what the numbers represent, making it easier to be understood (Figure 3A in the revised manuscript).

- Figure 2B: y-axis is % of methylated promoters compared to all promoters?

Thank you for your suggestion. The y-axis in Figure 2B indeed represents the percentage of de novo methylated promoters relative to all promoters. As you have suggested, we have clarified this labeling in the revised manuscript (Figure 3B in the revised manuscript).

- What is the delta DNA methylation gain specifically for X-linked promoters?

Thanks so much for your reminder. To provide more convincing evidence. We have reanalyzed a single cell COOL-seq data, we also specifically reanalyzed the DNA methylation changes on the X chromosomal promoter in female embryos. The X chromosome showed a more notable increase in the de novo methylated promoters than that on autosomes, and the female X chromosome showed higher DNA methylation levels than that of the male (Figure 3—figure supplement 2A-B; Lines 203-206 in the revised manuscript).

- Figure 2C: include representative images of separate channels to better see the signal of CDX2 and H3K27me3. Quantification would be better represented with box plots.

Thank you for your helpful suggestions. We have added separate channel images in the revised manuscript. Additionally, we have adjusted the quantification to be represented as box plots, as you have suggested, to improve the accuracy and interpretability of the data presentation (Figure 3D-F in the revised manuscript).

- Figure 2C: Does the H3K27me3 signal overlap with the location of the inactive X-chromosome (is there maybe denser DAPI or do IF combined with Xist RNA-FISH)?

Thanks so much for your insightful comments. Despite the global enrichment of H3K27me3, the H3K27me3 domain detected by immunostaining is a classic marker for establishment of XCI by achieving X chromosome wide heterochromatinization of transcriptional depression (Chow and Heard, 2009; Heard et al., 2004; Huynh and Lee, 2005). Thus, we have used immunostaining for H3K27me3 domains to evaluate the iXCI establishment in the blastocysts, as previously reported (Fukuda et al., 2014; Gontan et al., 2018; Inoue et al., 2010; Tan et al., 2016). We have taken effort to perform co-staining of H3K27me3 IF and Xist FISH, but was hindered by the technical challenge, we wish to get your understanding. However, as we aforementioned, H3K27me3 is a well-accepted maker to clarify the XCI status.

In addition, to make our results more convincing, we have added an alternative statistical method to quantify the establishment of iXCI, i.e., the percentage of H3K27me3-positive and -negative trophoblast cells to total trophoblast cells in female blastocysts subject to Dnmt3b knockdown or not (Figure 3F; Lines 243-244 in the revised manuscript)

- Figure 2 figure supplement 2A: relative expression of Dnmt3b?

Thanks for your detailed reminder. The data represent the relative expression level of Dnmt3b, as noted in the original figure legend. Based on your comments, we have added the gene name in the label of the Y-axis. Similarly, the protein name has been also added to make the results more informative (Figure 2 figure supplement 2A, C, E in the revised manuscript).

- Figure 2 figure supplement 2B/C: in the text, line 153, it is stated that "Dnmt3b mRNA and protein levels were significantly reduced in morulae, but not in blastocysts compared to those of negative control (NC) group". These figures do not support that statement. The IF images show a loss of DNMT3B in the Dnmt3b KD blastocysts. The IF quantification seems to have fewer datapoints for the blastocyst, and looking at the bar graphs, there seems to be a trend towards reduced DNMT3B in both the morula and blastocyst, which would also explain the reduction in DNA methylation in both stages as shown in Figure 2 figure supplement 2D/E.

Thanks so much for your careful reviewing that makes our statements more accurate. We have rewritten the sentence in the revised manuscript as follows: Dnmt3b mRNA and protein levels were significantly reduced in morulae, and tended to be lower in blastocysts compared to those of the negative control (NC) group. In addition, we have removed “transient” from the original statement “The transient inhibition of Dnmt3b” (Lines 168-170 in the revised manuscript).

- Figure 2 figure supplement 2F/G: include representative IF images with separation of all channels and the merged image.

Thank you for your suggestion. We have added the representative immunofluorescence (IF) images with separate channels and merged image in the revised manuscript (Figure 3—figure supplement 3B, F in the revised manuscript).

- Figure 2 figure supplement 2H: Instead of showing log2FC in methylation levels, delta methylation would be more informative. Are these genes already inactivated at the 8-cell stage? Or are they active and become inactivated by the gain in DNA methylation? Doing qPCR for these genes, or looking at published RNAseq data would be informative. What happens to the expression of these genes in the Dnmt3b KD?

Thanks for your suggestions. We have represented DNA methylation changes as “ΔDNA methylation”. During mouse preimplantation development, iXCI is initiated in earlier cleavage female embryos dependent on Xist upregulation around 4-8-cell stage, and then Xist specifically coats paternal X chromosome and finally leads to chromosome-wide silencing via heterochromatinization in early blastocysts. Thus, these non-escaping genes, which are subject to XCI, would not be inactivated at 8-cell stage

Author response image 1.

The processes of iXCI initiation and establishment (left panel), and dynamics of total expression levels of X chromosome in male and female preimplantation embryos (right panel, note that X-dosage is balanced between sexes until the early blastocyst stage).

As you expected, most of these representative non-escaping is downregulated upon the transition of 8-cell to blastocyst stage, consistent with their gain of DNA methylation. Additionally, since preimplantation iXCI status maintains extraembryonic cells (Galupa and Heard, 2015; Schulz and Heard, 2013), we further reanalyzed the published RNA-seq data in extraembryonic ectoderm (ExE) of E6.5 single embryos that underwent DNA methyltransferase knockout (Chen et al., 2019). The results showed that chromosome-wide loss of DNA methylation led to a chromosome-wide transcriptional upregulation, including the locus of these non-escaping genes, on paternal X chromosome. We have added this result in the revised manuscript (Figure 3—figure supplement 3J; Figure 3—figure supplement 4A-B; Lines 253-261 in the revised manuscript).

Figure 3:

- Figure 3 figure supplement 1: representative IF image missing.

Thanks for your kind reminder. We have added the representative IF images in the revised manuscript to provide a clearer illustration of the data (Figure 4—figure supplement 1A in the revised manuscript).

- Figure 3 figure supplement 2B: scales are missing for the H3K27me3 ChIP-seq data (are the 8-cell and ICM tracks set to the same scale?). It looks like the ICM track is cut off at the top (peaks not fully displayed) and the data looks very sparse. A more informative analysis would be to do peak calling over promoters and compare 8-cell with ICM.

Thanks for your detailed reminder. We apologize for the missing of scale bars in the H3K27me3 ChIP-seq data. The 8-cell and ICM tracks were set to the same scale, and we have now added scales to the figure in the revised manuscript to improve the result presentation. As you have speculated, the visual effect of the flatted peak is not caused by track cutting off, but rather by zooming into a specific region in the extended IGV files.

These results are based on the reanalysis of publicly available data of pooled embryos, which just provided suggestive but not direct evidence to support the role of DNA methylation in promoting X-linked H3K27me3 enrichment in iXCI.

To provide more convincing evidence. we have reanalyzed the RNA-seq and H3K27me3 CHIP-seq data in extraembryonic ectoderm (ExE) of E6.5 female embryos that underwent Dnmt3a/3b knockout because preimplantation iXCI status maintains extraembryonic cells (Chen et al., 2019; Galupa and Heard, 2015; Schulz and Heard, 2013). The results showed that Dnmt knockout led to a nearly complete loss of H3k27me3 on paternal (specifically inactivated in iXCI), which showed a notable transcriptional upregulation cross the chromosome. By contrast, these changes cannot be not observed on maternal X chromosome (Figure 3—figure supplement 4 in the revised manuscript). We have added these results in the revised manuscript.

- Figure 3E: Given all tested proteins give a positive signal, it would have been good to include a negative control chromatin protein that is known to not interact with DNMT3B. Given both PRC2 and DNMT3B are chromatin-binding proteins, can the signal be a result of close proximity instead of a direct interaction?

In the present study, to test the interaction between DNMT3B and PRC2 core components, we have used in situ proximity ligation assay (PLA), an increasingly popular technique for detecting the close proximity of two proteins in fixed samples using two primary antibodies (Alsemarz et al., 2018).

Author response image 2.

Schematic diagram of the principle of the in situ PLA.

Compared with classical co-Immunoprecipitation (Co-IP) method, in situ PLA has advantages in (1) detecting low input samples or proteins expressed at low levels, which is extremely difficult using Co-IP; (2) providing in situ or subcellular information of protein-protein interaction. However, it should be noted that the maximal distance allowing this reaction is 40 nm, which is not quite small enough to demonstrate a physical interaction between the two antigens, but sufficient to support a very close “proximity”.

In our study, in situ PLA, including the experimental design of negative control, was performed in the accordance with the manufacturer’s instruction of Duolink® In Situ Red Starter Kit (MilliporeSigma): “Technical negative controls included incubation with each primary antibody separately and no primary antibody”. We have refined the relevant sentence in the revised manuscript (Lines 308-310 in the revised manuscript)

- Figure 3G: It would have been good to include a negative control, and DNase/benzonase to exclude DNA/RNA-mediated protein interaction.

- (Of note, there have been previous studies reporting an interaction between PRC2 and DNMT3B in other cell types, such as in Weigert et al. 2023, but unfortunately, they don't seem to use DNase/benzonase either).

The Co-IP analysis of DNMT3B and PRC2 core components in differentiated female ES cells was presented as additional supportive evidence. Because the Co-IP analysis is extremely difficult for preimplantation embryos, we have used in situ PLA to detect their interaction. However, the maximal distance allowing in situ PLA reaction is 40 nm, which is not quite small enough to demonstrate a physical interaction (Alsemarz et al., 2018). Thus, we have added a Co-IP analysis using differentiated female ES cells, in which rXCI occurs upon the differentiation.

Based on this consideration of the importance and contribution of this result, we have moved this result from the main figure, to the supplemental figure (Figure 4—figure supplement 3H in the revised manuscript).

- Figure 3 figure supplement 3G: what were the ESCs differentiated into? Did the Dnmt3b KO or Dnmt3a/b DKO show any differentiation defect?

The mouse ESC line PGK12.1 was a well-established ex vivo model of rXCI. Under the standard culture condition, PGK12.1 is normally fated to neuroectodermal commitment.

Author response image 3.

Immunostaining of NESTIN, a neuroectodermal stem cell marker molecule, and NANOG in undifferentiated and differentiated PGK12.1 ESCs respectively.

No differentiation defects have been observed in either Dnmt3b KO or Dnmt3a/3b DKO ESCs in our study. Dnmt KO/DKO/TKO ES cell lines have been successfully used as the model of interaction of DNA methylation and H3K27me3 deposition (King et al., 2016).

Figure 4:

- Figure 4B: Is there an explanation for seeing similar total cell numbers in Figure 4B, but showing decreased proliferation in Figure 4A?

Thank you for your insightful comments. The EdU cell proliferation assays labels cells during the S phase of cell cycle, as the 5-ethynyl 2´-deoxyuridine (EdU) is incorporated into newly synthesized DNA. This labeling identifies cells undergoing DNA synthesis, but these cells may not have completed mitosis at the time of detection. As a result, the total cell number may not immediately reflect the decrease in proliferation observed in the treated group. To address this point, we have rewritten the sentences in the revised manuscript (Lines 174-175 in the revised manuscript).

References

Alsemarz, A., Lasko, P. and Fagotto, F. J. B. (2018). Limited significance of the in situ proximity ligation assay. bioRxiv, 411355.

Auclair, G., Guibert, S., Bender, A. and Weber, M. (2014). Ontogeny of CpG island methylation and specificity of DNMT3 methyltransferases during embryonic development in the mouse. Genome Biol. 15, 545.

Balaton, B. P. and Brown, C. J. (2021). Contribution of genetic and epigenetic changes to escape from X-chromosome inactivation. Epigenetics Chromatin 14, 30.

Bartke, T., Vermeulen, M., Xhemalce, B., Robson, S. C., Mann, M. and Kouzarides, T. (2010). Nucleosome-interacting proteins regulated by DNA and histone methylation. Cell 143, 470-484.

Borgel, J., Guibert, S., Li, Y., Chiba, H., Schubeler, D., Sasaki, H., Forne, T. and Weber, M. (2010). Targets and dynamics of promoter DNA methylation during early mouse development. Nat. Genet. 42, 1093-1100.

Chen, Z., Yin, Q., Inoue, A., Zhang, C. and Zhang, Y. (2019). Allelic H3K27me3 to allelic DNA methylation switch maintains noncanonical imprinting in extraembryonic cells. Sci Adv 5, eaay7246.

Chiba, H., Hirasawa, R., Kaneda, M., Amakawa, Y., Li, E., Sado, T. and Sasaki, H. (2008). De novo DNA methylation independent establishment of maternal imprint on X chromosome in mouse oocytes. Genesis 46, 768-774.

Choi, S. H., Heo, K., Byun, H. M., An, W., Lu, W. and Yang, A. S. (2011). Identification of preferential target sites for human DNA methyltransferases. Nucleic Acids Res. 39, 104-118.

Chow, J. and Heard, E. (2009). X inactivation and the complexities of silencing a sex chromosome. Curr. Opin. Cell Biol. 21, 359-366.

Dahlet, T., Argueso Lleida, A., Al Adhami, H., Dumas, M., Bender, A., Ngondo, R. P., Tanguy, M., Vallet, J., Auclair, G., Bardet, A. F., et al. (2020). Genome-wide analysis in the mouse embryo reveals the importance of DNA methylation for transcription integrity. Nat Commun 11, 3153.

Duymich, C. E., Charlet, J., Yang, X. J., Jones, P. A. and Liang, G. N. (2016). DNMT3B isoforms without catalytic activity stimulate gene body methylation as accessory proteins in somatic cells. Nat Commun 7, 11453.

Fukuda, A., Tomikawa, J., Miura, T., Hata, K., Nakabayashi, K., Eggan, K., Akutsu, H. and Umezawa, A. (2014). The role of maternal-specific H3K9me3 modification in establishing imprinted X-chromosome inactivation and embryogenesis in mice. Nat Commun 5, 5464.

Galupa, R. and Heard, E. (2015). X-chromosome inactivation: new insights into cis and trans regulation. Curr. Opin. Genet. Dev. 31, 57-66.

Gontan, C., Mira-Bontenbal, H., Magaraki, A., Dupont, C., Barakat, T. S., Rentmeester, E., Demmers, J. and Gribnau, J. (2018). REX1 is the critical target of RNF12 in imprinted X chromosome inactivation in mice. Nat Commun 9, 4752.

Hagarman, J. A., Motley, M. P., Kristjansdottir, K. and Soloway, P. D. (2013). Coordinate regulation of DNA methylation and H3K27me3 in mouse embryonic stem cells. PLoS One 8, e53880.

Heard, E., Chaumeil, J., Masui, O. and Okamoto, I. (2004). Mammalian X-chromosome inactivation: an epigenetics paradigm. Cold Spring Harb. Symp. Quant. Biol. 69, 89-102.

Huynh, K. D. and Lee, J. T. (2005). X-chromosome inactivation: a hypothesis linking ontogeny and phylogeny. Nat. Rev. Genet. 6, 410-418.

Inoue, A., Jiang, L., Lu, F. and Zhang, Y. (2017). Genomic imprinting of Xist by maternal H3K27me3. Genes Dev. 31, 1927-1932.

Inoue, K., Kohda, T., Sugimoto, M., Sado, T., Ogonuki, N., Matoba, S., Shiura, H., Ikeda, R., Mochida, K., Fujii, T., et al. (2010). Impeding Xist expression from the active X chromosome improves mouse somatic cell nuclear transfer. Science 330, 496-499.

Jermann, P., Hoerner, L., Burger, L. and Schubeler, D. (2014). Short sequences can efficiently recruit histone H3 lysine 27 trimethylation in the absence of enhancer activity and DNA methylation. Proc. Natl. Acad. Sci. U. S. A. 111, E3415-3421.

King, A. D., Huang, K., Rubbi, L., Liu, S., Wang, C. Y., Wang, Y., Pellegrini, M. and Fan, G. (2016). Reversible Regulation of Promoter and Enhancer Histone Landscape by DNA Methylation in Mouse Embryonic Stem Cells. Cell Rep. 17, 289-302.

Maslov, A. Y., Lee, M., Gundry, M., Gravina, S., Strogonova, N., Tazearslan, C., Bendebury, A., Suh, Y. and Vijg, J. (2012). 5-aza-2'-deoxycytidine-induced genome rearrangements are mediated by DNMT1. Oncogene 31, 5172-5179.

Oikawa, M., Inoue, K., Shiura, H., Matoba, S., Kamimura, S., Hirose, M., Mekada, K., Yoshiki, A., Tanaka, S., Abe, K., et al. (2014). Understanding the X chromosome inactivation cycle in mice: a comprehensive view provided by nuclear transfer. Epigenetics-Us 9, 204-211.

Oka, M., Meacham, A. M., Hamazaki, T., Rodic, N., Chang, L. J. and Terada, N. (2005). De novo DNA methyltransferases Dnmt3a and Dnmt3b primarily mediate the cytotoxic effect of 5-aza-2'-deoxycytidine. Oncogene 24, 3091-3099.

Pintacuda, G. and Cerase, A. (2015). X Inactivation Lessons from Differentiating Mouse Embryonic Stem Cells. Stem Cell Rev Rep 11, 699-705.

Schulz, E. G. and Heard, E. (2013). Role and control of X chromosome dosage in mammalian development. Curr. Opin. Genet. Dev. 23, 109-115.

Smith, Z. D., Chan, M. M., Mikkelsen, T. S., Gu, H. C., Gnirke, A., Regev, A. and Meissner, A. (2012). A unique regulatory phase of DNA methylation in the early mammalian embryo. Nature 484, 339-344.

Tada, T., Obata, Y., Tada, M., Goto, Y., Nakatsuji, N., Tan, S., Kono, T. and Takagi, N. (2000). Imprint switching for non-random X-chromosome inactivation during mouse oocyte growth. Development 127, 3101-3105.

Tan, K., An, L., Miao, K., Ren, L., Hou, Z., Tao, L., Zhang, Z., Wang, X., Xia, W., Liu, J., et al. (2016). Impaired imprinted X chromosome inactivation is responsible for the skewed sex ratio following in vitro fertilization. Proc. Natl. Acad. Sci. U. S. A. 113, 3197-3202.

Vire, E., Brenner, C., Deplus, R., Blanchon, L., Fraga, M., Didelot, C., Morey, L., Van Eynde, A., Bernard, D., Vanderwinden, J. M., et al. (2006). The Polycomb group protein EZH2 directly controls DNA methylation. Nature 439, 871-874.

Zhao, J., Yao, K., Yu, H., Zhang, L., Xu, Y., Chen, L., Sun, Z., Zhu, Y., Zhang, C., Qian, Y., et al. (2021). Metabolic remodelling during early mouse embryo development. Nat Metab 3, 1372-1384.

Author response:

The following is the authors’ response to the original reviews

Reviewer #1:

Despite the strengths, multiple analytical decisions have to be explained, justified, or clarified. Also, there is scope to enhance the clarity and coherence of the writing - as it stands, readers will have to go back and forth to search for information. Last, it would be helpful to add line numbers in the manuscript during the revision, as this will help all reviewers to locate the parts we are talking about.

We thank the reviewer’s suggestions have added the line numbers to the revised manuscript.

(1) Introduction:

The introduction is somewhat unmotivated, with key terms/concepts left unexplained until relatively late in the manuscript. One of the main focuses in this work is "hyperaltruistic", but how is this defined? It seems that the authors take the meaning of "willing to pay more to reduce other's pain than their own pain", but is this what the task is measuring? Did participants ever need to PAY something to reduce the other's pain? Note that some previous studies indeed allow participants to pay something to reduce other's pain. And what makes it "HYPER-altruistic" rather than simply "altruistic"?

As the reviewer noted, we adopted a well-established experimental paradigm to study the context-dependent effect on hyper-altruism. Altruism refers to the fact that people take others’ welfare into account when making decisions that concern both parties. Research paradigms investigating altruistic behavior typically use a social decision task that requires participants to choose between options where their own financial interests are pitted against the welfare of others (FeldmanHall et al., 2015; Hu et al., 2021; Hutcherson et al., 2015; Teoh et al., 2020; Xiong et al., 2020). On the other hand, the hyperaltruistic tendency emphasizes subjects’ higher valuation to other’s pain than their own pain (Crockett et al., 2014, 2015, 2017; Volz et al., 2017). One example for the manifestation of hyperaltruism would be the following scenario: the subject is willing to forgo $2 to reduce others’ pain by 1 unit (social-decision task) and only willing to forgo $1 to reduce the same amount of his/her own pain (self-decision task) (Crockett et al., 2014). On the contrary, if the subjects are willing to forgo less money to reduce others’ suffering in the social decision task than in the self-decision task, then it can be claimed that no hyperaltruism is observed. Therefore, hyperaltruistic preference can only be measured by collecting subjects’ choices in both the self and social decision tasks and comparing the choices in both tasks.

In our task, as in the studies before ours (Crockett et al., 2014, 2015, 2017; Volz et al., 2017), subjects in each trial were faced with two options with different levels of pain on others and monetary payoffs on themselves. Based on subjects’ choice data, we can infer how much subjects were willing to trade 1 unit of monetary payoff in exchange of reducing others’ pain through the regression analysis (see Figure 1 and methods for the experimental details). We have rewritten the introduction and methods sections to make this point clearer to the audience.  

Plus, in the intro, the authors mentioned that the "boundary conditions" remain unexplored, but this idea is never touched again. What do boundary conditions mean here in this task? How do the results/data help with finding out the boundary conditions? Can this be discussed within wider literature in the Discussion section?

Boundary conditions here specifically refer to the variables or decision contexts that determine whether hyperaltruistic behavior can be elicited. Individual personality trait, motivation and social relationship may all be boundary conditions affecting the emergence of hyperaltruistic behavior. In our task, we specifically focused on the valence of the decision context (gain vs. loss) since previous studies only tested the hyperaltruistic preference in the gain context and the introduction of the loss context might bias subjects’ hyperaltruistic behavior through implicit moral framing.

We have explained the boundary conditions in the revised introduction (Lines 45 ~ 49).

“However, moral norm is also context dependent: vandalism is clearly against social and moral norms yet vandalism for self-defense is more likely to be ethically and legally justified (the Doctrine of necessity). Therefore, a crucial step is to understand the boundary conditions for hyperaltruism.”

Last, what motivated the authors to examine the decision context? It comes somewhat out of the blue that the opening paragraph states that "We set out to [...] decision context", but why? Are there other important factors? Why decision context is more important than studying those others?

We thank the reviewer for the comment. The hyperaltruistic preference was originally demonstrated between conditions where subjects’ personal monetary gain was pitted against others’ pain (social-condition) or against subjects’ own suffering (self-condition) (Crockett et al., 2014). Follow up studies found that subjects also exhibited strong egoistic tendencies if instead subjects needed to harm themselves for other’s benefit in the social condition (by flipping the recipients of monetary gain and electric shocks) (Volz et al., 2017). However, these studies have primarily focused on the gain contexts, neglecting the fact that valence could also be an influential factor in biasing subjects’ behavior (difference between gain and loss processing in humans). It is likely that replacing monetary gains with losses in the money-pain trade-off task might bias subjects’ hyperaltruistic preference due to heightened vigilance or negative emotions in the face of potential loss (such as loss aversion) (Kahneman & Tversky, 1979; Liu et al., 2020; Pachur et al., 2018; Tom et al., 2007; Usher & McClelland, 2004; Yechiam & Hochman, 2013). Another possibility is that gain and loss contexts may elicit different subjective moral perceptions (or internal moral framings) in participants, affecting their hyperaltruistic preferences (Liu et al., 2017; Losecaat Vermeer et al., 2020; Markiewicz & Czupryna, 2018; Wu et al., 2018). In our manuscript, we did not strive to compare which factors might be more important in eliciting hyperaltruistic behavior, but rather to demonstrate the crucial role played by the decision context and to show that the internal moral framing could be the mediating factor in driving subjects’ hyperaltruistic behavior. In fact, we speculate that the egoistic tendencies found in the Volz et al. 2017 study was partly driven by the subjects’ failure to engage the proper internal moral framing in the social condition (harm for self, see Volz et al., 2017 for details).

(2) Experimental Design:

(2a) The experiment per se is largely solid, as it followed a previously well-established protocol. But I am curious about how the participants got instructed? Did the experimenter ever mention the word "help" or "harm" to the participants? It would be helpful to include the exact instructions in the SI.

In the instructions, we avoided words such as “harm”, “help”, or other terms reminding subjects about the moral judgement of the decisions they were about to make. Instead, we presented the options in a neutral and descriptive manner, focusing only on the relevant components (shocks and money). The instructions for all four conditions are shown in supplementary Fig. 9.

(2b) Relatedly, the experimental details were not quite comprehensive in the main text. Indeed, the Methods come after the main text, but to be able to guide readers to understand what was going on, it would be very helpful if the authors could include some necessary experimental details at the beginning of the Results section.

We thank the reviewer’s suggestion. We have now provided a brief introduction of the experimental details in the revised results section (Lines 125 ~132).

“Prior to the money-pain trade-off task, we individually calibrated each subject’s pain threshold using a standard procedure[4–6]. This allowed us to tailor a moderate electric stimulus that corresponded to each subject’s subjective pain intensity. Subjects then engaged in 240 decision trials (60 trials per condition), acting as the “decider” and trading off between monetary gains or losses for themselves and the pain experienced by either themselves or an anonymous “pain receiver” (gain-self, gain-other, loss-self and loss-other, see Supplementary Fig. 8 for the instructions and also see methods for details).”

(3) Statistical Analysis<br /> (3a) One of the main analyses uses the harm aversion model (Eq1) and the results section keeps referring to one of the key parameters of it (ie, k). However, it is difficult to understand the text without going to the Methods section below. Hence it would be very helpful to repeat the equation also in the main text. A similar idea goes to the delta_m and delta_s terms - it will be very helpful to give a clear meaning of them, as nearly all analyses rely on knowing what they mean.

We thank the reviewer’s suggestion. We have now added the equation of the harm aversion model and provided more detailed description to the equations in the main text (Lines 150 ~155).

“We also modeled subjects’ choices using an influential model where subjects’ behavior could be characterized by the harm (electric shock) aversion parameter κ, reflecting the relative weights subjects assigned to ∆m and ∆s, the objective difference in money and shocks between the more and less painful options, respectively (∆V=(1-κ)∆m - κ∆s Eq.1, See Methods for details)[4–6]. Higher κ indicates that higher sensitivity is assigned to ∆s than ∆m and vice versa.”

(3b) There is one additional parameter gamma (choice consistency) in the model. Did the authors also examine the task-related difference of gamma? This might be important as some studies have shown that the other-oriented choice consistency may differ in different prosocial contexts.

To examine the task-related difference of choice consistency (γ), we compared the performance of 4 candidate models:

Model 1 (M1): The choice consistency parameter γ remains constant across shock recipients (self vs. other) and decision contexts (gain vs. loss).

Model 2 (M2): γ differs between the self- and other-recipient conditions, with γ_self and γ_other representing the choice consistency when pain is inflicted on him/her-self or the other-recipient.

Model 3 (M3): γ differs between the gain and loss conditions, with γ_gain and γ_loss representing the choice consistencies in the gain and loss contexts, respectively.

Model 4 (M4): γ varies across four conditions, with γ_self-gain, γ_other-gain, γ_self-loss and γ_other-loss capturing the choice consistency in each condition.

Supplementary Fig. 10 shows, after fitting all the models to subjects’ choice behavioral data, model 1 (M1) performed the best among all the four candidate models in both studies (1 & 2) with the lowest Bayesian Information Criterion (BIC). Therefore, we conclude that factors such as the shock recipients (self vs. other) and decision contexts (gain vs. loss) did not significantly influence subjects’ choice consistency and report model results using the single choice consistency parameter.

(3c) I am not fully convinced that the authors included two types of models: the harm aversion model and the logistic regression models. Indeed, the models look similar, and the authors have acknowledged that. But I wonder if there is a way to combine them? For example:

Choice ~ delta_V * context * recipient (*Oxt_v._placebo)

The calculation of delta_V follows Equation 1.

Or the conceptual question is, if the authors were interested in the specific and independent contribution of dalta_m and dalta_s to behavior, as their logistic model did, why did the authors examine the harm aversion first, where a parameter k is controlling for the trade-off? One way to find it out is to properly run different models and run model comparisons. In the end, it would be beneficial to only focus on the "winning" model to draw inferences.

The reviewer raised an excellent point here. According to the logistic regression model, we have:

Where P is the probability of selecting the less harmful option. Similarly, if we combine Eq.1 (∆V=1-κ)∆m-κ∆s) and Eq.2 ) of the harm aversion model, we have:

If we ignore the constant term β₀ from the logistic regression model, the harm aversion model is simply a reparameterization of the logistic regression model. The harm aversion model was implemented first to derive the harm aversion parameter (κ), which is an parameter in the range of [0 1] to quantify how subjects value the relative contribution of Δm and Δs between options in their decision processes. Since previous studies used the term κ_other-κ_self to define the magnitude of hyperaltruistic preference, we adopted similar approach to compare our results with previous research under the same theoretical framework. However, in order to investigate the independent contribution of Δm and Δs, we will have to take γ into account (we can see that the β_∆m and β_∆s in the logistic regression model are not necessarily correlated by nature; however, in the harm aversion model the coefficients (1-κ) and κ is always strictly negatively correlated (see Eq. 1). Only after multiplying γ, the correlation between γ(1-κ) and γκ will vary depending on the specific distribution of γ and κ). In summary, we followed the approach of previous research to estimate harm aversion parameter κ to compare our results with previous studies and to capture the relative influence between Δm and Δs. When we studied the contextual effects (gain vs. loss or placebo vs. control) on subjects’ behavior, we further investigated the contextual effect on how subjects evaluated Δm and Δs, respectively. The two models (logistic regression model and harm aversion model) in our study are mathematically the same and are not competitive candidate models. Instead, they represent different aspects from which our data can be examined.

We also compared the harm aversion model with and without the constant term β₀ in the choice function. Adding a constant term β₀ the above Equation 2 becomes:

As the following figure shows, the hyperaltruistic parameters (κ_other-κ_self) calculated from the harm aversion model with the constant term (panels A & B) have almost identical patterns as the model without the constant term (panels C & D, i.e. Figs. 2B & 4B in the original manuscript) in both studies.

Author response image 1.

Figs. 2B & 4B in the original manuscript) in both studies.

(3d) The interpretation of the main OXT results needs to be more cautious. According to the operationalization, "hyperaltruistic" is the reduction of pain of others (higher % of choosing the less painful option) relative to the self. But relative to the placebo (as baseline), OXT did not increase the % of choosing the less painful option for others, rather, it decreased the % of choosing the less painful option for themselves. In other words, the degree of reducing other's pain is the same under OXT and placebo, but the degree of benefiting self-interest is reduced under OXT. I think this needs to be unpacked, and some of the wording needs to be changed. I am not very familiar with the OXT literature, but I believe it is very important to differentiate whether OXT is doing something on self-oriented actions vs other-oriented actions. Relatedly, for results such as that in Figure 5A, it would be helpful to not only look at the difference but also the actual magnitude of the sensitivity to the shocks, for self and others, under OXT and placebo.

We thank the reviewer for this thoughtful comment. As the reviewer correctly pointed out, “hyperaltruism” can be defined as “higher % of choosing the less painful option to the others relative to the self”. Closer examination of the results showed that both the degrees of reducing other’s pain as well as reducing their own pain decreased under OXT (Figure 4A). More specifically, our results do not support the claim that “In other words, the degree of reducing others’ pain is the same under OXT and placebo, but the degree of benefiting self-interest is reduced under OXT.” Instead, the results show a significant reduction in the choice of less painful option under OXT treatment for both the self and other conditions (the interaction effect of OXT vs. placebo and self vs. other: F_1.45= 16.812, P < 0.001, η² = 0.272, simple effect OXT vs. placebo in the self- condition: F_1.45=59.332, P < 0.001, η² = 0.569, OXT vs. placebo in the other-condition: F_1.45= 14.626, P < 0.001, η² = 0.245, repeated ANOVA, see Figure 4A).

We also performed mixed-effect logistic regression analyses where subjects’ choices were regressed against  and  in different valences (gain vs. loss) and recipients (self vs. other) conditions in both studies 1 & 2 (Supplementary Figs. 1 & 6). As we replot supplementary Fig. 6 and panel B (included as Supplementary Fig. 8 in the supplementary materials) in the above figure, we found a significant treatment × ∆_s (differences in shock magnitude between the more and less painful options) interaction effect β=0.136±0.029P < =0.001, 95% CI=[-0.192, -0.079]), indicating that subject’s sensitivities towards pain were indeed different between the placebo and OXT treatments for both self and other conditions. Furthermore, the significant four-way ∆_s × treatment (OXT vs. Placebo) × context (gain vs. loss) × recipient (self vs. other) interaction effect (β=0.125±0.053, P=0.018 95% CI=[0.022, 0.228]) in the regression analysis, followed by significant simple effects (In the OXT treatment: ∆_s × recipient effect in the gain context: F_1.45= 7.622, P < 0.008, η² = 0.145; ∆_s × recipient effect in the loss context: F_1.45= 7.966, P 0.007, η² = 0.150, suggested that under OXT treatment, participants showed a greater sensitivity toward ∆_s (see asterisks in the OXT condition in panel B) in the other condition than the self-condition, thus restoring the hyperaltruistic behavior in loss context.

As the reviewer suggested, OXT’s effect on hyperaltruism does manifest separately on subjects’ harm sensitivities on self- and other-oriented actions. We followed the reviewer’s suggestions and examined the actual magnitude of the sensitivities to shocks for both the self and other treatments (panel B in the figure above). It’s clear that the administration of OXT (compared to the Placebo treatment, panel B in the figure above) significantly reduced participants’ pain sensitivity (treatment × ∆_s: β=-0.136±0.029, P < 0.001, 95% CI=[-0.192,-0.079]), yet also restored the harm sensitivity patterns in both the gain and loss conditions. These results are included in the supplementary figures (6 & 8) as well as in the main texts.

Recommendations:

(1) For Figures 2A-B, it would be great to calculate the correlation separately for gain and loss, as in other figures.

We speculate that the reviewer is referring to Figures 3A & B. Sorry that we did not present the correlations separately for the gain and loss contexts because the correlation between an individual’s IH (instrumental harm), IB (impartial beneficence) and hyperaltruistic preferences was not significantly modulated by the contextual factors. The interaction effects in both Figs. 3A & B and Supplementary Fig.5 (also see Table S1& S2) are as following: Study1 valence × IH effect: β=0.016±0.022, t₁₅₂=0.726, P=0.469; valence × IB effect: β=0.004±0.031, t₁₅₂=0.115, P=0.908; Study2 placebo condition: valence × IH effect: β=0.018±0.024, t₈₄=0.030 P=0.463; valence × IB effect: β=0.051±0.030, t₈₄=1.711, P=0.702. We have added these statistics to the main text following the reviewer’s suggestions.

(2) "by randomly drawing a shock increment integer ∆s (from 1 to 19) such that [...] did not exceed 20 (𝑆+ {less than or equal to} 20)." I am not sure if a random drawing following a uniform distribution can guarantee S is smaller than 20. More details are needed. Same for the monetary magnitude.

We are sorry for the lack of clarity in the method description. As for the task design, we followed adopted the original design from previous literature (Crockett et al., 2014, 2017). More specifically:

“Specifically, each trial was determined by a combination of the differences of shocks (Δs, ranging from 1 to 19, with increment of 1) and money (Δm, ranging from ¥0.2 to ¥19.8, with increment of ¥0.2) between the two options, resulting in a total of 19×99=1881 pairs of [Δs, Δm]. for each trial. To ensure the trials were suitable for most subjects, we evenly distributed the desired ratio Δm / (Δs + Δm) between 0.01 and 0.99 across 60 trials for each condition. For each trial, we selected the closest [Δs, Δm] pair from the [Δs, Δm] pool to the specific Δm / (Δs + Δm) ratio, which was then used to determine the actual money and shock amounts of two options. The shock amount (S_less) for the less painful option was an integer drawn from the discrete uniform distribution [1-19], constraint by S_less + ∆s < 20. Similarly, the money amount (M_less) for the less painful option was drawn from a discrete uniform distribution [¥0.2 - ¥19.8], with the constraint of M_less + ∆m < 20. Once the S_lessand M_less were selected, the shock (S_more) and money (M_more) magnitudes for the more painful option were calculated as: S_more = S_less + ∆s, M_more = M_less + ∆m”  

We have added these details to the methods section (Lines 520-533).

Reviewer #2:

(1) The theoretical hypothesis needs to be better justified. There are studies addressing the neurobiological mechanism of hyperaltruistic tendency, which the authors unfortunately skipped entirely.

Also in recommendation #1:

(1) In the Introduction, the authors claim that "the mechanistic account of the hyperaltruistic phenomenon remains unknown". I think this is too broad of a criticism and does not do justice to prior work that does provide some mechanistic account of this phenomenon. In particular, I was surprised that the authors did not mention at all a relevant fMRI study that investigates the neural mechanism underlying hyperaltruistic tendency (Crockett et al., 2017, Nature Neuroscience). There, the researchers found that individual differences in hyperaltruistic tendency in the same type of moral decision-making task is better explained by reduced neural responses to ill-gotten money (Δm in the Other condition) in the brain reward system, rather than heightened neural responses to others' harm. Moreover, such neural response pattern is related to how an immoral choice would be judged (i.e., blamed) by the community. Since the brain reward system is consistently involved in Oxytocin's role in social cognition and decision-making (e.g., Dolen & Malenka, 2014, Biological Psychiatry), it is important to discuss the hypothesis and results of the present research in the context of this literature.

We totally agree with the reviewer that the expression “mechanistic account of the hyperaltruistic phenomenon remains unknown” in our original manuscript can be misleading to the audience. Indeed, we were aware of the major findings in the field and cited all the seminal work of hyperaltruism and its related neural mechanism (Crockett et al., 2014, 2015, 2017). We have changed the texts in the introduction to better reflect this point and added further discussion as to how oxytocin might play a role:

“For example, it was shown that the hyperaltruistic preference modulated neural representations of the profit gained from harming others via the functional connectivity between the lateral prefrontal cortex, a brain area involved in moral norm violation, and profit sensitive brain regions such as the dorsal striatum6.” (Lines 41~45)

“Oxytocin has been shown to play a critical role in social interactions such as maternal attachment, pair bonding, consociate attachment and aggression in a variety of animal models[42,43]. Humans are endowed with higher cognitive and affective capacities and exhibit far more complex social cognitive patterns[44]. ” (Lines 86~90)

(2) There are some important inconsistencies between the preregistration and the actual data collection/analysis, which the authors did not justify.

Also in recommendations:

(4) It is laudable that the authors pre-registered the procedure and key analysis of the Oxytocin study and determined the sample size beforehand. However, in the preregistration, the authors claimed that they would recruit 30 participants for Experiment 1 and 60 for Experiment 2, without justification. In the paper, they described a "prior power analysis", which deviated from their preregistration. It is OK to deviate from preregistration, but this needs to be explicitly mentioned and addressed (why the deviation occurred, why the reported approach was justifiable, etc.).

We sincerely appreciate the reviewer’s thorough assessment of our manuscript. In the more exploratory study 1, we found that the loss decision context effectively diminished subjects’ hyperaltruistic preference. Based on this finding, we pre-registered study 2 and hypothesized that: 1) The administration of OXT may salvage subject’s hyperaltruistic preference in the loss context; 2) The administration of OXT may reduce subjects’ sensitivities towards electric shocks (but not necessarily their moral preference), due to the well-established results relating OXT to enhanced empathy for others (Barchi-Ferreira & Osório, 2021; Radke et al., 2013) and the processing of negative stimuli(Evans et al., 2010; Kirsch et al., 2005; Wu et al., 2020); and 3) The OXT effect might be context specific, depending on the particular combination of valence (gain vs. loss) and shock recipient (self vs. other) (Abu-Akel et al., 2015; Kapetaniou et al., 2021; Ma et al., 2015).

As our results suggested, the administration of OXT indeed restored subjects’ hyperaltruistic preference (confirming hypothesis 1, Figure 4A). Also, OXT decreased subjects’ sensitivities towards electric shocks in both the gain and loss conditions (supplementary Fig. 6 and supplementary Fig. 8), consistent with our second hypothesis. We must admit that our hypothesis 3 was rather vague, since a seminal study clearly demonstrated the context-dependent effect of OXT in human cooperation and conflict depending on the group membership of the subjects (De Dreu et al., 2010, 2020). Although our results partially validated our hypothesis 3 (supplementary Fig. 6), we did not make specific predictions as to the direction and the magnitude of the OXT effect.

The main inconsistency is related to the sample size. When we carried out study 1, we recruited both male and female subjects. After we identified the context effect on the hyperaltruistic preference, we decided to pre-register and perform study 2 (the OXT study). We originally made a rough estimate of 60 male subjects for study 2. While conducting study 2, we also went through the literature of OXT effect on social behavior and realized that the actual subject number around 45 might be enough to detect the main effect of OXT. Therefore, we settled on the number of 46 (study 2) reported in the manuscript. Correspondingly, we increased the subject number in study 1 to the final number of 80 (40 males) to make sure the subject number is enough to detect a small-to-medium effect, as well as to have a fair comparison between study 1 and 2 (roughly equal number of male subjects). It should be noted that although we only reported all the subjects (male & female) results of study 1 in the manuscript, the main results remain very similar if we only focus on the results of male subjects in study 1 (see the figure below). We believe that these results, together with the placebo treatment group results in study 2 (male only), confirmed the validity of our original finding.

Author response image 2.

Author response image 3.

We have included additional texts (Lines 447 ~ 452) in the Methods section for the discrepancy between the preregistered and actual sample sizes in the revised manuscript:

“It should be noted that in preregistration we originally planned to recruit 60 male subjects for Study 2 but ended up recruiting 46 male subjects (mean age =  years) based on the sample size reported in previous oxytocin studies[57,69]. Additionally, a power analysis suggested that the sample size > 44 should be enough to detect a small to median effect size of oxytocin (Cohen’s d=0.24, α=0.05, β=0.8) using a 2 × 2 × 2 within-subject design[76].”

(3) Some of the exploratory analysis seems underpowered (e.g., large multiple regression models with only about 40 participants).

We thank the reviewer’s comments and appreciate the concern that the sample size would be an issue affecting the results reliability in multiple regression analysis.

In Fig. 2, the multiple regression analyses were conducted after we observed a valence-dependent effect on hyperaltruism (Fig. 2A) and the regression was constructed accordingly:

Choice ~ ∆s *context*recipient + ∆m *context*recipient+(1+ ∆s *context*recipient + ∆s*context*recipient | subject)

Where ∆s and ∆m indicate the shock level and monetary reward difference between the more and loss painful options, context as the monetary valence (gain vs. loss) and recipient as the identity of the shock recipient (self vs. other).

Since we have 240 trials for each subject and a total of 80 subjects in Study 1, we believe that this is a reasonable regression analysis to perform.

In Fig. 3, the multiple regression analyses were indeed exploratory. More specifically, we ran 3 multiple linear regressions:

hyperaltruism~EC*context+IH*context+IB*context

Relative harm sensitivity~ EC*context+IH*context+IB*context

Relative money sensitivity~ EC*context+IH*context+IB*context

Where Hyperaltruism is defined as κ_other - κ_self, Relative harm sensitivity as otherβ_∆s - selfβ_∆s and Relative monetary sensitivity as otherβ_∆m - selfβ_∆m. EC (empathic concern), IH (instrumental harm) and IB (impartial beneficence) were subjects’ scores from corresponding questionnaires.

For the first regression, we tested whether EC, IH and IB scores were related to hyperaltruism and it should be noted that this was tested on 80 subjects (Study 1). After we identified the effect of IH on hyperaltruism, we ran the following two regressions. The reason we still included IB and EC as predictors in these two regression analyses was to remove potential confounds caused by EC and IB since previous research indicated that IB, IH and EC could be correlated (Kahane et al., 2018).

In study 2, we performed the following regression analyses again to validate our results (Placebo treatment in study 2 should have similar results as found in study 1).

Relative harm sensitivity~ EC*context+IH*context+IB*context

Relative money sensitivity~ EC*context+IH*context+IB*context

Again, we added IB and EC only to control for the nuance effects by the covariates. As indicated in Fig. 5 C-D, the placebo condition in study 2 replicated our previous findings in study 1 and OXT administration effectively removed the interaction effect between IH and valence (gain vs. loss) on subjects’ relative harm sensitivity.

To more objectively present our data and results, we have changed the texts in the results section and pointed out that the regression analysis:

hyperaltruism~EC*context+IH*context+IB*context

was exploratory (Lines 186-192).

“We tested how hyperaltruism was related to both IH and IB across decision contexts using an exploratory multiple regression analysis. Moral preference, defined as κ_other - κ_self, was negatively associated with IH (β=-0.031±0.011, t₁₅₆=-2.784, P =0.006) but not with IB (β=0.008±0.016, t₁₅₆=0.475, P=0.636) across gain and loss contexts, reflecting a general connection between moral preference and IH (Fig. 3A & B).”

(4) Inaccurate conceptualization of utilitarian psychology and the questionnaire used to measure it.

Also in recommendations:

(2) Throughout the paper, the authors placed lots of weight on individual differences in utilitarian psychology and the Oxford Utilitarianism Scale (OUS). I am not sure this is the best individual difference measure in this context. I don't see a conceptual fit between the psychological construct that OUS reflects, and the key psychological processes underlying the behaviors in the present study. As far as I understand it, the conceptual core of utilitarian psychology that OUS captures is the maximization of greater goods. Neither the Instrumental Harm (IH) component nor the Impartial Beneficence (IB) component reflects a tradeoff between the personal interests of the decision-making agent and a moral principle. The IH component is about the endorsement of harming a smaller number of individuals for the benefit of a larger number of individuals. The IB component is about treating self, close others, and distant others equally. However, the behavioral task used in this study is neither about distributing harm between a smaller number of others and a larger number of others nor about benefiting close or distant others. The fact that IH showed some statistical association with the behavioral tendency in the present data set could be due to the conceptual overlap between IH and an individual's tendency to inflict harm (e.g., psychopathy; Table 7 in Kahane et al., 2018, which the authors cited). I urge the authors to justify more why they believe that conceptually OUS is an appropriate individual difference measure in the present study, and if so, interpret their results in a clearer and justifiable manner (taking into account the potential confound of harm tendency/psychopathy).

We thank the reviewer for the thoughtful comment and agree that “IH component is about the endorsement of harming a smaller number of individuals for the benefit of a larger number of individuals. The IB component is about treating self, close others, and distant others equally”. As we mentioned in the previous response to the reviewer, we first ran an exploratory multiple linear regression analysis of hyperaltruistic preference (κ_other - κ_self) against IB and IH in study 1 based on the hypothesis that the reduction of hyperaltruistic preference in the loss condition might be due to 1) subjects’ altered altitudes between IB and hyperaltruistic preference between the gain and loss conditions, and/or 2) the loss condition changed how the moral norm was perceived and therefore affected the correlation between IH and hyperaltruistic preference. As Fig. 3 shows, we did not find a significant IB effect on hyperaltruistic preference (κ_other - κ_self), nor on the relative harm or money sensitivity (supplementary Fig. 3). These results excluded the possibility that subjects with higher IB might treat self and others more equally and therefore show less hyperaltruistic preference. On the other hand, we found a strong correlation between hyperaltruistic preference and IH (Fig. 3A): subjects with higher IH scores showed less hyperaltruistic preference. Since the hyperaltruistic preference (κ_other - κ_self) is a compound variable and we further broke it down to subjects’ relative sensitivity to harm and money (other β_∆s - self β_∆s and other β_∆m - self β_∆m, respectively). The follow up regression analyses revealed that the correlation between subjects’ relative harm sensitivity and IH was altered by the decision contexts (gain vs. loss, Fig. 3C-D). These results are consistent with our hypothesis that for subjects to engage in the utilitarian calculation, they should first realize that there is a moral dilemma (harming others to make monetary gain in the gain condition). When there is less perceived moral conflict (due to the framing of decision context as avoiding loss in the loss condition), the correlation between subjects’ relative harm sensitivity and IH became insignificant (Fig. 3C). It is worth noting that these results were further replicated in the placebo condition of study 2, further indicating the role of OXT is to affect how the decision context is morally framed.

The reviewer also raised an interesting possibility that the correlation between subject’s behavioral tendency and IH may be confounded by the fact that IH is also correlated with other traits such as psychopathy. Indeed, in the Kahane et al., 2018 paper, the authors showed that IH was associated with subclinical psychopathy in a lay population. Although we only collected and included IB and Empathic concern (EC) scores as control variables and in principle could not rule out the influence of psychopathy, we argue it is unlikely the case. First, psychopaths by definition “only care about their own good” (Kahane et al., 2018). However, subjects in our studies, as well as in previous research, showed greater aversion to harming others (compared to harming themselves) in the gain conditions. This is opposite to the prediction of psychopathy. Even in the loss condition, subjects showed similar levels of aversion to harming others (vs. harming themselves), indicating that our subjects valuated their own and others’ well-being similarly. Second, although there appears to be an association between utilitarian judgement and psychopathy(Glenn et al., 2010; Kahane et al., 2015), the fact that people also possess a form of universal or impartial beneficence in their utilitarian judgements suggest psychopathy alone is not a sufficient variable explaining subjects’ hyperaltruistic behavior.

We have thus rewritten part of the results to clarify our rationale for using the Oxford Utilitarianism Scale (especially the IH and IB) to establish the relationship between moral traits and subjects’ decision preference (Lines 212-215):

“Furthermore, our results are consistent with the claim that profiting from inflicting pains on another person (IH) is inherently deemed immoral1. Hyperaltruistic preference, therefore, is likely to be associated with subjects’ IH dispositions.”

(3) Relatedly, in the Discussion, the authors mentioned "the money-pain trade-off task, similar to the well-known trolley dilemma". I am not sure if this statement is factually accurate because the "well-known trolley dilemma" is about a disinterested third-party weighing between two moral requirements - "greatest good for the greatest number" (utilitarianism) and "do no harm" (Kantian/deontology), not between a moral requirement and one's own monetary interest (which is the focus of the present study). The analogy would be more appropriate if the task required the participants to trade off between, for example, harming one person in exchange for a charitable donation, as a recent study employed (Siegel et al., 2022, A computational account of how individuals resolve the dilemma of dirty money. Scientific reports). I urge the authors to go through their use of "utilitarian/utilitarianism” in the paper and make sure their usage aligns with the definition of the concept and the philosophical implications.

We thank the reviewer for prompting us to think over the difference between our task and the trolley dilemma. Indeed, the trolley dilemma refers to a disinterested third-party’s decision between two moral requirements, namely, the utilitarianism and deontology. In our study, when the shock recipient was “other”, our task could be interpreted as either the decision between “moral norm of no harm (deontology) and one’s self-interest maximization (utilitarian)”, or a decision between “greatest good for both parties (utilitarian) vs. do no harm (deontology)”, though the latter interpretation typically requires differential weighing of own benefits versus the benefits of others(Fehr & Schmidt, 1999; Saez et al., 2015). In fact, it could be argued that the utilitarianism account applies not only to the third party’s well-being, but also to our own well-being, or to “that of those near or dear to us” (Kahane et al., 2018).

We acknowledge that there may lack a direct analogy between our task and the trolley dilemma and therefore have deleted the trolley example in the discussion.

(5) Related to the above point, the sample size of Study 2 was calculated based on the main effect of oxytocin. However, the authors also reported several regression models that seem to me more like exploratory analyses. Their sample size may not be sufficient for these analyses. The authors should: a) explicitly distinguish between their hypothesis-driven analysis and exploratory analysis; b) report achieved power of their analysis.

We appreciate the reviewer’s thorough reading of our manuscript. Following the reviewer’s suggestions, we have explicitly stated in the revised manuscript which analyses were exploratory, and which were hypothesis driven. Following the reviewer’s request, we added the achieved power into the main texts (Lines 274-279):

“The effect size (Cohen’s f²) for this exploratory analysis was calculated to be 0.491 and 0.379 for the placebo and oxytocin conditions, respectively. The post hoc power analysis with a significance level of α = 0.05, 7 regressors (IH, IB, EC, decision context, IH×context, IB×context, and EC×context), and sample size of N = 46 yielded achieved power of 0.910 (placebo treatment) and 0.808 (oxytocin treatment).”

(6) Do the authors collect reaction times (RT) information? Did the decision context and oxytocin modulate RT? Based on their procedure, it seems that the authors adopted a speeded response task, therefore the RT may reflect some psychological processes independent of choice. It is also possible (and recommended) that the authors use the drift-diffusion model to quantify latent psychological processes underlying moral decision-making. It would be interesting to see if their manipulations have any impact on those latent psychological processes, in addition to explicit choice, which is the endpoint product of the latent psychological processes. There are some examples of applying DDM to this task, which the authors could refer to if they decide to go down this route (Yu et al, 2021, How peer influence shapes value computation in moral decision-making. Cognition.)

We did collect the RT information for this experiment. As demonstrated in the figure below, participants exhibited significantly longer RT in the loss context compared to the gain context (Study1: the main effect of decision context: F_1,79=20.043, P < 0.001, η² =0.202; Study2-placebo: F_1.45=17.177, P < 0.001, η² =0.276). In addition to this effect of context, decisions were significantly slower in the other-condition compared to the self-condition

(Study1: the main effect of recipient: F_1,79=4.352, P < 0.040, η² =0.052; Study2-placebo: F_1,45=5.601, P < 0.022, η² =0.111) which replicates previous research findings (Crockett et al., 2014). However, the differences in response time between recipients was not modulated by decision context (Study1: context × recipient interaction: F_1,79=1.538, P < 0.219, η² =0.019; Study2-placebo: F_1,45=2.631, P < 0.112, η² =0.055). Additionally, the results in the oxytocin study (study 2) revealed no evidence supporting any effect of oxytocin on reaction time. Neither the main effect (treatment: placebo vs. oxytocin) nor the interaction effect of oxytocin on response time was statistically significant (main effect of OXT treatment: F_1,45=2.380, P < 0.230, η² =0.050; treatment × context: F_1,45=2.075, P < 0.157η² =0.044; treatment × recipient: F_1,45=0.266, P < 0.609, η² =0.006; treatment × context × recipient: F_1,45=2.909, P < 0.095, η² =0.061).;

Author response image 4.

We also agree that it would be interesting to also investigate how the OXT might impact the dynamics of the decision process using a drift-diffusion model (DDM). However, we have already showed in the original manuscript that the OXT increased subjects’ relative harm sensitivities. If a canonical DDM is adopted here, then such an OXT effect is more likely to correspond to the increased drift rate for the relative harm sensitivity, which we feel still aligns with the current framework in general. In future studies, including further manipulations such as time pressure might be a more comprehensive approach to investigate the effect of OXT on DDM related decision variables such as attribute drift rate, initial bias, decision threshold and attribute synchrony.

(7) This is just a personal preference, but I would avoid metaphoric language in a scientific paper (e.g., rescue, salvage, obliterate). Plain, neutral English terms can express the same meaning clearly (e.g., restore, vanish, eliminate).

Again, we thank the reviewer for the suggestion and have since modified the terms.

Reviewer #3:

The primary weakness of the paper concerns its framing. Although it purports to be measuring "hyper-altruism" it does not provide evidence to support why any of the behavior being measured is extreme enough to warrant the modifier "hyper" (and indeed throughout I believe the writing tends toward hyperbole, using, e.g., verbs like "obliterate" rather than "reduce"). More seriously, I do not believe that the task constitutes altruism, but rather the decision to engage, or not engage, in instrumental aggression.

We agree with the reviewer (and reviewer # 2) that plain and clear English should be used to describe our results and have since modified those terms.

However, the term “hyperaltruism”, which is the main theme of our study, was originally proposed by a seminal paper (Crockett et al., 2014) and has since been widely adopted in related studies (Crockett et al., 2014, 2015, 2017; Volz et al., 2017; Zhan et al., 2020). The term “hyperaltruism” was introduced to emphasize the difference from altruism (Chen et al., 2024; FeldmanHall et al., 2015; Hu et al., 2021; Hutcherson et al., 2015; Lockwood et al., 2017; Xiong et al., 2020). Hyperaltruism does not indicate extreme altruism. Instead, it simply reflects the fact that “we are more willing to sacrifice gains to spare others from harm than to spare ourselves from harm” (Volz et al., 2017). In other words, altruism refers to people’s unselfish regard for or devotion to the welfare of others, and hyperaltruism concerns subject’s own cost-benefit preference as the reference point and highlights the “additional” altruistic preference when considering other’s welfare. For example, in the altruistic experimental design, altruism is characterized by the degree to which subjects take other people’s welfare into account (left panel). However, in a typical hyperaltruism task design (right panel), hyperaltruistic preference is operationally defined as the difference (κ_other - κ_self) between the degrees to which subjects value others’ harm (κ_other) and their own harm (κ_self).

Author response image 5.

I found it surprising that a paradigm that entails deciding to hurt or not hurt someone else for personal benefit (whether acquiring a financial gain or avoiding a loss) would be described as measuring "altruism." Deciding to hurt someone for personal benefit is the definition of instrumental aggression. I did not see that in any of the studies was there a possibility of acting to benefit the other participant in any condition. Altruism is not equivalent to refraining from engaging in instrumental aggression. True altruism would be to accept shocks to the self for the other's benefit (e.g., money).  The interpretation of this task as assessing instrumental aggression is supported by the fact that only the Instrumental Harm subscale of the OUS was associated with outcomes in the task, but not the Impartial Benevolence subscale. By contrast, the IB subscale is the one more consistently associated with altruism (e.g,. Kahane et al 2018; Amormino at al, 2022) I believe it is important for scientific accuracy for the paper, including the title, to be re-written to reflect what it is testing.

Again, as we mentioned in the previous response, hyperaltruism is a term coined almost a decade ago and has since been widely adopted in the research field. We are afraid that switching such a term would be more likely to cause confusion (instead of clarity) among audience.

Also, from the utilitarian perspective, the gain or loss (or harm) occurred to someone else is aligned on the same dimension and there is no discontinuity between gains and losses. Therefore, taking actions to avoid someone else’s loss can also be viewed as altruistic behavior, similar to choices increasing other’s welfare (Liu et al., 2020).

Relatedly: in the introduction I believe it would be important to discuss the non-symmetry of moral obligations related to help/harm--we have obligations not to harm strangers but no obligation to help strangers. This is another reason I do not think the term "hyper altruism" is a good description for this task--given it is typically viewed as morally obligatory not to harm strangers, choosing not to harm them is not "hyper" altruistic (and again, I do not view it as obviously altruism at all).

We agree with the reviewer’s point that we have the moral obligations not to harm others but no obligation to help strangers (Liu et al., 2020). In fact, this is exactly what we argued in our manuscript: by switching the decision context from gains to losses, subjects were less likely to perceive the decisions as “harming others”. Furthermore, after the administration of OXT, making decisions in both the gain and loss contexts were more perceived by subjects as harming others (Fig. 6A).

The framing of the role of OT also felt incomplete. In introducing the potential relevance of OT to behavior in this task, it is important to pull in evidence from non-human animals on origins of OT as a hormone selected for its role in maternal care and defense (including defensive aggression). The non-human animal literature regarding the effects of OT is on the whole much more robust and definitive than the human literature. The evidence is abundant that OT motivates the defensive care of offspring of all kinds. My read of the present OT findings is that they increase participants' willingness to refrain from shocking strangers even when incurring a loss (that is, in a context where the participant is weighing harm to themselves versus harm to the other). It will be important to explain why OT would be relevant to refraining from instrumental aggression, again, drawing on the non-human animal literature.

We thank the reviewer’s comments and agree that the current understanding of the link between our results of OT with animal literature can be at best described as vague and intriguing. Current literature on OT in animal research suggests that the nucleus accumbens (NAc) oxytocin might play the critical role in social cognition and reinforcing social interactions (Dölen et al., 2013; Dölen & Malenka, 2014; Insel, 2010). Though much insight has already been gained from animal studies, in humans, social interactions can take a variety of different forms, and the consociate recognition can also be rather dynamic. For example, male human participants with self-administered OT showed higher trust and cooperation towards in-group members but more defensive aggression towards out-group members (De Dreu et al., 2010). In another human study, participants administered with OT showed more coordinated out-group attack behavior, suggesting that OT might increase in-group efficiency at the cost of harming out-group members (Zhang et al., 2019). It is worth pointing out that in both experiments, the participant’s group membership was artificially assigned, thus highlighting the context-dependent nature of OT effect in humans.

In our experiment, more complex and higher-level social cognitive processes such as moral framing and moral perception are involved, and OT seems to play an important role in affecting these processes. Therefore, we admit that this study, like the ones mentioned above, is rather hard to find non-human animal counterpart, unfortunately. Instead of relating OT to instrumental aggression, we aimed to provide a parsimonious framework to explain why the “hyperaltruism” disappeared in the loss condition, and, with the OT administration, reappeared in both the gain and loss conditions while also considering the effects of other relevant variables.  

We concur with the reviewer’s comments about the importance of animal research and have since added the following paragraph into the revised manuscript (Line 86~90) as well as in the discussion:

“Oxytocin has been shown to play a critical role in social interactions such as maternal attachment, pair bonding, consociate attachment and aggression in a variety of animal models[42,43]. Humans are endowed with higher cognitive and affective capacities and exhibit far more complex social cognitive patterns[44].”

Another important limitation is the use of only male participants in Study 2. This was not an essential exclusion. It should be clear throughout sections of the manuscript that this study's effects can be generalized only to male participants.

We thank the reviewer’s comments. Prior research has shown sex differences in oxytocin’s effects (Fischer-Shofty et al., 2013; Hoge et al., 2014; Lynn et al., 2014; Ma et al., 2016; MacDonald, 2013). Furthermore, with the potential confounds of OT effect due to the menstrual cycles and potential pregnancy in female subjects, most human OT studies have only recruited male subjects (Berends et al., 2019; De Dreu et al., 2010; Fischer-Shofty et al., 2010; Ma et al., 2016; Zhang et al., 2019). We have modified our manuscript to emphasize that study 2 only recruited male subjects.

Recommendations:

I believe the authors have provided an interesting and valuable dataset related to the willingness to engage in instrumental aggression - this is not the authors' aim, although also an important aim. Future researchers aiming to build on this paper would benefit from it being framed more accurately.

Thus, I believe the paper must be reframed to accurately describe the nature of the task as assessing instrumental aggression. This is also an important goal, as well-designed laboratory models of instrumental aggression are somewhat lacking.

Please see our response above that to have better connections with previous research, we believe that the term hyperaltruism might align better with the main theme for this study.

The research literature on other aggression tasks should also be brought in, as I believe these are more relevant to the present study than research studies on altruism that are primarily donation-type tasks. It should be added to the limitations of how different aggression in a laboratory task such as this one is from real-world immoral forms of aggression. Arguably, aggression in a laboratory task in which all participants are taking part voluntarily under a defined set of rules, and in which aggression constrained by rules is mutual, is similar to aggression in sports, which is not considered immoral. Whether responses in this task would generalize to immoral forms of aggression cannot be determined without linking responses in the task to some real-world outcome.

We agree with the reviewer that “aggression in a lab task …. is similar to aggression in sports”. Our starting point was to investigate the boundary conditions for the hyperaltruism (though we don’t deny that there is an aggression component in hyperaltruism, given the experiment design we used). In other words, the dependent variable we were interested in was the difference between “other” and “self” aggression, not the aggression itself. Our results showed that by switching the decision context from the monetary gain environment to the loss condition, human participants were willing to bear similar amounts of monetary loss to spare others and themselves from harm. That is, hyperaltruism disappeared in the loss condition. We interpreted this result as the loss condition prompted subjects to adopt a different moral framework (help vs. harm, Fig. 6A) and subjects were less influenced by their instrumental harm personality trait due to the change of moral framework (Fig. 3C). In the following study (study 2), we further tested this hypothesis and verified that the administration of OT indeed increased subjects’ perception of the task as harming others for both gain and loss conditions (Fig. 6A), and such moral perception mediated the relationship between subject’s personality traits (instrumental harm) and their relative harm sensitivities (the difference of aggression between the other- and self-conditions). We believe the moral perception framework and that OT directly modulates moral perception better account for subjects’ context-dependent choices than hypothesizing OT’s context-dependent modulation effects on aggression.

The language should also be toned down--the use of phrases like "hyper altruism" (without independent evidence to support that designation) and "obliterate" rather than "reduce" or "eliminate" are overly hyperbolic.

We have changed terms such as “obliterate” and “eliminate” to plain English, as the reviewer suggested.

Reference

Abu-Akel, A., Palgi, S., Klein, E., Decety, J., & Shamay-Tsoory, S. (2015). Oxytocin increases empathy to pain when adopting the other- but not the self-perspective. Social Neuroscience, 10(1), 7–15.

Barchi-Ferreira, A., & Osório, F. (2021). Associations between oxytocin and empathy in humans: A systematic literature review. Psychoneuroendocrinology, 129, 105268.

Berends, Y. R., Tulen, J. H. M., Wierdsma, A. I., van Pelt, J., Feldman, R., Zagoory-Sharon, O., de Rijke, Y. B., Kushner, S. A., & van Marle, H. J. C. (2019). Intranasal administration of oxytocin decreases task-related aggressive responses in healthy young males. Psychoneuroendocrinology, 106, 147–154.

Chen, J., Putkinen, V., Seppälä, K., Hirvonen, J., Ioumpa, K., Gazzola, V., Keysers, C., & Nummenmaa, L. (2024). Endogenous opioid receptor system mediates costly altruism in the human brain. Communications Biology, 7(1), 1–11.

Crockett, M. J., Kurth-Nelson, Z., Siegel, J. Z., Dayan, P., & Dolan, R. J. (2014). Harm to others outweighs harm to self in moral decision making. Proceedings of the National Academy of Sciences of the United States of America, 111(48), 17320–17325.

Crockett, M. J., Siegel, J. Z., Kurth-Nelson, Z., Dayan, P., & Dolan, R. J. (2017). Moral transgressions corrupt neural representations of value. Nature Neuroscience, 20(6), 879–885.

Crockett, M. J., Siegel, J. Z., Kurth-Nelson, Z., Ousdal, O. T., Story, G., Frieband, C., Grosse-Rueskamp, J. M., Dayan, P., & Dolan, R. J. (2015). Dissociable Effects of Serotonin and Dopamine on the Valuation of Harm in Moral Decision Making. Current Biology, 25(14), 1852–1859.

De Dreu, C. K. W., Greer, L. L., Handgraaf, M. J. J., Shalvi, S., Van Kleef, G. A., Baas, M., Ten Velden, F. S., Van Dijk, E., & Feith, S. W. W. (2010). The Neuropeptide Oxytocin Regulates Parochial Altruism in Intergroup Conflict Among Humans. Science, 328(5984), 1408–1411.

De Dreu, C. K. W., Gross, J., Fariña, A., & Ma, Y. (2020). Group Cooperation, Carrying-Capacity Stress, and Intergroup Conflict. Trends in Cognitive Sciences, 24(9), 760–776.

Dölen, G., Darvishzadeh, A., Huang, K. W., & Malenka, R. C. (2013). Social reward requires coordinated activity of nucleus accumbens oxytocin and serotonin. Nature, 501(7466), 179–184.

Dölen, G., & Malenka, R. C. (2014). The Emerging Role of Nucleus Accumbens Oxytocin in Social Cognition. Biological Psychiatry, 76(5), 354–355.

Evans, S., Shergill, S. S., & Averbeck, B. B. (2010). Oxytocin Decreases Aversion to Angry Faces in an Associative Learning Task. Neuropsychopharmacology, 35(13), 2502–2509.

Fehr, E., & Schmidt, K. M. (1999). A Theory of Fairness, Competition, and Cooperation*. The Quarterly Journal of Economics, 114(3), 817–868.

FeldmanHall, O., Dalgleish, T., Evans, D., & Mobbs, D. (2015). Empathic concern drives costly altruism. Neuroimage, 105, 347–356.

Fischer-Shofty, M., Levkovitz, Y., & Shamay-Tsoory, S. G. (2013). Oxytocin facilitates accurate perception of competition in men and kinship in women. Social Cognitive and Affective Neuroscience, 8(3), 313–317.

Fischer-Shofty, M., Shamay-Tsoory, S. G., Harari, H., & Levkovitz, Y. (2010). The effect of intranasal administration of oxytocin on fear recognition. Neuropsychologia, 48(1), 179–184.

Glenn, A. L., Koleva, S., Iyer, R., Graham, J., & Ditto, P. H. (2010). Moral identity in psychopathy. Judgment and Decision Making, 5(7), 497–505.

Hoge, E. A., Anderson, E., Lawson, E. A., Bui, E., Fischer, L. E., Khadge, S. D., Barrett, L. F., & Simon, N. M. (2014). Gender moderates the effect of oxytocin on social judgments. Human Psychopharmacology: Clinical and Experimental, 29(3), 299–304.

Hu, J., Hu, Y., Li, Y., & Zhou, X. (2021). Computational and Neurobiological Substrates of Cost-Benefit Integration in Altruistic Helping Decision. Journal of Neuroscience, 41(15), 3545–3561.

Hutcherson, C. A., Bushong, B., & Rangel, A. (2015). A Neurocomputational Model of Altruistic Choice and Its Implications. Neuron, 87(2), 451–462.

Insel, T. R. (2010). The Challenge of Translation in Social Neuroscience: A Review of Oxytocin, Vasopressin, and Affiliative Behavior. Neuron, 65(6), 768–779.

Kahane, G., Everett, J. A. C., Earp, B. D., Caviola, L., Faber, N. S., Crockett, M. J., & Savulescu, J. (2018). Beyond sacrificial harm: A two-dimensional model of utilitarian psychology. Psychological Review, 125(2), 131–164.

Kahane, G., Everett, J. A. C., Earp, B. D., Farias, M., & Savulescu, J. (2015). ‘Utilitarian’ judgments in sacrificial moral dilemmas do not reflect impartial concern for the greater good. Cognition, 134, 193–209.

Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica, 47(2), 263.

Kapetaniou, G. E., Reinhard, M. A., Christian, P., Jobst, A., Tobler, P. N., Padberg, F., & Soutschek, A. (2021). The role of oxytocin in delay of gratification and flexibility in non-social decision making. eLife, 10, e61844.

Kirsch, P., Esslinger, C., Chen, Q., Mier, D., Lis, S., Siddhanti, S., Gruppe, H., Mattay, V. S., Gallhofer, B., & Meyer-Lindenberg, A. (2005). Oxytocin Modulates Neural Circuitry for Social Cognition and Fear in Humans. The Journal of Neuroscience, 25(49), 11489–11493.

Liu, J., Gu, R., Liao, C., Lu, J., Fang, Y., Xu, P., Luo, Y., & Cui, F. (2020). The Neural Mechanism of the Social Framing Effect: Evidence from fMRI and tDCS Studies. The Journal of Neuroscience, 40(18), 3646–3656.

Liu, Y., Li, L., Zheng, L., & Guo, X. (2017). Punish the Perpetrator or Compensate the Victim? Gain vs. Loss Context Modulate Third-Party Altruistic Behaviors. Frontiers in Psychology, 8, 2066.

Lockwood, P. L., Hamonet, M., Zhang, S. H., Ratnavel, A., Salmony, F. U., Husain, M., & Maj, A. (2017). Prosocial apathy for helping others when effort is required. Nature Human Behaviour, 1(7), 131–131.

Losecaat Vermeer, A. B., Boksem, M. A. S., & Sanfey, A. G. (2020). Third-party decision-making under risk as a function of prior gains and losses. Journal of Economic Psychology, 77, 102206.

Lynn, S. K., Hoge, E. A., Fischer, L. E., Barrett, L. F., & Simon, N. M. (2014). Gender differences in oxytocin-associated disruption of decision bias during emotion perception. Psychiatry Research, 219(1), 198–203.

Ma, Y., Liu, Y., Rand, D. G., Heatherton, T. F., & Han, S. (2015). Opposing Oxytocin Effects on Intergroup Cooperative Behavior in Intuitive and Reflective Minds. Neuropsychopharmacology, 40(10), 2379–2387.

Ma, Y., Shamay-Tsoory, S., Han, S., & Zink, C. F. (2016). Oxytocin and Social Adaptation: Insights from Neuroimaging Studies of Healthy and Clinical Populations. Trends in Cognitive Sciences, 20(2), 133–145.

MacDonald, K. S. (2013). Sex, Receptors, and Attachment: A Review of Individual Factors Influencing Response to Oxytocin. Frontiers in Neuroscience, 6. 194.

Markiewicz, Ł., & Czupryna, M. (2018). Cheating: One Common Morality for Gain and Losses, but Two Components of Morality Itself. Journal of Behavior Decision Making. 33(2), 166-179.

Pachur, T., Schulte-Mecklenbeck, M., Murphy, R. O., & Hertwig, R. (2018). Prospect theory reflects selective allocation of attention. Journal of Experimental Psychology: General, 147(2), 147–169.

Radke, S., Roelofs, K., & De Bruijn, E. R. A. (2013). Acting on Anger: Social Anxiety Modulates Approach-Avoidance Tendencies After Oxytocin Administration. Psychological Science, 24(8), 1573–1578.

Saez, I., Zhu, L., Set, E., Kayser, A., & Hsu, M. (2015). Dopamine modulates egalitarian behavior in humans. Current Biology, 25(7), 912–919.

Teoh, Y. Y., Yao, Z., Cunningham, W. A., & Hutcherson, C. A. (2020). Attentional priorities drive effects of time pressure on altruistic choice. Nature Communications, 11(1), 3534.

Tom, S. M., Fox, C. R., Trepel, C., & Poldrack, R. A. (2007). The neural basis of loss aversion in decision-making under risk. Science, 315(5811), 515–518.

Usher, M., & McClelland, J. L. (2004). Loss Aversion and Inhibition in Dynamical Models of Multialternative Choice. Psychological Review, 111(3), 757–769.

Volz, L. J., Welborn, B. L., Gobel, M. S., Gazzaniga, M. S., & Grafton, S. T. (2017). Harm to self outweighs benefit to others in moral decision making. Proceedings of the National Academy of Sciences of the United States of America, 114(30), 7963–7968.

Wu, Q., Mao, J., & Li, J. (2020). Oxytocin alters the effect of payoff but not base rate in emotion perception. Psychoneuroendocrinology, 114, 104608.

Wu, S., Cai, W., & Jin, S. (2018). Gain or non-loss: The message matching effect of regulatory focus on moral judgements of other-orientation lies. International Journal of Psychology, 53(3), 223-227.

Xiong, W., Gao, X., He, Z., Yu, H., Liu, H., & Zhou, X. (2020). Affective evaluation of others’ altruistic decisions under risk and ambiguity. Neuroimage, 218, 116996.

Yechiam, E., & Hochman, G. (2013). Losses as modulators of attention: Review and analysis of the unique effects of losses over gains. Psychological Bulletin, 139(2), 497–518.

Zhan, Y., Xiao, X., Tan, Q., Li, J., Fan, W., Chen, J., & Zhong, Y. (2020). Neural correlations of the influence of self-relevance on moral decision-making involving a trade-off between harm and reward. Psychophysiology, 57(9), e13590.

Zhang, H., Gross, J., De Dreu, C., & Ma, Y. (2019). Oxytocin promotes coordinated out-group attack during intergroup conflict in humans. eLife, 8, e40698.

Author response:

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

Public Reviews:

Reviewer 1 (Public Review):

Summary:

The authors present a mean-field model that describes the interplay between (protein) aggregation and phase separation. Different classes of interaction complexity and aggregate dimensionality are considered, both in calculations concerning (equilibrium) phase behavior and kinetics of assembly formation.

Strengths:

The present work is, although purely theoretical, of high interest to understanding biological processes that occur as a result of a coupling between protein aggregation and phase separation. Of course, such processes are abundant, in the living cell as well as in in-vitro experiments. I appreciate the consideration of aggregates with various dimensionality, as well as the categorization into different ”interaction classes”, together with the mentioning of experimental observations from biology. The model is convincing and underlines the complexity associated with the distribution of proteins across phases and aggregates in the living cell.

Weaknesses:

There are a few minor weaknesses.

Reviewer 2 (Public Review):

This work deals with a very difficult physical problem: relating the assembly of building blocks on a molecular scale to the appearance of large, macroscopic assemblies. This problem is particularly difficult to treat, because of the large number of units involved, and of the complex way in which these units-monomers-interact with each other and with the solvent. In order to make the problem treatable, the authors recur to a number of approximations: Among these, there is the assumption that the system is spatially homogeneous, i.e., its features are the same in all regions of space. In particular, the homogeneity assumption may not hold in biologically relevant systems such as cells, where the behavior close to the cell membrane may strongly differ from the one in the bulk. As a result, this hypothesis calls for a cautious consideration and interpretation of the results of this work. Another notable simplification introduced by the authors is the assumption that the system can only follow two possible behaviors: In the first, each monomer interacts equally with the solvent; no matter the size of the cluster of which it is part. In the second case, monomers in the bulk of a cluster and monomers at the assembly boundary interact with the solvent in a different way. These two cases are considered not only because they simplify the problem, but also because they are inspired by biologically relevant proteins.

With these simplifications, the authors trace the phase diagram of the system, characterizing its phases for different fractions of the volume occupied by the monomers and solvent, and for different values of the temperature. The results qualitatively reproduce some features observed in recent experiments, such as an anomalous distribution of cluster sizes below the system saturation threshold, and the gelation of condensed phases above such threshold.

Reviewer 3 (Public Review):

Summary:

The authors combine classical theories of phase separation and self-assembly to establish a framework for explaining the coupling between the two phenomena in the context of protein assemblies and condensates. By starting from a mean-field free energy for monomers and assemblies immersed in solvent and imposing conditions of equilibrium, the authors derive phase diagrams indicating how assemblies partition into different condensed phases as temperature and the total volume fraction of proteins are varied. They find that phase separation can promote assembly within the protein-rich phase, providing a potential mechanism for spatial control of assembly. They extend their theory to account for the possibility of gelation. They also create a theory for the kinetics of self-assembly within phase separated systems, predicting how assembly size distributions change with time within the different phases as well as how the volumes of the different phases change with time.

Strengths:

The theoretical framework that the authors present is an interesting marriage of classic theories of phase separation and self-assembly. Its simplicity should make it a powerful general tool for understanding the thermodynamics of assembly coupled to phase separation, and it should provide a useful framework for analyzing experiments on assembly within biomolecular condensates.

The key advance over previous work is that the authors now account for how self-assembly can change the boundaries of the phase diagram.

A second interesting point is the explicit theoretical consideration for the possibility that gelation (i.e. self-assembly into a macroscopic aggregate) could account for widely observed solidification of condensates. While this concept has been broadly discussed, to date I have yet to see a rigorous theoretical analysis of the possibility.

The kinetic theory in sections 5 and 6 is also interesting as it extends on previous work by considering the kinetics of phase separation as well as those of self-assembly.

Weaknesses:

A key point the authors make about their theory is that it allows, as opposed to previous research, to study non-dilute limits. It is true that they consider gelation when the 3D assemblies become macroscopic. However, dilute solution theory assumptions seem to be embedded in many aspects of their theory, and it is not always clear where else the non-dilute limits are considered. Is it in the inter-species interaction χij? Why then do they never explore cases for which χij is nonzero in their analysis?

We explicitly consider that monomers and aggregates are non-dilute with respect to solvent. This is evident in accounting for the mixing entropy of all components, including the solvent. Moreover, we account for interactions among the monomers and the different aggregates with the solvent. We consider the case where each monomeric unit, independent in aggregate it is part of, interacts the same way with the solvent. Please note that this case corresponds to a non-dilute scenario where interactions indeed drive phase separation.

The connection between this theory and biological systems is described in the introduction but lost along the main text. It would be very helpful to point out, for instance, that the presence of phase separation might induce aggregation of proteins. This point is described formally at the end of Section 3, but a more qualitative connection to biological systems would be very useful here.

We thank the referee for the useful comment, we now mention this in the introduction (line 80) and point out the biological relevance of assembly formation and localization via the presence of phase separation (lines 268 and 283).

Building on the previous point, it would be helpful to give an intuitive sense of where the equations derived in the Appendices and presented in the main text come from and to spell out clear physical interpretations of the results. For example, it would be helpful to point out that Eq. 4 is a form of the law of mass action, familiar from introductory chemistry. It would be useful to better explain how the current work extends on existing previous work from these authors as well as others. Along these lines, closely related work by W. Jacobs and B. Rogers [O. Hedge et al. 2023, https://arxiv.org/abs/2301.06134; T. Li et al. 2023, https://arxiv.org/abs/2306.13198] should be cited in the introduction. The results discussed in the first paragraph of Section 3 on assembly size distributions in a homogeneous system are well-known from classic theories of self-assembly. This should be acknowledged and appropriate references should be added; see for instance, Rev. Mod. Phys. 93, 025008 and Statistical Thermodynamics Of Surfaces, Interfaces, And Membranes by Sam Safran. Equation 14 for the kinetic of volume fractions is given with reference to Bauermann et al. 2022, but it should be accompanied by a better intuitive interpretation of its terms in the main text. In particular, how should one understand the third term in this equation? Why does the change in volume impact the change of volume fraction in this way?

We thank the referee for the suggestions. We have included the missing references, with a particular emphasis on DNA nanostars that inhibit phase separation in DNA liquids in the definition of class II. We added intuitive explanations of the main equations, such as Eqs. (4),(8),(14), (17), and (18). Notice that, according to Mysels, Karol J., J. Chem. Educ., 33, 178 (1956) (https://pubs-acs-org.sire.ub.edu/doi/epdf/10.1021/ed033p178) we refer to (18) as the law of mass action.

The discussion in the last paragraph of Section 6 should be clarified. How can the total amount of protein in both phases decrease? This would necessarily violate either mass or volume conservation. Also, the discussion of why the volume is non-monotonic in time is not clear.

A decrease in the total amount of protein in both phases does not violate mass conservation, if the volume of the phases varies accordingly. In particular, the volume of the denser phase should grow. This given, in the case presented the total protein amount in the dense phase decreases, while in the dilute phase increases. For this reason, we revised the paragraph and now explain the results in more detail (see lines starting from 407). The nonmonotonic volume change is indeed a puzzling finding that, as we now state in the manuscript, requires further investigation. Given the lack of analytical approaches available to tackle the complex kinetics in the presence of coexisting phases, we believe that this analysis goes beyond the scope of the present paper.

Recommendations for the authors

Reviewer 1 (Recommendations For The Authors):

Line 96: I feel a mentioning/definition/explanation and perhaps some discussion on the parameter M (limiting aggregate size) would have been in place in the introduction of Equation (1). Furthermore, in the usual interpretation, Flory interaction parameters (symbolized χ) are dimensionless, as, classically, they represent an exchange energy (normalized by kT), defined on a monomeric basis. Here they seem to carry the dimension of energy.

We thank the reviewer for the observation. We have included a brief comment on M and mentioned that we use χ parameters that carry the dimension of energy such that, varying kBT, we scale at the same time the term containing interaction propensities (χ) and the one containing internal energies (_e_int). See the comment on line 127

Line 150: The choice of ρi \= i physically implies that a single protein is assumed to have the same as a solvent molecule. This may be a bit of a stretch. This assumption leads to an overestimation of the translational entropy of the aggregates (first term in Equation (1)). Acknowledging that ρ_1 >> ρs_ would give a pronounced desymmetrization of the phase diagram (I suspect).

Indeed, in the case of monomers only, the assumption leads to a symmetric phase diagram which may be unrealistic. Once assemblies form, however, the phase diagram becomes asymmetric and for this reason we decided to assume ρi \= i, simplifying the theoretical analysis. We have added a clarifying sentence in the manuscript, see line 163

Furthermore, the pictures in Figure 1a-c suggest the presence of a disordered residue, the degree of swelling of which might affect binding strength (see for instance: https://doi.org/10.3389/fnmol.2022.962526).

We added a comment on the possible coupling between internal free energies and interaction propensities, such as the swelling mechanism that affects binding sites, and included the reference above (line 215).

Line 154-156: It’s unclear what is meant with ”an internal bond that keeps each assembly together”. How should this be interpreted on an intuitive physical level?

We apologise for being unclear. We meant the internal bonds that lead to the formation of assemblies. We have now rephrased this sentence in the main text (lines starting from 169).

Line 254: The fact that ϕsg is defined below does not mean it does not fall out of the air here. The same holds for the consideration of the limit M →∞. Ideally, the main text should stand on its own, in particular with respect to physical intuitiveness, as well as the necessity and interest of discussion topics. Technical details, derivations and additional information can be in an appendix.

We agree with the referee and added some physical insights about the limit. We now also state clearly in the main text (line 298) that _ϕ_sg is affected by temperature and the free energy of internal bonds.

Line 257: ”Since we do not explicitly include the solvent in assembly formation we will consider the gel as a phase without solvent and thus ϕtot \= 1”. I’m not sure if I can agree with this. I would say, a gel, certainly in biological context, almost per definition contains a large fraction of solvent, i.e. here water. The situation ”ϕtot \= 1” would rather be a solid precipitate. Is gelation properly captured by this model?

We thank the referee for this very relevant observation. We now state in the main text that the model predicts a macroscopic assembly which we call ’the gel phase’, in agreement with previous literature. Then, to clarify, we added the sentence ”Please note that, since we do not explicitly include the solvent in assembly formation (see reaction scheme in Fig.1a), in our model the gel corresponds to a phase without solvent, _ϕ_tot \= 1. To account for biological gels that can be rich in water, our theory can be straightforwardly extended by incorporating the solvent into the reaction scheme.”, see main text line 300.

Line 268: Shouldn’t ”solvent” be ”solution”? If fsol is given by Equation (1), surely not only the solvent is considered.

Indeed, this is a typo, and we now use the term ’solution’ instead of ’solvent.’

Line 273: At this stage, the only information provided in the main text is that ω∞ is ”a constant that does not affect chemical nor phase equilibrium, except in the limit M →∞” (see lines 153-154). This is a little bit too abstract for me. Again, the main text should stand on its own, meaning the reader should not have to rely on an appendix to at least have an intuitive physical understanding of any modeling or input parameter discussed in the main text.

We thank the reviewer for pointing this out. We now comment on the physical interpretation of ω∞ in the main text, see lines from 320 on.

Figure 4. appears in Equation (39) but it is not defined.

We thank the reviewer for pointing this out. We have reshaped appendix 6A, making use of chemical activities and clarified the origin of the rate .

Line 317. I don’t fully understand the intention of the remark on the model being adaptable for ”primary and secondary nucleation”. How/in what way is this different from association and dissociation? For instance, classical nucleation theory is based on association and dissociation of monomeric units to and from clusters.

We agree that the kinetic rate coefficients kij (appearing in the association and dissociation rates ∆rij, Eq. 17) in our manuscript already depend on assembly length, see Appendix 6 B, where we now clarified their definition. Please note that, however, that secondary nucleation is a special kind of association, for which the kinetic rate coefficients corresponding to associations of small assemblies, i.e. kij with_i,j_ ≪ M, explicitly depend on the presence of large assemblies with sizes l ≫ 1. In our manuscript, we have not accounted for such a dependence. We now make this aspect clear in the manuscript, see Appendix 6 B.

Line 321. Why is ∆rij called the ”monomer exchange rate”? In line 318 the same parameter is defined as the ”reaction rate for the formation of a (i+j)-mer”. Why should these be the same?

We thank the reviewer for spotting this typo.

Line 323. Why do these calculations use M = 15?

The exploration of a 15-dimensional phase space is already numerically challenging. We are currently working on a generalization of the numerical scheme to work with larger values of M but, to discuss the fundamental physical principles, we kept M \= 15.

Reviewer 2 (Recommendations For The Authors):

The manuscript presents several issues, on both the scientific and presentational level, which need to be carefully addressed. Please find below a list of the points that need to be addressed by the authors, divided into major and minor points. Major issues:

• A general, major concern about the results in the paper is the homogeneity assumption. I do understand that repeating the whole analysis presented in the manuscript by allowing for spatial inhomogeneities partially goes beyond the scope of this paper. However, the authors should at least discuss how such inhomogeneities may alter the results in a qualitative way, and treat explicitly the presence of inhomogeneity in one prototypical case treated in the manuscript. Namely, what happens if the volume fractions and relative molecular volumes in the free energy (1) depend on space, e.g., ϕi → ϕi(x)?

We would like to stress that, in the present paper, we do account for spatial inhomogeneities. Indeed, in the case of phase separation, we consider systems which are divided into two phases, characterized by different values of the assemblies’ volume fractions ϕi. We do, however, consider the system to be homogeneous inside the phases, implying a jump in the value of the volume fraction at the interface between the two phases. In this sense, the analysis we carry out is valid in the thermodynamic limit, where gradients of the volume fractions ϕi(x) within the phases, can be neglected. On the other hand, considering the full spatial problem, i.e. solving the equations for M \= 15 spatially varying fields, would be numerically extremely challenging.

• The authors’ results relate molecular assembly- a phenomenon at the molecular scale-to phase separation-a mesoscopic or macroscopic phenomenon. The authors should stress the conceptual importance of this connection between scales, and present their results from the perspective of a multi-scale model.

We thank the reviewer for pointing this out. We now emphasize the multi-scale feature of our model in the introduction (line 80).

• Starting from Section 1, the reader is not well guided through the sections that follow. The authors should provide an outline of the line of though that they are going to follow in the following sections, and logically connect each section to the next one with a short paragraph at the end of each section. This paragraph should resume what has been addressed in the current section, and the connection with the topic that will be addressed in the next one.

We agree with the reviewer and have added a transitioning sentence at the end of each paragraph.

• ’We focus on linear assemblies (d = 1)’: Given the striking differences of the results between d = 1 and d > 1 shown above, the authors should discuss what happens for d > 1 as well.

• ’In figure Fig. 5a, we show the initial and final equilibrium binodals (black and coloured curve, respectively), for the case of linear assemblies (d = 1) belonging to class 1’: Again, show what happens for d > 1.

We agree with the reviewer, the kinetics in d > 1 would be definitely interesting. However, in this case, one assembly can become macroscopic (i.e. M must be set to ∞). This requires some substantial modification in the kinetic scheme, like introducing an absorbing boundary condition for monomers ’sucked in’ the gel. We prefer to leave this for future work, and now state it explicitly in the manuscript (line 383).

• ’This difference arises because, within class 2, monomers in the bulk of an assembly have reduced interaction propensity with respect to the boundary ones. As a consequence, the formation of large clusters shifts the onset of phase separation to higher ϕtot values.’: To prove this argument, the authors should show Fig. 2g and h for d > 1. In fact, by varying d, the effect of the boundary vs. bulk also varies.

We prefer to discuss the thermodynamics of d > 1 in section 4 on gelation. There we present only a single phase diagram so as not to blow up the discussion on equilibrium too much.

• ’referring for simplicity to systems belonging to Class 1’: The authors should do the same analysis for Class 2.

We agree with the reviewer. However, again not to blow up the discussion on equilibrium, we leave it for future work.

• ’other, implying that the corresponding Flory-Huggins parameter χij vanishes’: Why?

The explanation based on a lattice model is reported in Appendix 2, and is now more clearly referenced (line 185).

Minor issues:

• Eq. (10): Here the authors should explain in the main text, possibly in a simple and intuitive way, why the number of monomers i and the space dimension d enter the righthand side of this equation in this particular way.

We thank the reviewer for pointing this out. We added the physical origin of the scaling with dimension in Eq. (10) and in Eq. (8), as pointed out by reviewer 3.

• ’The second and fifth terms of fsol characterize the internal free energies’: What do you mean by ’characterize the internal free energies’? Please clarify.

As we now state more clearly (lines 114-120), these two contributions include the internal free energies ω_s and _ωi, stemming from the free energy of internal bonds that lead to assembly formation.

• ’depend on the scaling form of the’: Scaling with respect to what ? Please clarify.

We have now clarified that the scaling is with respect to the assembly size i.

• Figure 2 is way too dense: it should be split into two figures, and the legend of each of the two figures should be expanded to properly guide the reader to understand the figures.

We understand the reviewer’s point of view. To avoid altering the present flow, we decided not to split the figure, but we have included shaded boxes to better guide the reader.

• ’this is a consequence of the gelation transition’: Please clarify

• ’and this limitation can be dealt with by introducing explicitly the infinite-sized gel in the free energy’: Why? Please clarify.

We have now rephrased these sentences, hopefully in a clearer way. We now state: ’We know that this divergence is physical, and is caused by the gelation transition. This limitation can be dealt with by introducing explicitly a term in the free energy that accounts for an infinite-sized assembly (the gel)’, see lines 320-322.

• Figure 4: Add plots of panels d, e, h and i with log scale on the y axis to make explicit an eventual exponential behavior, and revise the text accordingly

Not to further complicate Figure 4, we preferred to display the logarithmic plots of the equilibrium distribution in the appendix, see Figure A3-1.

• ’... an equilibrium distribution which monotonously decreases with assembly size’: It is not the distributions that decreases but the cluster volume fraction, please rephrase.

We thank the reviewer for pointing this out and have now rephrased this sentence (line 394).

Reviewer 3 (Recommendations For The Authors):

I could not obtain the exact form of Eq 29 in App 3, can the authors elaborate on this calculation. App 3: What does it mean binodal agrees well with ϕsg? And doesn’t ϕsg depend on temperature through phi tilde? What temperature is this result for?

We apologise for the unclear explanation. We now state in detail that Eq. (29) is obtained by plugging the expression of ϕi given in Eq. (24) into Eq. (1), in the main text. The dependence of ϕ₁ on ϕ_tot is expressed in Eq. (26), and we have omitted linear terms in ϕ_tot, since they do not affect phase equilibrium (see lines 802-809). Moreover, ϕsg depends indeed on k_BT. We refer to the comparison between the full curve ϕsg in the k_BT−ϕ_tot plane, and the branch of the binodal between the triple point (indicated now with a cross) and ϕ_tot \= 1. The two curves are close, as expected since both correspond to the boundary between homogeneous mixtures and the gel state, obtained with different methods.

The references to Figures in the appendices are confusing. Please make it clear whether Figures in the main text or the appendices are being referenced. On a related note, the Appendix figures seem to be placed in appendices whose text describes something else - Appendix 2, Figure 1 should be moved to Appendix 3; Appendix 3, Figure 1 should be moved to Appendix 4; etc.

We revised the appendix, corrected the figure positions and clarified their references.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

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

Strengths:

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

We thank the reviewer for summarizing the importance and significance of our study. 

Weaknesses:

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

We have further improved our description of the methods in the revised reviewed preprint, and here in the response letter, we respond point-by-point to ‘Reviewer #1 (Recommendations For The Authors)’ below. 

Reviewer #2 (Public Review):

Summary:

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

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

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

Strengths:

Understanding how orexin interacts with the dopamine system is an important question and this paper contains several novel findings along these lines. Specifically:

(1) The distribution of orexin receptor subtypes in VTA and SN is explored thoroughly.

(2) Use of the genetic model that knocks out a specific orexin receptor subtype from only dopamine neurons is a useful model and helps to narrow down the behavioral significance of this interaction.

(3) PET studies showing how central administration of orexin evokes dopamine release across the brain is intriguing, especially since two key areas are pursued - BNST and LPGi - where the dopamine projection is not as well described/understood.

We thank the reviewer for the careful summary and highlighting the novelty of our study.

Weaknesses:

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

We agree that focusing on some orexin-dopamine targeting areas, such as dBNST or LPGi, is important to further reveal the anatomy-behavior links and underlying mechanisms. While we are very interested in further investigations, in the present manuscript we mainly aim to give an overview of the behavioral roles of orexin-dopamine interaction and to propose some promising downstream pathways in a relatively broad and systematical way. 

We have explained the physiological meanings of our results in more detail in the discussion in the revised reviewed preprint (lines 282-293, 318-332, ). Novelty-induced behavioral response should be at proper levels under normal physiological conditions. The orexin-dopamine interaction blunting novelty-induced locomotion could be important to keep attention on the main task without being distracted too much by other random stimuli in the environment. When this balance is disrupted, behavioral deficit may happen, such as attention deficit and hyperactivity disorder (ADHD).  

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

For the reward-related behaviors described in this study, we did not find significant changes between [mutant] and control mice. We agree that it will be helpful for readers by describing the behavioral tests in more details. In the revised reviewed preprint, we have described in more detail in the results and Materials and Methods section how the control and [mutant] mice behave to the reward (lines 162-165, 171-181, 526-528).  

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

We described the CPP analysis in the Materials and Methods section (lines 523-528 ) as below: ‘The percentage of time spent in the reward-paired compartment was calculated: 100 x time spent in the compartment / (total time - time spent in the middle area). The CPP score was then analyzed using the calculated percentage of time: 100 x (time on the test day – time on pre-test days)/ time on pre-test days. The pre-test and test days were before and after the conditioning, respectively. Thus, the CPP score above zero indicates that the CPP preference has developed.’ In Figure 2—figure supplement 4 C and F, it was shown that most control and knockout mice had a CPP score above zero. The control and knockout groups both developed a preference and there was no significant difference between the groups. 

For the sucrose/sucralose preference tests, in Figure 2—figure supplement 4 A and D, we present values as the percentages of sucrose/sucralose consumption in total daily drinking amount (sucrose/sucralose solution + water). Thus, percentages above 50% indicates mice prefer sucrose/sucralose to water. As shown in the figure, male mice only showed weak preference of 0.5% sucrose, compared to water, and under all other tested conditions, the mice showed strong preference of the sweet solution. There was no significant difference between control and knockout mice. 

We have described this in more details in the Results and Materials and Methods section in the revised reviewed preprint. 

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

(1) Figure 1, A-I. It is difficult to depict the anatomical subdivision of VTA in Figure 1, panels A and B. It is recommended to add a panel showing a schematic illustration of the SNc and subregions of VTA: PN, PIF, PBP, IF (providing more detail than in Figure 1, panel J). It is also recommended to show lower magnification images (as in Figure 1 - supplement 1), including both hemispheres, and to delineate the outline of the different subregions using curved lines, based on reference atlases (similar to Figure 1, panel I, please include distance from bregma). It would be helpful to indicate in Figure 1 that panel A is a control mouse and panel B is a Ox1RΔDAT mouse and include C-F letters to show corresponding insets. Anatomically, the paraintrafasicular nucleus (PIF) is positioned between the paranigral nucleus (PN) and the parabrachial pigmented nucleus (PBP). The authors have depicted the PIF ventral to the PN in Figure 1 panels A, B, and I. These panels and the quantification of Ox1R/2R positive cells within the different subdivisions need to be corrected accordingly. The image analysis method used to quantify RNAscope fluorescent images is not described in sufficient detail. Please expand this section.

According to the reviewer’s suggestions, we have refined Figure 1 in the revised reviewed preprint. We are now showing the schematic illustration of the SN and subregions of VTA in panel I, with blue squares to label the regions shown in panels A and B, and the distance from bregma is included. The outlines to delineate SN and the subregions of VTA are adjusted from straight to curved lines based on reference atlases. As suggested, we have also indicated panel A is a control and panel B is a Ox1RΔDAT mouse and included C-F letters to show corresponding insets. We apologize for the mistake about labeling PIF and PN positions in Figure A. We have corrected the labeling of their positions and double checked the quantification accordingly. This does not change our discussion or conclusion since both PIF and PN are the medial part of VTA, where both Ox1R and Ox2R are observed. The description of the image analysis in Matierials and Methods section has been improved (lines 378-385). We decided not to show lower magnification images than in Figure 1—supplement 1 to include both hemispheres, in the interests of clarity and reader-friendliness.  

(2) Figure 1, J-L. The claim that orexin activates dopaminergic SN and VTA neurons is weakly supported by the data provided. Calcium imaging of SN dopaminergic neurons in control mice suggests a discrete effect of 100 nM orexin-A application compared to baseline. Application of 300 nM shows a slightly bigger effect, but none of these results are statistically analysed. 

We are surprised by this comment and thank the reviewer for pointing out our apparent lack of clarity in the previous version (lines 96-106 and legend of Figure 1K, L). In more detail, we explain the data analysis in the new version (lines 119-133, 451-465) and the legend of Figure 1K, L and Figure 1-figure supplement 3).

The main goal of this part of the project was to functionally validate the Ox1R knockout in dopaminergic (DAT-expressing) neurons. This was a prerequisite for the behavioral and PET imaging experiments. We used GCaMP-mediated Ca2+ imaging in acute brain slices to reach this goal. This analysis was performed on the dopaminergic SN neurons, which we used as an "indicator population" because a large number of these neurons express Ox1R, but only a few express Ox2R. 

The analysis consisted of two parts:

a) For each neuron, we tested whether it responded to orexin A. At the single cell level, a neuron was considered orexin A-responsive if the change in fluorescence induced by orexin A was three times larger than the standard deviation (3 σ criterion) of the baseline fluorescence, corresponding to a Zscore of 3. We found that 56% of the neurons tested responded to orexin A, while 44% of the neurons did not respond to orexin A (Figure 1L, top). These data agree with the number of Ox1R-expressing neurons (Figure 1J). 

b) We also determined the orexin A-induced GCaMP fluorescence for each neuron, expressed as a percentage of GCaMP fluorescence induced upon application of high K+ saline. Accordingly, the "population response" of all analyzed neurons was expressed as the mean ± SEM of these responses. The significance of this mean response was tested for each group (control and Ox1R KO) using a onesample t-test. We found a marked and highly significant (p < 0.0001, n = 71) response of control neurons to 100 nM orexin A, while the Ox1R KO neurons did not respond (p = 0.5, n = 86). Note that, as described in a), 44% of the neurons contributing to the mean do not respond to orexin. Thus, the orexin responses of most responders are significantly higher than the mean. This is also evident in the example recordings in Figure 1K and Figure1—figure supplement 3. The orexin A-induced change in fluorescence was increased by increasing the orexin A concentration to 300 nM.

Note: As mentioned above, the orexin A response was expressed for each neuron individually as a percentage of its high K+saline-induced GCaMP fluorescence. This value is a solid reference point, reflecting the GCaMP fluorescence at maximal voltage-activated Ca2+ influx. Obviously, the Ca2+ concentration at this point is extremely high and not typically reached under physiological conditions. Therefore, as shown in Figure1—figure supplement 3 for completeness, the physiologically relevant responses may appear relatively minor at first glance when presented together in one figure (compare Figure1—figure supplement 3 A and B).

The authors should provide more evidence of the orexin-induced activation of dopaminergic neurons in the SN to support this claim and investigate whether a similar activation is observed in VTA neurons. 

Following the reviewer's suggestion, we confirmed orexin A-induced activation of dopaminergic neurons in the mouse SN by using perforated patch clamp recordings (Figure1—figure supplement 2).

This finding is consistent with previous extracellular in vivo recordings in rats (Liu et al., 2018).

The activation of dopaminergic neurons in the mouse VTA by orexin A has been shown repeatedly in earlier studies (e.g., Baimel et al., 2017; Korotkova et al., 2003; Tung et al., 2016).

In addition, Figure 3-Figure Supplement 2 shows that injection of orexin does not induce c-Fos expression in SN and VTA dopaminergic neurons of control and Ox1RΔDAT mice, which further weakens the claim made by the authors.

Figure 3—Figure Supplement 2 in the original submission is now Figure 3—Figure Supplement 3 in the revised reviewed preprint. It shows low c-Fos expression in SN and VTA dopaminergic neurons, and orexin-induced c-Fos expression was observed in Th-negative cells in SN and VTA. 

Technically relatively straightforward, Fos analysis is widely (and successfully) used in studies to reveal neuronal activation. However, this approach has limitations, e.g., regarding sensitivity and temporal resolution. Electrophysiological or optical imaging techniques can circumvent these shortcomings. The electrophysiological and Ca2+ imaging studies presented here, along with previous electrophysiological studies by others, clearly show that orexin A acutely and directly stimulates SN and VTA dopaminergic neurons.

In vivo, the injection of orexin A induced a pronounced c-Fos activity in non-dopaminergic cells of the VTA and SN but not in dopaminergic neurons. This result shows that the detection of c-Fos has worked in principle. Whether the absent c-Fos staining in dopaminergic neurons is due to lack of sensitivity, whether other IEGs would have worked better here, or whether there are other, e.g., cell type-specific reasons for the absence of staining, cannot be determined from the current data.

(3) Figure 2, I-L. The fact that ICV injection of both saline and orexin causes a sustained increase of locomotion (around 20 minutes in males, and over 30 minutes in females) is problematic and could mask the effects of orexin, particularly in females. It is unclear what panels J and L are showing. To be appropriately analysed, the authors should plot the pre- and post-injection AUC data for all groups and analyse it as a two-way mixed ANOVA, with the within-subjects factor "pre/post injection activity" and between-subjects factor "group". The authors can only warrant a statistically meaningful hyperlocomotor effect in Ox1RΔDAT mice if a significant interaction is found.

Though mice were habituated to the injection, it still makes sense to see the injection-induced increase in locomotion to some extent. We described in the figure legend that the AUC was calculated for the period after orexin injection, which meant 5 – 90 min in Figure 2 I, K. We have clearly observed significant differences between genotypes and between saline and orexin application, which means the genotype and orexin impact is strong enough to pop up despite of the injection effect. 

As the reviewer’s suggests, we have now plotted the pre- and post-injection AUC data for all groups and analyzed it as a two-way mixed ANOVA, with the within-subjects factor "pre/post injection activity" and between-subjects factor "group". To match the pre- and post-injection duration, we are now comparing AUC for around 60 min before and after the injection. A significant interaction is found here. Panels I-L are renewed, and the differences induced by Ox1R knockout and orexin confirmed the results shown in the initially submitted manuscript.  

(4) Figure 3. The literature has robustly shown that one of the main projection areas of VTA and SN dopaminergic neurons is the striatum, in particular its ventral part. It is surprising to see that this region is not affected by the lack of OX1R or by the injection of orexin. How can the authors explain that identified regions with significantly different activity include neighbouring brain structures with heterogenous composition? See for example, in panel A, section bregma 0.62mm, a significant region is seen expanding across the cortex, corpus callosum, and striatum. While the data from PET studies is potentially interesting, it may not be adequate to provide enough resolution to allow examination of the anatomical distribution of orexin-mediated neuronal activation.

While the striatum is a major projection area of dopaminergic neurons in VTA and SN, the projection and function of Ox1R-positive dopaminergic neurons is not clear. We have improved the description of dopamine function diversity in the revised reviewed preprint (lines 46-58), and it was reported before that the projection-defined dopaminergic populations in the VTA exhibited different responses to orexin A (Baimel et al., 2017). Moreover, the striatum activity is modulated by the indirect effect via other brain regions affected by Ox1R-positive dopaminergic neurons. It is unknown how the striatum activity should change after Ox1R deletion in dopaminergic neurons. We could not rule out the possibility that the striatum is indeed modulated by the Ox1R-positive dopaminergic neurons, though there was only a trend of genotype difference (Ox1RΔDAT vs. ctrl) in the ventral striatum in the section bregma 1.42 mm in Figure 3A. The ICV injection of orexin is potentially acting on Ox1R and Ox2R in the whole brain, so projections from other brain regions to the striatum also affect striatum activity and could have masked the effect of Ox1R-positive dopaminergic neurons. 

The spatial resolution of the PET data is in the order of ~1 mm^3. As we also explained in the Materials and Methods section, the size of a voxel in the original PET data is 0.4mm x 0.4mm x 0.8 mm. All calculations were performed on this grid. The higher-resolved images shown in Figure 3 are for presentation purposes only inspired by a request of the reviewer who asked us to show this in the Jais et al. 2016 manuscript. To make this clearer we now added the p-map images with the original voxel size to the supplement (Figure 3—figure supplement 1). For the interest in specific brain areas, more precise identification of anatomical sub-regions requires using methods with higher spatial resolution such as staining of brain slices for c-Fos-positive cells as we do in Figure 4.

PET is a powerful tool to identify global regions of activation/inhibition. In the manuscript, we have described in the results and discussion section that the activity in brain regions with related functions were changed. In panel A, Ox1RΔDAT showed activity increase in MPA, Pir and endopiriform claustrum, which are important for olfactory sensation; spinal trigeminal nucleus, sp5, and IRt, which regulates mastication and sensation of the oral cavity and the surface of the face; SubCV and Gi, which regulates sleeping and motion-related arousal and motivation. In panel B, changes in HDB, MCPO, Pir, DEn, S1, V2L and V1 are related to sensation, and changes in BNST, LPGi and M2 are important for emotion, exploration, and action selection. 

(5) Figure 4. As in Figure 1, the authors should consider including a schematic illustration of the brain areas that are being analysed using a reference atlas. It is also recommended to provide more details describing the quantification of the images. Without such information, the data is not convincing, in particular, the claim that Ox1R depletion causes a decrease in DRD1 in BNST is unclear. Additional unbiased quantitative approaches could be used to strengthen this point.

We have added Figure 4—figure supplement 1 as a schematic illustration of the brain areas that were being analyzed using a reference atlas. More details describing the unbiased quantification of the images have been added to Materials and Methods. We have added Figure 4—figure supplement 3, to show DRD1, DRD2 and the merged signal separately.  

(6) The discussion starts by stating that the main findings of this study are based on RNAscope and optophysiological experiments, however, the latter are not presented anywhere in the manuscript. This sentence (line 192) should be revised. The authors state in line 193 that OX1R is the only orexin receptor in the SN, but they show in Figure 1 that in the SN, 3% of neurons express OX2R and 2% co-express both receptors. 

We thank the reviewer for the input. We have rephrased the beginning of the discussion to clarify the objectives (lines 238 - 246). In doing so, we changed "optophysiological experiments" and "single orexin receptor" (lines 192 and 193 in the original manuscript) to " Ca2+ imaging experiments" and "main subtype of orexin receptors ", respectively. In this context, it should be noted that Ca2+ imaging is considered an optophysiological method - optophysiology generally refers to techniques that combine optical methods with physiological measurements.

The results of LPGi and BNST dopamine receptors in control and Ox1RΔDAT mice are poorly discussed. The authors should justify why these two regions were selected for further validation and how these may be related to the behavioural effects found in Ox1RΔDAT regarding exposure to a novel context.

Ox1RΔDAT mice exhibited increased novelty- and orexin-induced locomotion compared to control mice. After orexin injection, PET imaging shows that the neural activity of BNST and LPGi was lower or higher than in control mice, respectively. We selected BNST and LPGi for further validation because we think their key functional roles in regulating emotion, exploratory behaviors and locomotor speed are related to novelty-induced locomotion. We confirmed changes in neural activity change by c-Fos staining and investigated the expression patterns of dopamine receptors in BNST and LPGi. Our findings suggested that Ox1R deletion in dopaminergic neurons results in the disinhibition of neural activity in LPGi via dopaminergic pathways and the decrease of dopamine-mediated neural activity in BNST. Emotion perception affects the decision of how to respond to the novelty. It is possible that novelty activates the orexin system and Ox1R signaling in dopaminergic neurons promotes emotion perception and inhibits exploration. Of course, further careful investigation is necessary to test this hypothesis in the future experiments. We have improved the rational description and discussion in the

‘Results’ and ‘Discussion’ section in the revised reviewed preprint (lines 210-213, 259-270, 293-308). 

Reviewer #2 (Recommendations For The Authors):

A major recommendation - if possible - would be to directly show that one or both of the two target areas - dBNST and LPGi - are associated with the behavioral effects caused by the deletion of the orexin receptor 1 in dopamine neurons.

We completely agree that it would be very valuable to directly show dBNST and LPGi are associated with the behavioral effects caused by the deletion of Ox1R in dopaminergic neurons. While we are very interested in carefully investigating specific orexin-dopamine targeting areas and related neural circuits in the future, in the present manuscript, we mainly aim to give an overview of the behavioral roles of orexin-dopamine interaction and propose some promising downstream pathways. 

The authors should state if data are corrected for multiple comparisons, e.g., in the PET study of different regions.

We have included information about the post-hoc tests for all 2-way ANOVA analyses in the submitted manuscript. For the PET study, the p-values in the p-maps were not corrected for multiple comparison, Figure 3—figure supplement 2 shows the raw data of each mouse and the analysis method (t-test). In the revised reviewed preprint, we include the information on the analysis method in the figure legends of Figure 3. 

We consider that saline and orexin injections mimic the resting and active state of mice, respectively, and would like to study genotype effect under each condition. Doing 2-way ANOVA takes in count the difference between orexin and saline injection, which could mask the genotype effect under a certain condition. Therefore, we decided to perform t-tests for each condition in Figure 3. While we provide readers with full information in Figure 3—figure supplement 2 with the raw data of each individual mouse, below we present the p-maps after multiple comparisons (Sidak’s post hoc test). After multiple comparisons, we could see changes in similar brain regions as in Figure 3, though significant values are reduced by the correction for multiple comparisons, and under orexin-injection condition, we fail to see significantly higher activity around the lateral paragigantocellular nucleus (LPGi), nucleus of the horizontal limb of the diagonal band (HDB) and magnocellular preoptic nucleus (MCPO) in Ox1RΔDAT mice. In order to more precisely identify the anatomical locations, we performed additional experiments to confirm the changes revealed by PET. For example, LPGi is a relatively small region confirmed and identified more precisely by c-Fos immunostaining (Figure 4A, C). 

Author response image 1.

PET imaging studies comparing Ox1RΔDAT and control mice, with post-hoc t-test to correct for multiple comparisons. 3D maps of p-values in PET imaging studies comparing Ox1RΔDAT and control mice, after intracerebroventricular (ICV) injection of (A) saline (NS) and (B) orexin A. Control-NS, n = 8; control-orexin, n = 6; Ox1RΔDAT, n = 8. M2, secondary motor cortex; MPA, medial preoptic area; Pir, piriform cortex; IEn, intermediate endopiriform claustrum; DEn, dorsal endopiriform claustrum; VEn, ventral endopiriform claustrum; LSS, lateral stripe of the striatum; BNST, the dorsal bed nucleus of the stria terminalis; S1Sh, primary somatosensory cortex, shoulder region; S1HL, primary somatosensory cortex, hindlimb region; S1BF, primary somatosensory cortex, barrel field; S1Tr, primary somatosensory cortex, trunk region; V1, primary visual cortex; V2L, secondary visual cortex, lateral area; SubCV, subcoeruleus nucleus, ventral part; Gi, gigantocellular reticular nucleus; IRt, intermediate reticular nucleus; sp5, spinal trigeminal tract.

Provide a rationale for following up on BNST and LPGi and not any of the regions identified in the PET study.

We thank the reviewer for the careful reading and important input. Ox1RΔDAT mice exhibited increased novelty- and orexin-induced locomotion compared to control mice. After orexin injection, PET imaging shows that the neural activity of BNST and LPGi was lower or higher than control mice, respectively.

We selected BNST and LPGi for further validation because we think their key functional roles in regulating emotion, exploratory behaviors and locomotor speed are related to novelty-induced locomotion. We confirmed the neural activity change by c-Fos staining and investigated the expression patterns of dopamine receptors in BNST and LPGi. Our findings suggested that Ox1R deletion in dopaminergic neurons results in the disinhibition of neural activity in LPGi via dopaminergic pathways and the decrease of dopamine-mediated neural activity in BNST. Emotion perception affects the decision how to respond to the novelty. It is possible that novelty activates the orexin system and Ox1R signaling in dopaminergic neurons promotes emotion perception and inhibits exploration. Of course, further investigation is necessary to test this hypothesis in future. We have improved the rational description and discussion in the ‘Results’ and ‘Discussion’ section in the revised reviewed preprint (lines 210-213, 259-270, 293-308). 

Heatmap in Fig. 1K should not have smoothing across the y-axis, individual cells should be discrete.

We thank the reviewer for bringing this issue to our attention. The data had not been intentionally smoothed (neither across the x-axis nor the y-axis), but it was probably a formatting issue. We have corrected this and separated individual cell traces with lines (Figure 1K, Figure 1—figure supplement 3).

Dopamine cells are well known to lack Fos expression in most cases. Did the authors consider using another IEG to show neural activation, e.g., pERK?

We did not use another IEG. The electrophysiological and Ca2+ imaging studies presented here, along with previous electrophysiological studies by others, clearly show that orexin A acutely and directly stimulates SN and VTA dopaminergic neurons. Please see also the response to a related comment of Reviewer 1.

Consider adding a lower magnification section to anatomical figures to aid the reader in orienting and identifying the location.

We have added the schematic illustration of SN, VTA, BNST and LPGi in Figure 1I and Figure 4— figure supplement 1. We hope this helps the reader in orienting and identifying the location.  

Data availability should be stated.

There are no restrictions on data availability. We have added this section to the revised reviewed preprint.

Line 50. Some more references both historical and recent could be given to support this statement about the function of dopamine.

We have improved the description and references to support the statement about dopamine function (lines 46-58). We have cited recent studies and some reviews in the revised reviewed preprint (lines 4658). 

The PET data (Fig. 3) might be easier to visualize and interpret if a white background was used. In addition, is there a more refined way of presenting the data in Fig 3, S1?

It is common to present imaging data such as PET and MRI on a black background. We also have already applied this color scheme in multiple publications and would therefore prefer to stick to this color scheme. 

While Figure 3 is the concise way to present PET data, we aim to show the original individual results of mice in Figure 3—figure supplement 2 and to demonstrate how we performed the statistical analysis. Therefore, we take an example voxel of the respective brain area, perform the t-test, and present the data as bars with individual dots. 

Line 97. State what type of Ca imaging here, e.g., "we performed Ca imaging in ex vivo slices of VTA and SN".

As the reviewer suggested, we have specified the type of Ca2+ imaging (line 112).

Line 165. State which groups this post-mortem analysis was performed on and if any differences were to be found (not expected to find differences in this anatomical tracing experiment but good to report this as both groups were used).

Postmortem analysis of c-Fos staining revealed low c-Fos expression in dopaminergic neurons in the VTA and SN of Ox1RΔDAT and control mice after ICV injection of saline or orexin A (1 nmol). No obvious changes were observed among the groups. We have improved the description in the revised reviewed preprint (lines 202-208).

Line 192. What do you mean by optophysiological here? The Ca imaging (which is a fairly small, confirmatory element of the manuscript).

We have changed ‘optophysiological experiments’ (line 192 in initial submitted manuscript) to ‘calcium imaging experiments’ and rephrased the beginning of the discussion to clarify the objectives (lines 238246).

The protein level in the diet is substantially higher than in most rodent diets (34% here vs 14-20% in most commercial rodent chows). Please comment on this.

This diet is for rat and mouse maintenance, purchased from ssniff Spezialdiäten GmbH (product V1554).

The percentage of calories supplied by protein is affected by the calculation methods. The company calculated with pig equation before and the value was 34% in the old instruction data sheet. They have updated the value to 23% in the new data sheet with calculations by Atwater factors. We thank the reviewer for reminding us and have updated the values in the revised reviewed preprint (lines 314-316). 

Editor's note:

Should you choose to revise your manuscript, please include full statistical reporting including exact p-values wherever possible alongside the summary statistics (test statistic and df) and 95% confidence intervals. These should be reported for all key questions and not only when the p-value is less than 0.05.

We have provided the source data and the statistical reporting for each Figure with the revision

References

Baimel, C., Lau, B. K., Qiao, M., & Borgland, S. L. (2017). Projection-target-defined effects of orexin and dynorphin on VTA dopamine neurons. Cell Rep, 18(6), 1346-1355.  https://doi.org/10.1016/j.celrep.2017.01.030

Korotkova, T. M., Eriksson, K. S., Haas, H. L., & Brown, R. E. (2002). Selective excitation of GABAergic neurons in the substantia nigra of the rat by orexin/hypocretin in vitro. Regul Pept, 104(1-3), 83-89. https://doi.org/10.1016/s0167-0115(01)00323-8 

Korotkova, T. M., Sergeeva, O. A., Eriksson, K. S., Haas, H. L., & Brown, R. E. (2003). Excitation of ventral tegmental area dopaminergic and nondopaminergic neurons by orexins/hypocretins. J Neurosci, 23(1), 7-11. https://www.ncbi.nlm.nih.gov/pubmed/12514194

Liu, C., Xue, Y., Liu, M. F., Wang, Y., Liu, Z. R., Diao, H. L., & Chen, L. (2018). Orexins increase the firing activity of nigral dopaminergic neurons and participate in motor control in rats. J Neurochem, 147(3), 380-394. https://doi.org/10.1111/jnc.14568 

Tung, L. W., Lu, G. L., Lee, Y. H., Yu, L., Lee, H. J., Leishman, E., Bradshaw, H., Hwang, L. L., Hung, M. S., Mackie, K., Zimmer, A., & Chiou, L. C. (2016). Orexins contribute to restraint stress-induced cocaine relapse by endocannabinoid-mediated disinhibition of dopaminergic neurons. Nat Commun, 7, 12199. https://doi.org/10.1038/ncomms12199

Author response:

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

eLife assessment

This useful manuscript reports mechanisms behind the increase in fecundity in response to sub-lethal doses of pesticides in the crop pest, the brown plant hopper. The authors hypothesize that the pesticide works by inducing the JH titer, which through the JH signaling pathway induces egg development. Evidence for this is, however, inadequate.

We greatly appreciate your valuable comments and constructive suggestions for our work. All in all, the manuscript has been carefully edited and improved following your suggestions. We also provide more evidence to support our statements by conducting new experiments. First, we found that also EB treatment of adult females can stimulate egg-laying. Second, EB treatment in female adults increases the number of mature eggs in the ovary and ovarioles. Third, EB treatment in females enhances the expression of the kr-h1 gene in the whole body of BPH. Finally, EB treatment in female adults increases the JHIII titer, but has no impact on the 20E titer.

Public Reviews:

Reviewer #1 (Public Review):

Summary:

Gao et al. have demonstrated that the pesticide emamectin benzoate (EB) treatment of brown planthopper (BPH) leads to increased egg-laying in the insect, which is a common agricultural pest. The authors hypothesize that EB upregulates JH titer resulting in increased fecundity.

Strengths:

The finding that a class of pesticide increases the fecundity of brown planthopper is interesting.

We greatly appreciate your positive comments on our work.

Weaknesses:

(1) EB is an allosteric modulator of GluCl. That means EB physically interacts with GluCl initiating a structural change in the cannel protein. Yet the authors' central hypothesis here is about how EB can upregulate the mRNA of GluCl. I do not know whether there is any evidence that an allosteric modulator can function as a transcriptional activator for the same receptor protein. The basic premise of the paper sounds counterintuitive. This is a structural problem and should be addressed by the authors by giving sufficient evidence about such demonstrated mechanisms before.

Thank you for your question. As the reviewer points out, EB physically interacts with its target protein GluCl and thus affects its downstream signaling pathway. In the manuscript, we reported that EB-treated brown planthoppers display increased expression of GluCl in the adult stage (Fig. 5A). Actually, there are many studies showing that insects treated with insecticides can increase the expression of target genes. For example, the relative expression level of the ryanodine receptor gene of the rice stem borer, Chilo suppressalis was increased 10-fold after treatment with chlorantraniliprole, an insecticide which targets the ryanodine receptor (Peng et al., 2017). Besides this, in Drosophila, starvation (and low insulin) elevates the transcription level of the sNPF and tachykinin receptors (Ko et al., 2015; Root et al., 2011). In brown planthoppers, reduction in mRNA and protein expression of a nicotinic acetylcholine receptor α8 subunit is associated with resistance to imidacloprid (Zhang et al., 2015). RNA interference knockdown of α8 gene decreased the sensitivity of N. lugens to imidacloprid (Zhang et al., 2015). Hence, expression of receptor genes can be regulated by diverse factors including insecticide treatment. In our case, we found that EB can upregulate its target gene GluCl. However, we did not claim that EB functions as transcriptional activator for GluCl, and we still do not know why EB treatment changes the expression of GluCl in the brown planthopper. Considering our experiments are lasting several days, it might be an indirect (or secondary) effect caused by other factors, which change the expression of GluCl gene upon EB action of the channel. One reason is maybe that the allosteric interaction with GluCl by EB makes it dysfunctional and the cellular response is to upregulate the channel/receptor to compensate. We have inserted text on lines 738 - 757 to explain these possibilities.

(2) I am surprised to see a 4th instar larval application or treatment with EB results in the upregulation of JH in the adult stages. Complicating the results further is the observation that a 4th instar EB application results in an immediate decrease in JH titer. There is a high possibility that this late JH titer increase is an indirect effect.

Thank you for your question. Treatment with low doses or sublethal doses of insecticides might have a strong and complex impact on insects (Gandara et al., 2024; Gong et al., 2022; Li et al., 2023; Martelli et al., 2022). We kept the 4th instar of brown planthoppers feeding on EB for four days. They will develop to 5th instar after four days treatment, which is the final nymphal stage of BPH. Since the brown planthopper is a hemimetabolous insect, we cannot rule out the possibility that an indirect effect of treatment with EB results in the upregulation of JH in the adult stages. In this new revised manuscript, we investigated the impact of EB treatment in the adult stage. We found that female adults treated with EB also laid more eggs than controls (Figure 1-figure supplement 1A). The following experiments were performed in adults to address how EB treated stimulates egg-laying in adult brown planthopper.

(1) We found that EB treatment in adults increases the number of mature eggs in ovary (new Figure 2-figure supplement 1). We add this results in lines 234 – 238 and 281-285.

(2) We measured the JH titer after the female adults had been treated with EB. We found that EB can also increase the JH titer but has no impact on the 20E titer in the female adult (Figure 3-S3A and B). We add this results in lines 351 – 356 and 281-285.

(3) EB treatment in adults increases the gene expression of JHAMT and Kr-h1 (Figure 3-S3C and D). We add this results in lines 378 – 379, lines 387-390 and lines 457-462.

(3) The writing quality of the paper needs improvement. Particularly with respect to describing processes and abbreviations. In several instances the authors have not adequately described the processes they have introduced, thus confusing readers.

Thank you for your suggestion. We have thoroughly revised the paper to improve clarity.

(4) In the section 'EB promotes ovarian development' the authors have shown that EB treatment results in increased detention of eggs which contradicts their own results which show that EB promotes egg laying. Again, this is a serious contradiction that nullifies their hypothesis.

Thank you for pointing this out. We revised the figure 2B to show number of mature eggs in the ovary. The number of mature eggs in ovaries of females that fed on EB was higher than in control females. We also show that BPH fed with EB laid more eggs than controls. Thus, our results suggest that EB promotes ovary maturation (and egg production) and also increases egg laying (Figure 1 and Table S1). Thus, we found that EB treatment can increase both the production of eggs and increase egg laying. We add this results in lines 234 – 238.

(5) Furthermore, the results suggest that oogenesis is not affected by EB application. The authors should devote a section to discussing how they are observing increased egg numbers in EB-treated insects while not impacting Oogenesis.

Thank you for your suggestions, and apologies for the lack of clarity in our initial explanation. First, we found that EB treatment led to an increase in the number of eggs laid by female brown planthoppers (Figure 1). Through dissection experiments, we observed that EB-treated females had more mature eggs in their ovaries (Figure 2A and B), indicating that the increased egg-laying was due to a larger production of mature eggs in the ovaries after EB treatment. This is now explained on lines 229-238.

Additionally, since there is no systematic description of oogenesis in the brown planthopper, we were the first to observe the oogenesis process in this species using immunohistochemistry and laser confocal microscopy. Based on the developmental characteristics, we defined the different stages of oogenesis (Figure 2C, Figure 2-figure supplement 2). We did not observe any significant effect of EB treatment on the various stages of oogenesis, indicating that EB treatment does not impair normal egg development (Figure 2D). Instead, the increase in vitellogenin accelerates the production of mature eggs. This is now explained on lines 243-262.

During the maturation process, eggs require uptake of vitellogenin, and an increase in vitellogenin (Vg) content can accelerate egg maturation, producing more mature eggs. Our molecular data suggest that EB treatment leads to an upregulation of vg expression. Based on these findings, we conclude that the increase in egg-laying caused by EB treatment is due to the upregulation of vg (Figure 3I), which raises vitellogenin content, promoting the uptake of vitellogenin by maturing eggs and resulting in the production of more mature eggs. We have revised the text on lines 389-395 to clarify this point.

(6) Met is the receptor of JH and to my understanding, remains mostly constant in terms of its mRNA or protein levels throughout various developmental periods in many different insects. Therefore, the presence of JH becomes the major driving factor for physiological events and not the presence of the receptor Met. Here the authors have demonstrated an increase in Met mRNA as a result of EB treatment. Their central hypothesis is that EB increases JH titer to result in enhanced fecundity. JH action will not result in the activation of Met. Although not contradictory to the hypothesis, the increase in mRNA content of Met is contrary to the findings of the JH field thus far.

Thank you for your comment. Our results showed that EB treatment can mildly increase (about 2-fold) expression of the Met gene in brown planthoppers (Figure 3G). And our data indicated that Met and FAMeT expression levels were not influenced so much by EB compared with kr-h1 and vg (Figure 3H and I). We agree that JH action will not result in the increase of Met. However, we cannot rule out the possibility of other factors (indirect effects), induced by EB treatment that increase the mRNA expression level of Met. One recent paper reported that downregulation of transcription factor CncC will increase met expression in beetles (see Figure 6A in this reference) (Jiang et al., 2023). Many studies have reported that insecticide treatment will activate the CncC gene signaling pathway, which regulates detoxification gene expression (Amezian et al., 2023; Fu et al., 2024; Hu et al., 2021). Hence, it is possible that EB might influence the CncC gene pathway which then induces met expression. This EB effect on met upregulation may be similar to the upregulation of GluCl and some other secondary effects. We have discussed this on lines 725-738.

(7) As pointed out before, it is hard to rationalize how a 4th instar exposure to EB can result in the upregulation of key genes involved in JH synthesis at the adult stage. The authors must consider providing a plausible explanation and discussion in this regard.

Thank you for your comments. It must be mentioned that although we exposed the BPH to EB at 4th instar, we make the insect feed on the EB-treated rice plants for four days. After that, the insect will develop into 5^th instar, the final nymphal stage of brown planthopper. Since brown planthoppers do not have a pupal stage, this might cause the EB presented to the insects last a longer time even in the adult stage. Besides this, we found that EB treatment will increase the weight of adult females (Figure 1-figure supplement 3E and F), which indicates that EB might increase food intake in BPHs that might produce more insulin peptide. Insulin might increase the JH synthesis at the adult stage. In our revised study we also investigate EB impairment in adult BPHs. We found that, similar to the nymphal stage, EB treatment in adult BPHs also increases the egg laying. Furthermore, the JH titer was increased after treatment of BPH with EB in adults. Besides this, GluCl and kr-h1 genes were also up-regulated after EB treatment in the adult stage. We have discussed this on lines 739-746.

(8) I have strong reservations against such an irrational hypothesis that Met (the receptor for JH) and JH-Met target gene Kr-h1 regulate JH titer (Line 311, Fig 3 supplemental 2D). This would be the first report of such an event on the JH field and therefore must be analysed in depth. I strongly suggest the authors remove such claims from the manuscript without substantiating it.

Thank you for your suggestions and comments. We have changed our claims in this revised MS. We found that EB treatment can enhance Kr-h1 expression. We have no evidence to support that JH can induce met expression. We have rewritten the manuscript to avoid confusion (see text on lines 725-735).

(9) Kr-h1 is JH/Met target gene. The authors demonstrate that silencing of Kr-h1 results in inhibition of FAMeT, which is a gene involved in JH synthesis. A feedback loop in JH synthesis is unreported. It is the view of this reviewer that the authors must go ahead with a mechanistic detail of Kr-h1 mediated JH upregulation before this can be concluded. Mere qPCR experiments are not sufficient to substantiate a claim that is completely contrary to the current understanding of the JH signalling pathway.

Thank you for your suggestions and comments. We agree that only qPCR experiments are not enough to provide this kind of claim. More evidences need to be provided to support this. We have revised the MS to avoid confusion (see text on lines 725-735).

(10) The authors have performed knockdowns of JHAMT, Met, and Kr-h1 to demonstrate the effect of these factors on fecundity in BPH. Additionally, they have performed rescue experiments with EB application on these knockdown insects (Figure 3K-M). This, I believe, is a very flawed experiment. The authors demonstrate EB works through JHAMT in upregulating JH titer. In the absence of JHAMT, EB application is not expected to rescue the phenotype. But the authors have reported a complete rescue here. In the absence of Met, the receptor of JH, either EB or JH is not expected to rescue the phenotype. But a complete rescue has been reported. These two experimental results contradict their own hypothesis.

Thank you for your comments. We thought that this rescue is possible since knockdown of the genes is incomplete when using dsRNA injection (and residual gene expression allows for EB action). It is not a total knockout and actually, these genes still have a low level of expression in the dsRNA-injected insects. Since EB can upregulate the expression of JHAMT, Met, and Kr-h1, it is reasonable that EB treatment can rescue the down-regulation effects of these three genes and make fecundity completely rescued. We have clarified this on lines 411-413).

(11) A significant section of the paper deals with how EB upregulates JH titer. JH is a hormone synthesized in the Corpora Allata. Yet the authors have chosen to use the whole body for all of their experiment. Changes in the whole body for mRNA of those enzymes involved in JH synthesis may not reflect the situation in Corpora Allata. Although working with Corpora Allata is challenging, discarding the abdomen and thorax region and working with the head and neck region of the insect is easily doable. Results from such sampling are always more convincing when it comes to JH synthesis studies.

Thank you for your suggestions. Because the head is very difficult to separate from the thorax region in brown planthoppers as you can see in Author response image 1. We are now trying to answer how EB regulates JH synthesis using Drosophila as a model.

Author response image 1.

The brown planthopper

(12) The phenomenon reported was specific to BPH and not found in other insects. This limits the implications of the study.

Thank you for your comments. The brown planthopper is a serious insect pest on rice in Asia. Our findings can guide the use of this insecticide in the field. Besides this, our findings indicated that EB, which targets GluCl can impair the JH titer. Our findings added new implications for how a neuronal system influences the JH signaling pathway. We will further investigate how EB influences JH in the future and will use Drosophila as a model to study the molecular mechanisms.

(13) Overall, the molecular experiments are very poorly designed and can at best be termed superficial. There are several contradictions within the paper and no discussion or explanation has been provided for that.

Thank you for your comments. We have revised the paper according to your suggestions and added further explanation of our results in the discussion parts and hope the conclusions are better supported in the new version. We have discussed this on lines 725-746 and 778-799.

Reviewer #2 (Public Review):

The brown plant hopper (BPH) is a notorious crop pest and pesticides are the most widespread means of controlling its population. This manuscript shows that in response to sublethal doses of the pesticide (EB), BPH females show enhanced fecundity. This is in keeping with field reports of population resurgence post-pesticide treatment. The authors work out the mechanism behind this increase in fecundity. They show that in response to EB exposure, the expression of its target receptor, GluCl, increases. This, they show, results in an increase in the expression of genes that regulate the synthesis of juvenile hormone (JH) and JH itself, which, in turn, results in enhanced egg-production and egg-laying. Interestingly, these effects of EB exposure are species-specific, as the authors report that other species of plant hoppers either don't show enhanced fecundity or show reduced fecundity. As the authors point out, it is unclear how an increase in GluCl levels could result in increased JH regulatory genes.

We greatly appreciate your valuable comments and constructive suggestion to our work. We will try to figure out how EB interacts with its molecular target GluCl and then increases JH regulatory genes in the future work using Drosophila as models.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Overall, the molecular experiments are very poorly designed and can at best be termed superficial. There are several contradictions within the paper and no discussion or explanation has been provided for that.

The authors should consider a thorough revision.

Thank you for your comments. We have thoroughly revised the paper according to your suggestions and added further experiments and explanations of our results in the discussion parts.

Reviewer #2 (Recommendations For The Authors):

It would help the reader to have more schematics along with the figures. The final figure is helpful, but knowing the JH pathway, and where it acts would help with the interpretations as one reads the manuscript and the figures. The pathways represented in 4N or 5J are helpful but could be improved upon for better presentation.

It would be nice to have some discussion on how the authors think EB exposure results in an increase in GluCl expression, and how that in turn affects the expression of so many genes.

Thank you for your comments. We have thoroughly revised the paper according to your suggestions and added further experiments and explanations of how we think EB exposure results in an increase in JH titer and other genes in the discussion parts. We have added the test on lines 753-761.

References

Amezian, D., Fricaux, T., de Sousa, G., Maiwald, F., Huditz, H.-I., Nauen, R., Le Goff, G., 2023. Investigating the role of the ROS/CncC signaling pathway in the response to xenobiotics in Spodoptera frugiperda using Sf9 cells. Pesticide Biochemistry and Physiology 195, 105563.

Fu, B., Liang, J., Hu, J., Du, T., Tan, Q., He, C., Wei, X., Gong, P., Yang, J., Liu, S., Huang, M., Gui, L., Liu, K., Zhou, X., Nauen, R., Bass, C., Yang, X., Zhang, Y., 2024. GPCR–MAPK signaling pathways underpin fitness trade-offs in whitefly. Proceedings of the National Academy of Sciences 121, e2402407121.

Gandara, L., Jacoby, R., Laurent, F., Spatuzzi, M., Vlachopoulos, N., Borst, N.O., Ekmen, G., Potel, C.M., Garrido-Rodriguez, M., Böhmert, A.L., Misunou, N., Bartmanski, B.J., Li, X.C., Kutra, D., Hériché, J.-K., Tischer, C., Zimmermann-Kogadeeva, M., Ingham, V.A., Savitski, M.M., Masson, J.-B., Zimmermann, M., Crocker, J., 2024. Pervasive sublethal effects of agrochemicals on insects at environmentally relevant concentrations. Science 386, 446-453.

Gong, Y., Cheng, S., Desneux, N., Gao, X., Xiu, X., Wang, F., Hou, M., 2022. Transgenerational hormesis effects of nitenpyram on fitness and insecticide tolerance/resistance of Nilaparvata lugens. Journal of Pest Science.

Hu, B., Huang, H., Hu, S., Ren, M., Wei, Q., Tian, X., Esmail Abdalla Elzaki, M., Bass, C., Su, J., Reddy Palli, S., 2021. Changes in both trans- and cis-regulatory elements mediate insecticide resistance in a lepidopteron pest, Spodoptera exigua. PLOS Genetics 17, e1009403.

Jiang, H., Meng, X., Zhang, N., Ge, H., Wei, J., Qian, K., Zheng, Y., Park, Y., Reddy Palli, S., Wang, J., 2023. The pleiotropic AMPK–CncC signaling pathway regulates the trade-off between detoxification and reproduction. Proceedings of the National Academy of Sciences 120, e2214038120.

Ko, K.I., Root, C.M., Lindsay, S.A., Zaninovich, O.A., Shepherd, A.K., Wasserman, S.A., Kim, S.M., Wang, J.W., 2015. Starvation promotes concerted modulation of appetitive olfactory behavior via parallel neuromodulatory circuits. eLife 4, e08298.

Li, Z., Wang, Y., Qin, Q., Chen, L., Dang, X., Ma, Z., Zhou, Z., 2023. Imidacloprid disrupts larval molting regulation and nutrient energy metabolism, causing developmental delay in honey bee Apis mellifera. eLife

Martelli, F., Hernandes, N.H., Zuo, Z., Wang, J., Wong, C.-O., Karagas, N.E., Roessner, U., Rupasinghe, T., Robin, C., Venkatachalam, K., Perry, T., Batterham, P., Bellen, H.J., 2022. Low doses of the organic insecticide spinosad trigger lysosomal defects, elevated ROS, lipid dysregulation, and neurodegeneration in flies. eLife 11, e73812.

Peng, Y.C., Sheng, C.W., Casida, J.E., Zhao, C.Q., Han, Z.J., 2017. Ryanodine receptor genes of the rice stem borer, Chilo suppressalis: Molecular cloning, alternative splicing and expression profiling. Pestic. Biochem. Physiol. 135, 69-77.

Root, Cory M., Ko, Kang I., Jafari, A., Wang, Jing W., 2011. Presynaptic facilitation by neuropeptide signaling mediates odor-driven food search. Cell 145, 133-144.

Zhang, Y., Wang, X., Yang, B., Hu, Y., Huang, L., Bass, C., Liu, Z., 2015. Reduction in mRNA and protein expression of a nicotinic acetylcholine receptor α8 subunit is associated with resistance to imidacloprid in the brown planthopper, Nilaparvata lugens. Journal of Neurochemistry 135, 686-694.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

The authors bring together implanted radiofrequency coils, high-field MRI imaging, awake animal imaging, and sensory stimulation methods in a technological demonstration. The results are very detailed descriptions of the sensory systems under investigation.

Strengths:

- The maps are qualitatively excellent for rodent whole-brain imaging. - The design of the holder and the coil is pretty clever.

Weaknesses:

- Some unexpected regions appear on the whole brain maps, and the discussion of these regions is succinct.

- The authors do not make the work and e ort to train the animals and average the data from several hundred trials apparent enough. This is important for any reader who would like to consider implementing this technology.

- The data is not available. This does not let the readers make their own assessment of the results.

Thank you for the comments on this manuscript. We have provided more detailed discussion of the unexpected regions(page 18 – line 491-494) and training procedures(page7-9 – line 172-236). We also uploaded the datasets to OpenNeuro 

Whisker (https://doi.org/10.18112/openneuro.ds005496.v1.0.1),  Visual (https://doi.org/10.18112/openneuro.ds005497.v1.0.0) and Zenodo:

SNR Line Profile Data & Data Processing Scripts:  (https://zenodo.org/doi/10.5281/zenodo.13821455). 

Reviewer #2 (Public Review):

Summary:

The manuscript by Hike et al. entitled 'High-resolution awake mouse fMRI at 14 Tesla' describes the implementation of awake mouse BOLD-fMRI at high field. This work is timely as the field of mouse fMRI is working toward collecting high-quality data from awake animals. Imaging awake subjects o ers opportunities to study brain function that are otherwise not possible under the more common anesthetized conditions. Not to mention the confounding e  ects that anesthesia has on neurovascular coupling. What has made progress in this area slow (relative to other imaging approaches like optical imaging) is the environment within the MRI scanner (high acoustic noise) - as well as the intolerance of head and body motion. This work adds to a relatively small, but quickly growing literature on awake mouse fMRI. The findings in the study include testing of an implanted head-coil (for MRI data reception). Two designs are described and the SNR of these units at 9.4T and 14T are reported. Further, responses to visual as well as whisker stimulation recorded in acclimated awake mice are shown. The most interesting finding, and most novel, is the observation that mice seem to learn to anticipate the presentation of the stimulus - as demonstrated by activations evident ~6 seconds prior to the presentation of the stimulus when stimuli are delivered at regular intervals (but not when stimuli are presented at random intervals). These kinds of studies are very challenging to do. The surgical preparation and length of time invested into training animals are grueling. I also see this work as a step in the right direction and evidence of the foundations for lots of interesting future work. However, I also found a few shortcomings listed below.

Weaknesses:

(1) The surface coil, although o ering a great SNR boost at the surface, ultimately comes at a cost of lower SNR in deeper more removed brain regions in comparison to commercially available Bruker coils (at room temperature). This should be quantified. A rough comparison in SNR is drawn between the implanted coils and the Bruker Cryoprobe - this should be a quantitative comparison (if possible) - including any di erences in SNR in deeper brain structures. There are drawbacks to the Cryoprobe, which can be discussed, but a more thorough comparison between the implanted coils, and other existing options should be provided (the Cryoprobe has been used previously in awake mouse experiments(Sensory evoked fMRI paradigms in awake mice - Chen, Physiological e ects of a habituation procedure for functional MRI in awake mice using a cryogenic radiofrequency probe – Yoshida, PREVIOUS REFERENCE). Further, the details of how to build the implanted coils should be provided (shared) - this should include a parts list as well as detailed instructions on how to build the units. Also, how expensive are they? And can they be reused?

Thank you for the comment. We did not use a Bruker Cryoprobe for this work but rather a Bruker 4array surface coil. We are unable to compare to a cryoprobe since we do not have access to one for our system. A comparison to previously published data using different scanners could be possible but would require the sequence contain identical parameters to avoid introducing an uncontrollable variable, we are planning to recruit different laboratories to test the implanted RF coils with their existing cryoprobes in the future study. 

We have included an updated figure comparing SNR at different depths across the Bruker 4-array coil and the implanted RF coils. As shown in Supplementary Figure 7B, there is significant SNR enhancement up to 4 mm cortical depth for both single loop and Figure 8 implanted RF coils in comparison to the Bruker 4-array coil.

Author response image 1.

Comparison between implanted and commercial coils. A shows representative coils in the single loop (left) and figure 8 styles (right). Supplementary Table 1 provides a parts list and cost for making these coils and Supplementary Figure 1 provides a circuit diagram to assemble. B presents the SNR line profile values as a function of distance from Pia Matter for each coil tested at 9.4T: commercial phased array surface coil (4 Array), implanted single loop, and implanted figure 8. SNR values were calculated by dividing the signal by the standard deviation of the noise. C-E shows a representative FLASH image with line profile of SNR measurements from each of the coils used to create the graph seen in B. Clear visual improvement in SNR can be seen in figures C-E. C – Commercial phased array. D – Single loop at 9.4T. E – Figure 8 at 9.4T. (N4 array = 6, Nsingle loop = 5, Nfigure 8 = 5)

Additionally, we have added a supplementary figure (supp fig 1) of a circuit diagram, in an effort to disseminate the prototype design of the coils to other laboratories. We have included a detailed parts list with the cost for construction of the coils configured for our scanner(supp table 1). These specifics though would need to be adjusted to the precise field strength/bore size/animal the coil was being built for. As for reusability, the copper wire is cemented to the animal skull and this implantable coil should be considered as consumables for the awake mouse experiments, though the PCB parts can be retrieved.  

(2) In the introduction, the authors state that "Awake mouse fMRI has been well investigated". I disagree with this statement and others in the manuscript that gives the reader the impression that awake experiments are not a challenging and unresolved approach to fMRI experiments in mice (or rodents). Although there are multiple labs (maybe 15 worldwide) that have conducted awake mouse experiments (with varying degrees of success/thoroughness), we are far from a standardized approach. This is a strength of the current work and should be highlighted as such. I encourage the authors to read the recent systematic review that was published on this topic in Cerebral Cortex by Mandino et al. There are several elements in there that should influence the tone of this piece including awake mouse implementations with the Bruker Cryoprobe, prevalence of surgical preparations, and evaluations of stress.

Thank you for the comment. We agree with the reviewer that the current stage of awake mouse fMRI studies remains to be improved.  And, we have revised the Introduction to highlight the state-of-theart of awake mouse fMRI (Page 4 – line 81-88). 

(3) The authors also comment on implanted coils reducing animal stress - I don't know where this comment is coming from, as this has not been reported in the literature (to my knowledge) and the authors don't appear to have evaluated stress in their mice. 

Since question 3 and 4 are highly related to the acclimation procedures, we will answer the two questions together.   

(4) Following on the above point, measures of motion, stress, and more details on the acclimation procedure that was implemented in this study should be included.

We thank the reviewer to raise the animal training issues.  

During the animal training, we have measured both pupil dynamic and eye motion features from training sessions, of which the detailed procedure is described in Methods (page 7-9 – line 172236). 

The training procedure is carried out over a total of 5 weeks with four phases of training: i. Holding animal in hands, ii. Head-fixation and pupillometry, iii. Head-fixation and pupillometry with mockMRI acoustic exposure, iv. Head-fixation and pupillometry with Echo-Planar-Imaging (EPI) in the MR scanner.

Author response table 1.

As shown in Supp Fig 2B, the spectral power of pupil dynamics (<0.02Hz) and eye movements gradually increased as a function of the training time for head-fixed mice exposed to the mock MRI acoustic environment during phase 3.  In phase 4, when head-fixed mice were put into the scanner for the first time, both eye movements and pupil dynamics were initially reduced during scanning but recovered to an acclimated state on Day 2, similar to the level on Day 8 of phase 3.  These behavioral outputs would provide an alternative way to monitor the stress levels of the mice. 

Author response image 2.

The eye movements (A) and power spectra of pupil dynamics (<0.02Hz) (B) change during different training phases.

It should be noted that stress may be related to increased frequency of eye blinking or twitching movements in human subjects(1–3). Whereas, the eyeblink of head-fixed mice has been used for behavioral conditioning to investigate motor learning in normal behaving mice(4–6). Importantly, head-fixed mouse studies have shown that eye movements are significantly reduced compared to the free-moving mice(7). The increased eye movement during acclimation process would indicate an alleviated stress level of the head-fixed mice in our cases. Meanwhile, stress-related pupillary dilation could dominate the pupil dynamics at the early phase of training(8). We have observed a gradually increased pupil dynamic power spectrum at the ultra-slow frequency during phase 3, presenting the alleviated stress-related pupil dilation but recovered pupil dynamics to other factors, including arousal, locomotion, startles, etc. in normal behaving mice.  Despite the extensive training procedure of the present work in comparison to the existing awake mouse fMRI studies (training strategies for awake mice fMRI have been reviewed by Mandino et al. to show the overall training duration of existing studies(9)), the stress remains a confounding factor for the brain functional mapping in head-fixed mice. In particular, a recent study(10) shows that the corticosterone concentration in the blood samples of head-fixed mice is significantly reduced on Day 25 following the training but remains higher than in the control mice. In the discussion section, we have discussed the potential issues of stress-related confounding factors for awake mouse fMRI studies (Page 16 – lines 436-458). 

(1) A. Marcos-Ramiro, D. Pizarro-Perez, M. Marron-Romera, D. Gatica-Perez, Automatic blinking detection towards stress discovery. ICMI 2014 - Proceedings of the 2014 International Conference on Multimodal Interaction 307–310 (2014). https://doi.org/10.1145/2663204.2663239/SUPPL_FILE/ICMI1520.MP4.

(2) M. Haak, S. Bos, S. Panic, L. Rothkrantz, DETECTING STRESS USING EYE BLINKS AND BRAIN ACTIVITY FROM EEG SIGNALS. Lance 21, 76 (2009).

(3) E. Del Carretto Di Ponti E Sessam, Exploring the impact of Stress and Cognitive Workload on Eye Movements: A Preliminary Study. (2023).

(4) S. A. Heiney, M. P. Wohl, S. N. Chettih, L. I. Ru olo, J. F. Medina, Cerebellar-dependent expression of motor learning during eyeblink conditioning in head-fixed mice. J Neurosci 34, 14845–14853 (2014).

(5) S. N. Chettih, S. D. Mcdougle, L. I. Ruffolo, J. F. Medina, Adaptive timing of motor output in the mouse: The role of movement oscillations in eyelid conditioning. Front Integr Neurosci 5, 12996 (2011).

(6) J. J. Siegel, et al., Trace Eyeblink Conditioning in Mice Is Dependent upon the Dorsal Medial Prefrontal Cortex, Cerebellum, and Amygdala: Behavioral Characterization and Functional Circuitry. eNeuro 2, 51–65 (2015).

(7) A. F. Meyer, J. O’Keefe, J. Poort, Two Distinct Types of Eye-Head Coupling in Freely Moving Mice. Current Biology 30, 2116-2130.e6 (2020).

(8) H. Zeng, Y. Jiang, S. Beer-Hammer, X. Yu, Awake Mouse fMRI and Pupillary Recordings in the UltraHigh Magnetic Field. Front Neurosci 16, 886709 (2022).

(9) F. Mandino, S. Vujic, J. Grandjean, E. M. R. Lake, Where do we stand on fMRI in awake mice? Cereb Cortex 34 (2024).

(10) K. Juczewski, J. A. Koussa, A. J. Kesner, J. O. Lee, D. M. Lovinger, Stress and behavioral correlates in the head-fixed method: stress measurements, habituation dynamics, locomotion, and motor-skill learning in mice. Scientific Reports 2020 10:1 10, 1–19 (2020).

(5) It wasn't clear to me at what times the loop versus "Figure 8" coil was being used, nor how many mice (or how much data) were included in each experiment/plot. There is also no mention of biological sex.

Thank you for the comment. We have clarified sex and number. The figure 8 coil was only used as part of development to show the improvement of the coil design for cortical measurements. The detailed information is described in Method (Page 6 – line 127-129 & Page 10 – line 269-270). Additionally animal numbers have been included in the figure captions.

(6) Building on the points above, the manuscript overall lacks experimental detail (especially since the format has the results prior to the methods).

Thank you for the comment. We have modified the manuscript to increase the experimental detail and moved the methods section before the results.

(7) An observation is made in the manuscript that there is an appreciable amount of negative BOLD signal. The authors speculate that this may come from astrocyte-mediated BOLD during brain state changes (and cite anesthetized rat and non-human primate experiments). This is very strange to me. First, the negative BOLD signal is not plotted (please do this), further, there are studies in awake mice that measure astrocyte activation eliciting positive BOLD responses (see Takata et al. in Glia, 2017).

We thank the reviewer to raise the negative BOLD fMRI observation issue.  We added a subplot of the negative BOLD signal changes in the revised Figure 4. This negative BOLD signals across cortical areas could be coupled with brain state changes upon air-pu -induced startle responses. Our future studies are focusing on elucidating the brain-wide activity changes of awake mice with fMRI.  We also provide a detailed discussion of the potential mechanism underlying the negative BOLD fMRI signals. First, as reported in the paper (suggested  by the reviewer),  astrocytic Ca2+ transients coincide with positive BOLD responses in the activated cortical areas, which is aligning with the neurovascular coupling (NVC) mechanism. However, there is emerging evidence to show that astrocytic Ca2+ transients are coupled with both positive and negative BOLD responses in anesthetized rats(11) and awake mice(12). An intriguing observation is that cortex-wide negative BOLD signals coupled with the spontaneous astrocytic Ca2+ transients could co-exist with the positive BOLD signal detected at the activated cortex.  Studies have shown that astrocytes are involved in regulating brain state changes(13), in particular, during locomotion(14) and startle responses(15). These brain state-dependent global negative BOLD responses are also related to the arousal changes of both non-human primates(16) and human subjects(17).  The established awake mouse fMRI platform with ultra-high spatial resolution will enable the brain-wide activity mapping of the functional nuclei contributing to the brain state changes of head-fixed awake mice in future studies. (Page 17-18 – Line 478-490)

(11) M. Wang, Y. He, T. J. Sejnowski, X. Yu, Brain-state dependent astrocytic Ca2+ signals are coupled to both positive and negative BOLD-fMRI signals. Proc Natl Acad Sci U S A 115, E1647–E1656 (2018).

(12) C. Tong, Y. Zou, Y. Xia, W. Li, Z. Liang, Astrocytic calcium signal bidirectionally regulated BOLD-fMRI signals in awake mice in Proc. Intl. Soc. Mag. Reson. Med. 32, (2024).

(13) K. E. Poskanzer, R. Yuste, Astrocytes regulate cortical state switching in vivo. Proc Natl Acad Sci U S A 113, E2675–E2684 (2016).

(14) M. Paukert, et al., Norepinephrine controls astroglial responsiveness to local circuit activity. Neuron 82, 1263–1270 (2014).

(15) R. Srinivasan, et al., Ca2+ signaling in astrocytes from IP3R2−/− mice in brain slices and during startle responses in vivo. Nat Neurosci 18, 708 (2015).

(16) C. Chang, et al., Tracking brain arousal fluctuations with fMRI. Proc Natl Acad Sci U S A 113, 4518– 4523 (2016).

(17) B. Setzer, et al., A temporal sequence of thalamic activity unfolds at transitions in behavioral arousal state. Nat Commun 13 (2022).

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

I really enjoyed this work. The maps shown are among the best-quality maps out there. Here are suggestions to the authors.

(1) Both the ACA and VRA are rather unexpected. The authors explain these briefly as being part of the associative cortical areas. Both the ACA and VRA are not canonical associative areas (or at least not to us). This warrants a stronger discussion.

To verify both ACA and VRA as associate areas, we provide the  connectivity map projections from the Allen Brain Atlas (seen below). These projections are derived from a Cre-dependent AAV tracing of axonal projections. We have included an explanation of this in the introduction. 

Author response image 3.

Representative images are shown indicating connections between the barrel cortex and retrosplenial area from an injection in the barrel cortex (Left panel) as well as the visual cortex and cingulate connection from an injection in the visual cortex (Right panel). Images are of connectivity map projections from the Allen Brain Atlas derived from a Cre-dependent AAV tracing of axonal projections

(2) This is a lot of work. But looking at the figures, this is not obvious. We read in the caption that several hundred trials were used. It would be good to also specify how many mice. It would be clearer to represent this info in the figure as well to support the fact that this is not a trivial acquisition.

Thank the reviewer to raise the e ort issue. We have edited the figure to include this information and included the numbers in the text as well

(3) The training protocol is seemingly extensive, but this is only visible by following another reference. Including a description in this work would help the reader make sense of the effort that went into this work.

We thank the reviewer to raise the training protocol issue. We have more thoroughly discussed the training method used for this study (page 7-9 – line 172-236)

(4) I really would love to see that dataset made freely available - this should be the norm.

The datasets have been uploaded to OpenNeuro 

Whisker (https://doi.org/10.18112/openneuro.ds005496.v1.0.1),  Visual (https://doi.org/10.18112/openneuro.ds005497.v1.0.0) and Zenodo:

SNR Line Profile Data & Data Processing Scripts: 

(https://zenodo.org/doi/10.5281/zenodo.13821455). 

(page 21 – line 573-579)

Reviewer #2 (Recommendations For The Authors):

(1) I'm a little confused about the stimulation paradigm and the effect of it causing an effective 2second TR (which is on the long side) - please elaborate (a figure might be helpful). The paradigm for visual stimulation also seems elaborate, can you please explain the logic and how it was developed?

Thank you for raising the detailed stimulation paradigm issues. The stimulation paradigm is independent and does not interfere with the setup of the effective 2-second TR. The 2-second TR is based on the usage of 2-segment EPI, each with a TR of 1-second. The application of 2-segment paradigm enables the echo spacing with 0.52 ms with effective image bandwidth with 3858Hz, assuring less image distortion.  The stimulation paradigm was defined by an “8s on, 32s o ” epoch such to elicit a strong BOLD response and could be used for any reasonable TR duration. 

We have included a figure outlining the stimulation paradigm (Supp Fig. 3)

(2) I had difficulties viewing the movies (on my MAC).

Thank you for this note. We have re-upload the videos in .mov format

Author response:

The following is the authors’ response to the original reviews

Reviewer #1 (Public Review):  

Summary:

In this study, Setogawa et al. employ an auditory discrimination task in freely moving rats, coupled with small animal imaging, electrophysiological recordings, and pharmacological inhibition/lesioning experiments to better understand the role of two striatal subregions: the anterior Dorsal Lateral Striatum (aDLS) and the posterior Ventrolateral Striatum (pVLS), during auditory discrimination learning. Attempting to better understand the contribution of different striatal subregions to sensory discrimination learning strikes me as a highly relevant and timely question, and the data presented in this study are certainly of major interest to the field. The authors have set up a robust behavioral task and systematically tackled the question about a striatal role in learning with multiple observational and manipulative techniques. Additionally, the structured approach the authors take by using neuroimaging to inform their pharmacological manipulation experiments and electrophysiological recordings is a strength.

However, the results as they are currently presented are not easy to follow and could use some restructuring, especially the electrophysiology. Also, the main conclusion that the authors draw from the data, that aDLS and pVLS contribute to different phases of discrimination learning and influence the animal's response strategy in different ways, is not strongly supported by the data and deserves some additional caveats and limitations of the study in the discussion. 

We appreciate the reviewer’s valuable feedback, which has been beneficial for improvement of our manuscript. In response to the reviewer’s comments, we have revised multiple parts of the manuscript, including explanations of electrophysiological data. We have also provided additional data to support our main conclusion and addressed caveats and limitations related to the data in the Discussion section. For more details, please refer to the responses to each comment.

Comment 1: The authors have rigorously used PET neuroimaging, which is an interesting noninvasive method to track brain activity during behavioral states. However, in the case of a freely moving behavior where the scans are performed ~30 minutes after the behavioral task, it is unclear what conclusions can be drawn about task-specific brain activity. The study hinges on the neuroimaging findings that both areas of the lateral striatum (aDLS and pVLS) show increased activity during acquisition, but the DMS shows a reduction in activity during the late stages of behavior, and some of these findings are later validated with complementary experiments. However, the limitations of this technique can be further elaborated on in the discussion and the conclusions.

As described in our response to the following two comments (a, b) from the reviewer, in the PET imaging study we first analyzed task-related activity by comparing ¹⁸F-FDG uptake on different days of the auditory discrimination task with that on Day 4 of the single lever press task as a control. Next, we analyzed learning-dependent activity by comparing the uptake on different days of the discrimination task with that on Day 2 of the same task. Based on the results of both analyses, we concluded that the activity in the striatal subregions changes during the progress of discrimination learning. The behavioral significance of striatal subregions was tested by excitotoxic lesion and pharmacological blockade experiments. The explanation of imaging data analysis may have been insufficient to fully communicate dynamic changes in the activity of striatal subregions. Therefore, we have clarified our voxel-based statistical parametric analysis method to better explain the dynamic activity changes in the striatal subregions. Please refer to the following responses to comments 1 (a, b).

Comment 1 (a): In commenting on the unilateral shifts in brain striatal activity during behavior, the authors use the single lever task as a control, where many variables affecting neuronal activity might be different than in the discriminatory task. The study might be better served using Day 2 measurements as a control against which to compare activity of all other sessions since the task structures are similar.

We initially analyzed task-related activity by comparing ¹⁸F-FDG uptake on one of Days 2, 6, 10, or 24 of auditory discrimination task with that on Day 4 of the single lever press task. This task was used as a control that does not require a decision process based on the auditory stimulus. We observed significant increases in the activity of the unilateral aDLS on Day 6 and in that of the bilateral pVLS on Day 10 of the discrimination task. We also observed a significant decrease in the unilateral DMS on Day 24 (see Figures 2F and 2G). Next, as suggested, we compared the uptake on one of Days 6, 10, or 24 with that on Day 2 as a control to evaluate learning-dependent activity. The activity showed significant increases in the bilateral aDLS on Day 6 and in the unilateral pVLS on Day 10, and a significant decrease in the bilateral DMS on Day 24 (see Figures 2H). 

The reviewer has suggested a discrepancy in the activity of the unilateral or bilateral striatal subregions under certain conditions between the image data (shown in Figures 2F–H) and plot data (Figures 2J–L). This discrepancy is also suggested in the following Comment 1 (b). For example, in the image data the brain activity was increased in the unilateral (left) aDLS on Day 6 of the discrimination task as compared to Day 4 of the single lever task (Figure 2F). In the plot data, ¹⁸F-FDG uptake reached a peak on Day 6 in both the left and right sides of the aDLS (Figure 2J), and the uptake in the left aDLS on Day 6 significantly increased relative to the value of the single lever press, whereas the value in the right aDLS on Day 6 tended to increase relative to that of the single lever press with no significant difference. The plot data showing the unilaterality in the aDLS activation relative to the single lever press are consistent with the image data. On the other hand, the ¹⁸F-FDG uptake in the aDLS on Day 6 compared to the value on Day 2 was significantly increased in both sides. Similar observations were made in the activity in the pVLS on Day 10 compared to that on Day 2, as well as in the DMS activity on Day 24 relative to that of the single lever press. 

Our analysis of both task-related and learning-dependent activities revealed dynamic changes in striatal subregions during discrimination learning. We investigated the brain regions in which ¹⁸F-FDG uptake significantly increased or decreased during the learning processes, applying a statistical significance threshold (p < 0.001, uncorrected) and an extent threshold, by using a voxel-based statistical parametric analysis. In the image data, the voxels showing significant differences between two conditions are visualized on the brain template. The plot data show the amount of ¹⁸FFDG uptake in the voxels, which was detected by the voxel-based analysis. The insufficient explanation of the data analysis of PET imaging in the initial manuscript may have led to a misunderstanding regarding the activity in the unilateral or bilateral striatal subregions. Therefore, we have revised the explanation for voxel-based statistical parametric analysis, adding a more detailed description of the thresholds in the text (page 7, lines 143–145) and Methods (page 27, lines 672–675).

Comment 1 (b): From the plots in J, K, and L, it seems that shifts in activity in the different substructures are not unilateral but consistently bilateral, in contrast to what is mentioned in the text. Possibly the text reflects comparisons to the single lever task, and here again, I would emphasize comparing within the same task.

Please see our response to the first comment (a) regarding our explanation of the consistency in the activity of the unilateral or bilateral striatal subregions between the image and plot data. We have also revised the explanation in the corresponding sections of the manuscript, as described above.

Comment 2: In Figure 2, the authors present compelling data that chronic excitotoxic lesions with ibotenic acid in the aDLS, pVLS, and DMS produce differential effects on discrimination learning. However, the significant reduction in success rate of performance happens as early as Day 6 in both IBO groups in both aDLS and pVLS mice. This would seem to agree with conclusions drawn about the role of aDLS in the middle stages of learning in Figure 2, but not the pVLS, which only shows an increased activity during the late stages of the behavior.  

Figure 3 shows the behavioral effects of ibotenic acid injections into striatal subregions in rats. For the aDLS injection, we performed two-way repeated ANOVA, which revealed a significant main effect of group or day and a significant interaction of group × day, and added the simple main effects between the treatments to the figure (Figure 3G). We observed significant differences in the success rate mainly at the middle stage of learning. In contrast, for the pVLS injection there was no significant interaction for group × day, although the main effects of group or day was significant by two-way repeated ANOVA (Figure 3H). Consequently, it was unclear as to when exactly the significant reduction occurred. These results indicate that the aDLS and pVLS are necessary for the acquisition of auditory discrimination, and that the aDLS is mainly required for the middle stage. Similar results were observed in the win-shift-win strategy in the aDLS and pVLS (Figures 3J and 3K).

Next, we performed temporal inhibition of neuronal activity in striatal subregions by muscimol treatment in order to examine whether the activity in the subregions is linked with learning processes at different stages. In this experiment, muscimol was injected into the aDLS or pVLS at the middle or late stage, and the resultant effects on the success rate were investigated. The success rate in the muscimol-injected groups into the aDLS significantly decreased at the middle stage, but not at the early and late stages (Figure 4C). In contrast, the rate in the muscimol groups into the pVLS significantly decreased at the late stage, but not at the early or middle stages (Figure 4D). The results indicate that the aDLS and pVLS are mainly involved in the processes at the middle and late stages, respectively, and support the PET imaging data showing the activation of two striatal subregions at the various stages.

We have now provided the results of simple main effects analysis for the aDLS lesion (Figures 3G and 3J) and revised the description of the Results section (page 8, lines 174–178, page 8, lines 186–188, and page 9, line 205-206) and Figure legend (page 44, lines 1000‒1003, and page 44, lines 1010–1013). We have also added the results of simple main effects analysis in Figure 3J.

Comment 3: In Figure 4, the authors show interesting data with transient inactivation of subregions of the striatum with muscimol, validating their findings that the aDLS mediates the middle and the pVLS the late stages of learning, and the function of each area serves different strategies. However, the inference that aDLS inactivation suppresses the WSW strategy "moderately" is not reflected in the formal statistical value p=0.06. While there still may be a subtle effect, the authors would need to revise their conclusions appropriately to reflect the data. In addition, the authors could try a direct comparison between the success rate during muscimol inhibition in the mid-learning session between the aDLS and pVLS-treated groups in Figure 4C (middle) and 4D (middle). If this comparison is not significant, the authors should be careful to claim that inhibition of these two areas differentially affects behavior.

In Figure 4E, aDLS inhibition showed a tendency to reduce slightly win-shift-win strategy at the middle stage (t[14] = 2.038, p = 0.061, unpaired Student’s t-test). In accordance with the reviewer’s comment, we changed the word “moderate” to “subtle” (page 12, line 272).

In the temporal inhibition of the striatal subregions, the aDLS and pVLS experiments (panels C and D, respectively) were conducted separately. Since it is difficult to directly compare the data obtained from different experiments, we did not carry out a direct comparison of the success rate between the aDLS and pVLS injections. 

Comment 4: The authors have used in vivo electrophysiological techniques to systematically investigate the roles of the aDLS and the pVLS in discriminatory learning, and have done a thorough analysis of responses with each phase of behavior over the course of learning. This is a commendable and extremely informative dataset and is a strength of the study. However, the result could be better organized following the sequence of events of the behavioral task to give the reader an easier structure to follow. Ideally, this would involve an individual figure to compare the responses in both areas to Cue, Lever Press, Reward Sound, and First Lick (in this order).

We first showed changes in the proportion of event-related neurons during the acquisition phase (Figure S5). Next, we conducted a detailed analysis of the characteristics of aDLS and pVLS neuronal activity. Specifically, we found several types of event-related neurons, including: (1) reward sound-related neurons representing behavioral outcomes in the aDLS; (2) first licking-related neurons showing sustained activity after the reward in the aDLS and pVLS; and (3) cue-onset and cue-response neurons associated with the beginning and ending of a behavior in the pVLS.

Descriptions of the characteristics of event-related neurons according to the sequence of events in a trial, as the reviewer has suggested, is another way to provide an easy structure for understandings on the electrophysiological data. However, we focused on the characteristics of aDLS neurons at the middle stage and pVLS neurons at the late stage of discrimination learning. Therefore, we explained the electrophysiological data based on the order of learning stages rather than the sequence of events in the trial, as described above.

Comment 5: An important conceptual point presented in the study is that the aDLS neurons, with learning, show a reduction in firing rates and responsiveness to the first lick as well as the behavioral outcome, and don't play a role in other task-related events such as cue onset. However, the neuroimaging data in Figure 2 seems to suggest a transient enhancement of aDLS activity in the mid-stage of discriminatory learning, that is not reflected in the electrophysiology data. Is there an explanation for this difference?

In the ¹⁸F-FDG PET imaging study, the brain activity in the aDLS reached a peak at the middle stage of the acquisition phase of auditory discrimination (Figure 2J). In the multi-unit electrophysiological recording experiment, the firing activity of the aDLS neuron subpopulations related to the behavioral outcome showed no significant differences among the three stages (Figure 5E), while the proportion of these subpopulations were gradually reduced through the progress of learning stages (Figure 5F). The extent of the firing activity and length of the firing period of other subpopulations showing sustained activation after the reward appeared to show a learning-dependent decrease (Figures 6B and 6C), although the proportion of these subpopulations indicated no correlation with the progress of the learning (Figure 6D). Patterns of the temporal changes in brain activity in striatal subregions across the learning stages did not match completely the time variation in the property or proportion of specific event-related neurons. In our electrophysiological analysis, we identified well-isolated neurons from the striatal subregions during the auditory discrimination task, focusing on putative medium spiny neurons (Figures S4E–S4G). Based on the combinatorial pattern of the tone instruction cue (high tone/H or low tone /L), and lever press (right/R or left/L), we categorized the electrophysiological data into the four trials, including the HR, LL. LR, and HL. We identified HR or LL type neurons showing significant changes in the firing rate related to specific events, such as cue onset, choice response, reward sound, and first licking compared to the baseline firing rate. These neurons were further divided into two groups with increased or decreased activity relative to the baseline firing (Figures S5A and S5B). In the present study, we focused on event-related neurons with increased activity. Because of the analysis limited to neuronal subpopulations related to specific events with the increased activity, it is difficult to fully explain dynamic shifts in the brain activity of striatal subregions dependent on the progress of learning by the time variation of firing activity of individual event-related neurons. The activity of other subpopulations in the striatum may be involved in the shift in brain activity during the learning processes. In addition, recent studies have reported that the activity of glial cells influences the uptake of ¹⁸FFDG (Zimmer et al., Nat Neurosci., 2017) and that these cells regulate spike timingdependent plasticity (Valtcheva and Venance, Nat Commun, 2016). Changes in glial cellular activity, through the control of synaptic plasticity, may partly contribute to the pattern formation of learning-dependent shifts in brain activity.

To explain the difference in the time course between the brain activity and the firing activity of specific event-related neurons, we have added the aforementioned information to the Limitations section (pages 21 to 22, lines 512–539). 

Comment 6: A significant finding of the study is that CO-HR and CO-LL responses are strikingly obvious in the pVLS, but not in the aDLS, in line with the literature that the posterior (sensory) striatum processes sound. This study also shows that responses to the highfrequency tone indicating a correct right-lever choice increase with learning in contrast to the low-frequency tone responses. To further address whether this difference arises from the task contingency, and not from the frequency representation of the pVLS, an important control would be to switch the cue-response association in a separate group of mice, such that high-frequency tones require a left lever press and vice versa. This would also help tease apart task-evoked responses in the aDLS, as I am given to understand all the recording sites were in the left striatum.

We did not conduct an experiment switching cue-response association in the auditory discrimination task. However, the transient activity of cue onset-related neurons in the pVLS, as the reviewer has suggested, did not appear at the early stage of learning, but was observed in a learning-dependent manner (Figures 7A and S8E). In addition, the cue onset-HR activity showed a slight but notable difference between the HR and LL trials at the middle and late stages (Figure 7B), but there was no difference in activity in the HL and LR incorrect trials at the corresponding stages (Wilcoxon signed rank test; early, p = 0.375, middle, p = 0.931, and late, p = 0.668). These results suggest that the activity of cue onset-related neurons in the pVLS is associated with the stimulus and response association (task contingency) rather than the tone frequency.

Reviewer #1 (Recommendations For The Authors):

Minor comment 1: The readability and appeal of this study would be improved by explaining the various neuronal response types, and task-related events in slightly more detail in the results section, and minimizing the use of non-standard abbreviations wherever possible.

As suggested, we have replaced the abbreviations related to electrophysiological events (CO, CR, RS, and FL) with the original terms, and improved the explanation for neuronal response types and event-related neurons. 

Minor comment 2: It would be helpful to label DLS and VLS recordings more clearly on the figures instead of only in the figure caption.

Thank you for pointing this out. The terms “aDLS” and “pVLS” have now been added to the panels showing firing pattern of neurons: “aDLS” in Figures 5D, 6A, S6A, S7A, S8A, S8B. S8C, and S8D; and “pVLS” in Figures 6F, 7A, 7D, S6D, S6E, S7F, S8E, and S8F.

Minor comment 3: The authors suggest that aDLS HR- and LL- neurons are more sensitive to the behavioral outcome than those in pVLS (Fig 5 and S5). However, their conclusions are based on sample sizes as low as n=3 for each response type.

We identified event-related neurons from single neurons detected in both the aDLS and pVLS using the same criteria. In the pVLS, we found a small number of neurons that increased their activity during the period when the reward sound is presented (Figures S6D and S6E) (6, 4, and 17 HR type neurons at the early, middle, and late stages, respectively; 3, 5, and 15 LL type neurons at the early, middle, and late stages, respectively). The number of LL type neurons at the early stage was particularly lower, as the reviewer has suggested. However, when we plotted the firing rates of these neurons around the event, their activity did not reflect behavioral outcome. In the aDLS, we detected a large number of reward sound-related neurons representing behavioral outcome (Figures 5 and S6A) (43, 37, and 44 HR type neurons at the early, middle, and late stages, respectively; 49, 62, and 59 LL type neurons at the early, middle, and late stages, respectively). These observations suggest that aDLS neurons are more sensitive to behavioral outcomes than pVLS neurons.

Minor comment 4: Typo in Figure 4C and D, right plots, y-axis label: "subtracted".

The typographic errors in Figures 4C–4H have now been corrected to “subtracted”.

Reviewer #2 (Public Reviews):

The study by Setogawa et al. aims to understand the role that different striatal subregions belonging to parallel brain circuits have in associative learning and discrimination learning (S-O-R and S-R tasks). Strengths of the study are the use of multiple methodologies to measure and manipulate brain activity in rats, from microPET imaging to excitotoxic lesions and multielectrode recordings across anterior dorsolateral (aDLS), posterior ventral lateral (pVLS)and dorsomedial (DMS) striatum. The main conclusions are that the aDLS promotes stimulus-response association and suppresses response-outcome associations. The pVLS is engaged in the formation and maintenance of the stimulus-response association. There is a lot of work done and some interesting findings however, the manuscript can be improved by clarifying the presentation and reasoning. The inclusion of important controls will enhance the rigor of the data interpretation and conclusions.

We appreciate the reviewer’s valuable feedback, which has been beneficial in our endeavor to improve our manuscript. In response to the comments, we have revised the description of the experimental methods and underlying rationale, as well as the Results section. We have also provided additional data for some of the experiments that support the conclusions. For more details, please refer to the responses to each comment, included below.

Reviewer #2 (Recommendations For The Authors):

Comment 1: Generally, the manuscript is hard to read because of the cumbersome sentence structure, overuse of poorly defined acronyms, and lack of clarity on the methods used.

According to the following comments (a)–(d), we have revised the corresponding text in the manuscript to clarify the sentence structure, definitions of terms, and methodology. 

Comment 1 (a): For example, the single lever task used as a control for the auditory discrimination task could be introduced better, explaining the reasoning and the strategy for subtracting it from the images obtained during the discrimination phase at the start of the section.

We analyzed task-related activity by comparing ¹⁸F-FDG uptake on Days 2, 6, 10, or 24 of auditory discrimination task with that on Day 4 of the single lever press task. This task was used as a control that does not require a decision process based on the auditory stimulus. For clarification, we have provided a more detailed explanation of the flow of the single lever press task used in the PET experiment, including the rationale for employing this task as a control (page 6, lines 129–135). We have also revised the explanation of voxel-based statistical parametric analysis, adding a more detailed description of the thresholds (page 7, lines 143–145).

Comment 1 (b): Another example is that important methodological information is buried deep in the text and complicates the interpretation of the results.

We have revised the following sentences in the manuscript in order to provide clearer methodological information.

(1) As described above, explanations for the single lever task (page 6, lines 129–135) and voxel-based statistical parametric analysis were added (page 7, lines 143–145). 

(2) Definition of the early, middle, and late stages were described in the initial behavioral experiment (page 6, lines 113–119). 

(3) Abbreviations related to behavioral strategies (WSW and LSL) and electrophysiological events (CO, CR, RS, and FL) were replaced with the original terms. 

Comment 1 (c): The specie being studied is not stated in the abstract, nor the introduction, and only in the middle of the result section. Please include the specie in the abstract and the first part of the result also for clarity.

We included the name of the species (rats) in the Abstract (page 3, line 47), at the end of the Introduction (page 5, lines 87–88) and at the beginning of the Results (page 5, line 109).

Comment 1 (d): The last part of the intro is copied/pasted from the abstract. Please revise.

The last part of the Introduction was revised accordingly (page 5, lines 97–104).

Comment 2: The glucose microPET imaging is carried out 30 mins after the rats performed the task and it is expected to capture activation during the task. Is this correct? This assumption has to be validated with an experiment, which is a control showing a validation of the microPET approach used, and this way can report activation of brain areas during the task completed 20-30 minutes before. For example, V1 or A1 would be a control that we would expect to be activated during the task.

Our PET experiment was conducted in accordance with previously established methods (Cui et al, Neuroimage, 2015), where rats received intravenous administration of ¹⁸FFDG solution just before the start of the behavioral session, which lasted for 30 min. The ¹⁸F-FDG uptake in the brain starts immediately and reaches the maximum level until 30 min after the administration, and the level is kept at least for 1 h (Mizuma et al., J Nucl Med, 2010). The rats were returned to their home cages, and a 30-min PET scan started 25 min after the session. The start time of the scan was chosen to allow for sufficient reduction of 18F radioactivity in arterial blood to increase the S/N ratio of the radioactivity (Mizuma et al., J Nucl Med, 2010). As shown in Table S1, we confirmed that the brain activity in the medial geniculate body (auditory thalamus) was increased on Days 6 and 10 in the acquisition phase, although the activity in the auditory cortex was not changed, which is consistent with the results of a previous study reporting that the auditory cortex does not show the causality for the pure-tone discrimination task (Gimenez et al., J Neurophysiol., 2015).

Comment 3: Why are Days 2, 6, 10, and 13 chosen and compared for the behavior? Why aren't these the same days chosen in the other part of the study? It is unclear why authors focused on these days and why the focus changed later.

We conducted daily training of the discrimination task. The success rate reached a plateau on Day 13 and was maintained until Day 24 (Figure 1B). Based on these results, we categorized the learning processes into the acquisition and learned phases, and then divided the acquisition phase into the early (< 60%), middle (60–80%), and late (> 80%) stages. In the PET experiment, we selected Days 2, 6, and 10 as the representatives of each stage during the acquisition phase. In addition, we also selected Day 24 for the learned phase.  However, no scan was performed on Day 13 due to the transition between the two phases.   

Comment 4: (A) Is the learning and acquisition of the single lever press and discrimination task completed by day 4? Or are rats still learning? The authors claimed no changes in DMS activity between single lever press & discrimination, and therefore DMS isn't involved in learning. But to make this claim we should have measures that the learning has already happened, which I am not sure have been provided. (B) On this same point, the DMS activity is elevated on Day 4 of a single lever press compared to the aDLS and pVLS. So is it possible that the activity in DMS was already elevated on Day 4 of single lever press training? Especially given that DMS is supposedly involved in goal-directed behavior?

(A) In the single lever press task, the number of lever presses plateaued on Day 2 (Figure 1C). In addition, we analyzed response time and its variability, which plateaued from Day 3 and Day 2, respectively (see Author response image 1). These results indicate that the learning in the task was completed by Day 4. In the auditory discrimination task, Day 4 corresponded to the transition period from the early-tomiddle stages of the acquisition phase, suggesting that learning was still progressing. 

In the imaging analysis, we examined task-related activity by comparing ¹⁸F-FDG uptake on either day of the discrimination task with that on Day 4 of the single lever press task, and did not find any changes in the brain activity in the DMS. In addition, we investigated learning-related activity, and the DMS activity did not change during acquisition phase. These results suggest that the DMS is not involved in the acquisition phase of learning. Furthermore, comparisons between Days 10 and Day 24 showed a decrease in DMS activity during the learned phase, suggesting that DMS activity was downregulated during the learned phase. In addition, chronic lesion in the DMS indicated that the success rate in the discrimination task was comparable between the control and lesioned groups (Figure 3I), whereas the response time lengthened throughout the learning in the lesioned group compared to the controls (Figure S1C). These results support our notion that the DMS contributes to the execution, but not learning, of discriminative behavior (Figure 3I and S1C).

Author response image 1.

Performance of single lever press task conducted before auditory discrimination task. (A) Number of lever presses. (B) Response time (Kruskal-Wallis test, χ² = 38.063, p = 2.7 × 10^-8, post hoc Tukey–Kramer test, p = 0.047 for Day 1 vs. Day 2; p = 2.3 × 10^-7 for Day 1 vs. Day 3; and p = 4.0 × 10^-6 for Day 1 vs. Day 4; p = 0.019 for Day 2 vs. Day 3; p = 0.082 for Day 2 vs. Day 4; p = 0.951 for Day 3 vs. Day 4). (C) Response time variability (Kruskal-Wallis test, χ² = 28.929, p = 2.3 × ^-6, post hoc Tukey–Kramer test, p = 0.077 for Day 1 vs. Day 2; p = 5.7 × 10^-6 for Day 1 vs. Day 3; and p = 1.3 × 10^-4 for Day 1 vs. Day 4; p = 0.060 for Day 2 vs. Day 3; p = 0.253 for Day 2 vs. Day 4; p = 0.912 for Day 3 vs. Day 4). Data obtained from the task shown in Figure 2C are plotted as the median and quartiles with the maximal and minimal values. *p < 0.05, **p < 0.01, and ***p < 0.001.

(B) We compared ¹⁸F-FDG uptakes among striatal subregions on Day 4 of the single lever press task (334.8 ± 2.86, 299.0 ± 1.71, and 336.8 ± 2.18 for the aDLS, pVLS, and DMS, respectively; one-way ANOVA, F[2,41] = 104.767, p = 2.1 × 10^-16). The uptake was comparable between the aDLS and DMS (post hoc Tukey-Kramer test, p = 0.058), but it was significantly lower in the pVLS compared to either of the other two subregions (post hoc Tukey-Kramer test, aDLS vs. pVLS, p = 5.1 × 10^-9, post hoc Tukey-Kramer test, pVLS vs. DMS, p = 5.1 × 10^-9). However, since we did not measure the brain activity in the single lever task outside of Day 4, it is unclear whether there was an increase in DMS activity during the acquisition of the task. Similarly, since we did not confirm the behavioral modes, which include goal-directed and habitual actions, it is difficult to conclude that the lever presses in the task were controlled by the goaldirected mode. However, our chronic lesion experiment suggests that the DMS is involved in the execution of discrimination behavior (Figure S1C). A clearer understanding of the DMS function in discrimination learning is an important challenge in the future.

Comment 5: It seems like the procedure of microPET imaging affects performance on the task. The anesthesia used maybe? Figures 2C and D show evidence that the behavior was negatively affected on the days on which microPET imaging was performed after the training. Can the author clarify/comment?

Isoflurane anesthesia may slightly reduce behavioral performance. We carried out anesthesia (median [interquartile range]: 6 [5–8] min) during the insertion of the catheter for FDG injection, and set a recovery period of at least 2 h until the beginning of the behavioral session, to minimize the impact of anesthesia. The performances in Figure 2E were similar to those in the intact rats (compared to Figures 1C–1F), suggesting that the procedure for PET scans does not affect the acquisition of discrimination. 

We have added detailed information on the isoflurane anesthesia to the Methods section (page 26, lines 649–653).

Comment 6: More on clarity. Section 3 of the results (muscimol inactivation) refers a lot to "the behavioral strategies" without really clarifying what these are - are they referring to WSW / LSL (which also could use a better introduction) or goal-directed/habitual or stimulus-response/stimulus-outcome?

The dorsal striatum is involved in both behavioral strategies based on stimulus-response association and the response-outcome association during instrumental learning. To assess the impact of striatal lesions on the behavioral strategies, we analyzed the proportion of response attributed to two strategies in all responses of each session. One is the “win-shift-win” strategy, which is considered to reflect the behavioral strategy based on the stimulus-response association. In this strategy, after a correct response in the previous trial, the rats press the opposite lever in the current trial in response to a shift of the instruction cue, resulting in the correct response.  Another strategy is the “lose-shift-lose” strategy, which is considered to appear as a consequence of the behavioral strategy based on the response-outcome association. In this strategy, after an error response in the previous trial, the rats press the opposite lever in the current trial despite a shift of the instruction cue, leading to another error response.

We have revised the explanations of the behavioral strategies in the section of the Results section (page 9, lines 192–201). 

Comment 7: Related to WSW / LSL needing a better introduction, on lines 192/193 authors describe a result where they saw the WSW and LSL strategies increase and decrease, respectively, in saline-injected mice. Is the change in performance expected or an undesired effect of the saline injection? This is not clear now and it should be clarified.

The explanations of the win-shift-win and lose-shift-lose strategies have been revised in the Results section on excitotoxic lesion experiment (page 9, lines 192–201) as described in our response to Comment 6. Win-shift-win is an indicator of correct responses, while lose-shift-lose indicates errors. Therefore, win-shift-win is predicted to increase, and lose-shift-lose decrease, as discrimination learning progresses. Indeed, in the results of the behavioral experiments, shown in Figure 1, both indicators change in a similar pattern to those in the results of the lesion experiments (Figure 3).

We have added the explanation of the proportions of both strategies in intact rats (page 9, lines 203–204) with a supplementary figure (Figure S2) and accompanying legend (page 56, lines 1173–1177).

Comment 8: Muscimol experiments - two questions/comments. How often do rats receive muscimol?

In this section, muscimol is given on day 2 and on days after the animals hit a 60% or 80% success rate. Can the authors provide a mean and SEM for when are those injections?

The first injection was conducted on Day 2 to target the early stage. The second and third injections were conducted on the days after the success rate had reached 60% and 80% for the first time through the training, respectively, to target the middle and late stage. respectively. These conditions are described in the Results (page 10, lines 234– 237) and Methods (page 26, lines 633–636). The mean and s.e.m. of the injection day at the middle and late stages were not significantly different between the saline and muscimol-injected groups into the aDLS (see Author response image 2A) and pVLS (see Author response image 2B).

Author response image 2.

Injection days during auditory discrimination learning. Injections with saline (SAL) and muscimol (MUS) into the aDLS (A) or pVLS (B) were performed after the success rate had reached 60% (middle stage) and 80% (late stage) for the first time through the training, respectively (A, Wilcoxon signed rank test, middle, Z = 65, p = 0.772, late, Z = 56.5, p = 0.242 for the aDLS; B, Wilcoxon signed rank test, middle, Z = 39, p = 1.000, late, Z = 43, p = 0.587). Data are indicated as the median and quartiles with the maximal and minimal values. 

Comment 9: Muscimol experiments. Can the authors comment on the effects on performance vs learning? What happens on the days after Muscimol? Does performance bounce back or is it still impaired?

We conducted a transient inhibition experiment with muscimol to examine whether the neuronal activity in the striatal subregions is linked with the processes at different stages. In this experiment, to lower the possibility that compensation of learning may occur during a session after the muscimol injection (Day N), we limited the session time to 15 min (45 trials) and evaluated the impact of the injection on the success rate at specific stages. The success rate in the muscimol-injected groups into the aDLS significantly decreased at the middle stage compared to the corresponding salineinjected groups, but not at the early and late stages (Figure 4C), and the rate in the muscimol groups into the pVLS significantly decreased at the late stage compared with the respective saline groups, but not at the early and middle stages (Figure 4D). Our results demonstrated that the aDLS and pVLS mainly function at the middle and late stages of the auditory discrimination task, respectively. 

In addition, we here reply to comment 10 as for the comparison of success rates before (Day N-1) and after (Day N+1) the injections (see Author response image 3). We focused on two injections into the aDLS at the middle stage and into the pVLS at the late stage, in which the rate was reduced soon after the muscimol injection on Day N. The success rate for the two injections showed no significant main effect regarding group (saline/muscimol) or day (Days N-1/N+1) and no significant interactions for group × day. Moreover, the success rate was not significantly increased on Day N+1 as compared to Day N-1, even in the saline-injected control group, probably because of the limited session time soon after the injection. Therefore, we consider that it was difficult to define the effects of drug injection on the learning of auditory discrimination in our behavioral protocol for the transient inhibition experiment, and that the reduced rates observed in the muscimol-injected group on Day N mostly reflect the impacts of muscimol at least partly on the performance of discriminative behavior. 

Author response image 3.

Comparison of success rate between days before (Day N1) and after (Day N+1) the injections into striatal subregions. Success rate in the saline (SAL)- and muscimol (MUS)-injected groups into the aDLS (A) or pVLS (B) at the early, middle, and late stages of auditory discrimination learning (two-way repeated ANOVA; early, day, F[1,14] = 5.266, p = 0.038, group, F[1,14] = 0.276, p = 0.608, day × group, F[1,14] = 0.118, p = 0.736; middle, day, F[1,14] = 4.110, p = 0.062, group, F[1,14] = 0.056, p = 0.816, day × group, F[1,14] = 1.150, p = 0.302; late, day, F[1,14] = 6.408, p = 0.024, group, F[1,14] = 0.229, p = 0.640, day × group, F[1,14] = 1.277, p = 0.278 for the aDLS; and early, day, F[1,10] = 0.115, p = 0.746, group, F[1,10] = 2.414, p = 0.151, day × group, F[1,10] = 0.157, p = 0.700; middle, day, F[1,10] = 0.278, p = 0.610, group, F[1,10] = 0.511, p = 0.491, day × group, F[1,10] = 4.144, p = 0.069; late, day, F[1,10] = 0.151, p = 0.705, group, F[1,10] = 0.719, p = 0.416, day × group, F[1,10] = 0.717, p = 0.417 for the pVLS). Data are indicated as the mean ± s.e.m.

Comment 10: Muscimol data has a pair before and after, can the authors show this comparison at early, middle, and late training? Not just the subtraction.

The comparison of success rates before and after drug injection is shown in Author response image 3.

Comment 11: Ephys recordings. These are complex figures and include a large number of acronyms. It would help to define them again and help the reader through these figures so the reader can focus on understanding the finding more than the figure presentation.

We replaced the abbreviations related to electrophysiological events (CO, CR, RS, and FL) with the original terms, and improved the explanation in the text and figures. 

Comment 12: Figure 7B/E - on correct trials, they see a difference in the cue response to high tone / low tone but no difference in the choice. This is the one that seemed like a topography issue.

The transient activity of cue onset-related neurons in the pVLS did not appear at the early stage of learning, but was observed in a learning-dependent manner (Figures 7A and S8E). In addition, the cue onset-HR activity showed a slight but notable difference between the HR and LL trials at the middle and late stages (Figure 7B), whereas there was no difference between activities in the HL and LR incorrect trials at the corresponding stages (Wilcoxon signed rank test; early, p = 0.375, middle, p = 0.931, and late, p = 0.668). These results suggest that the cue onset-related neurons in the pVLS represents the stimulus and response association (task contingency) rather than the topography of tone frequency.

Comment 13: Animals were normally trained for 60 minutes but on muscimol days only trained for 15 mins. On PET days only trained for 30 minutes. Ephys sessions were 60 mins. Is this correct? Why?

We determined the session time for each experiment by considering both technical and behavioral aspects. In the initial behavioral experiment, the session time was set to 60 min per day. Under this condition, the rats acquired the discrimination learning within 13 days. In the imaging experiment, the session without a PET scan was conducted for 60 min, while the session with a PET scan was carried out for 30 min as described previously (Cui et al, Neuroimage, 2015). This time schedule produced a learning curve similar to that of the initial behavioral experiment. In the transient inhibition experiment, the sessions without drug injections lasted for 60 min. As described in our response to the comment 2, the time of the session soon after the injection was limited to 15 min to lower the possibility of compensation of learning during the session. In the chronic lesion and electrophysiological experiments, all sessions were conducted for 60 min, corresponding to the initial experiment. 

References

Mizuma, H., Shukuri, M., Hayashi, T., Watanabe, Y. & Onoe, H. Establishment of in vivo brain imaging method in conscious mice. Journal of Nuclear Medicine 51, 10681075 (2010).

Cui, Y., et al. A voxel-based analysis of brain activity in high-order trigeminal pathway in the rat induced by cortical spreading depression. Neuroimage 108, 17-22 (2015).

Zimmer, E.R., et al. [18 F] FDG PET signal is driven by astroglial glutamate transport. Nat Neurosci 20, 393-395 (2017).

Valtcheva, S. & Venance, L. Astrocytes gate Hebbian synaptic plasticity in the striatum. Nature communications 7, 13845 (2016).

Gimenez T.L., Lorenc M., Jaramillo S. Adaptive categorization of sound frequency does not require the auditory cortex in rats. J Neurophysiol 114:1137-1145 (2015).

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public review):

Summary:

This study puts forth the model that under IFN-B stimulation, liquid-phase WTAP coordinates with the transcription factor STAT1 to recruit MTC to the promoter region of interferon-stimulated genes (ISGs), mediating the installation of m⁶A on newly synthesized ISG mRNAs. This model is supported by strong evidence that the phosphorylation state of WTAP, regulated by PPP4, is regulated by IFN-B stimulation, and that this results in interactions between WTAP, the m⁶A methyltransferase complex, and STAT1, a transcription factor that mediates activation of ISGs. This was demonstrated via a combination of microscopy, immunoprecipitations, m⁶A sequencing, and ChIP. These experiments converge on a set of experiments that nicely demonstrate that IFN-B stimulation increases the interaction between WTAP, METTL3, and STAT1, that this interaction is lost with the knockdown of WTAP (even in the presence of IFN-B), and that this IFN-B stimulation also induces METTL3-ISG interactions.

Strengths:

The evidence for the IFN-B stimulated interaction between METTL3 and STAT1, mediated by WTAP, is quite strong. Removal of WTAP in this system seems to be sufficient to reduce these interactions and the concomitant m⁶A methylation of ISGs. The conclusion that the phosphorylation state of WTAP is important in this process is also quite well supported.

Weaknesses:

The evidence that the above mechanism is fundamentally driven by different phase-separated pools of WTAP (regulated by its phosphorylation state) is weaker. These experiments rely relatively heavily on the treatment of cells with 1,6-hexanediol, which has been shown to have some off-target effects on phosphatases and kinases (PMID 33814344).

Given that the model invoked in this study depends on the phosphorylation (or lack thereof) of WTAP, this is a particularly relevant concern.

We are grateful for the reviewer’s positive comment and constructive feedback. 1,6-hexanediol (hex) was considered an inhibitor of hydrophobic interaction, thereby capable of dissolving protein phase separation condensates. Hex (5%-10% w/v) was still widely used to explore the phase separation characteristic and function on various protein, including some phosphorylated proteins such as pHSF1, or kinases such as NEMO1-3. Since hydrophobic interactions involved in various types of protein-protein interaction, the off-target effects of hex were inevitable. It has been reported that hex impaired RNA polymerase II CTD-specific phosphatase and kinase activity at 5% concentration4, which led to the reviewer’s concern. In response to this concern, we investigated the phosphorylation level of WTAP under hex concentration gradient treatment. Surprisingly, we found that both 2% and 5% hex maintained the PPP4c-mediated dephosphorylation level of WTAP but still led to the dissolution of WTAP LLPS condensates (Figure 5-figure supplement 1H; Author response image 1), indicating that hex dispersed WTAP phase separation in a phosphorylation-independent manner. Consistent with our findings, Ge et al. used 10% hex to dissolve WTAP phase separation condensates5. Additionally, we found the phosphorylation level of STAT1 was not affected by both 2% and 5% hex, revealing the off-target and impairment function of hex on kinases or phosphatases might not be universal (Figure 5-figure supplement 1H). Collectively, since the 5% hex we used did not influence the de-phosphorylation event of WTAP, function of WTAP LLPS mediating interaction between methylation complex and STAT1 revealed by our results was independent from its phosphorylation status.

Author response image 1.

mCherry-WTAP-rescued HeLa cells were treated with 10 ng/mL IFN-β together with or without 2% or 5% hex and 20 μg/mL digitonin for 1 hour or left untreated. Phase separation of mCherry-WTAP was observed through confocal microscopy. The number of WTAP condensates that diameter over 0.4 μm of n = 20 cells were counted through ImageJ and shown. Scale bars indicated 10 μm. All error bars, mean values ± SD, P-values were determined by unpaired two-tailed Student’s t-test of n = 20 cells in (B). For (A), similar results were obtained for three independent biological experiments.

Related to this point, it is also interesting (and potentially concerning for the proposed model) that the initial region of WTAP that was predicted to be disordered is in fact not the region that the authors demonstrate is important for the different phase-separated states.

A considerable number of proteins undergo phase separation via interactions between intrinsically disordered regions (IDRs). IDR contains more charged and polar amino acids to present multiple weakly interacting elements, while lacking hydrophobic amino acids to show flexible conformations6. In our study, we used PLAAC websites (http://plaac.wi.mit.edu/) to predict IDR domain of WTAP, and a fragment (234-249 amino acids) was predicted as prion-like domain (PLD). However, deletion of this fragment failed to abolish the phase separation properties of WTAP, which might be the main confusion to reviewers. To explain this issue, we checked the WTAP structure (within part of MTC complex) from protein data bank (https://www.rcsb.org/structure/7VF2) and found that the prediction of IDR has been renewed due to the update of different algorithm. IDR of WTAP expanded to 245-396 amino acids, encompassing the entire CTD region. Our results demonstrate that the CTD was critical for WTAP LLPS, as CTD deficiency significantly inhibited the formation of liquid condensates both in vitro and in cells. Also, phosphorylation sites on CTD were important for the phase transition of WTAP. These observations highlight the phosphorylation status on CTD region as a key driving force of phase separation, consistent with the established role of IDR in regulating LLPS. We have revised our description on WTAP IDR in our article following the reviewers’ suggestion.

Taking all the data together, it is also not clear to me that one has to invoke phase separation in the proposed mechanism.

In this article, we observed that WTAP underwent phase transition during virus infection and IFN-β treatment, and proposed a novel mechanism whereby post translational modification (PTM)-driven WTAP LLPS was required for the m⁶A modification of ISG mRNAs. To verify the hypothesis, we first demonstrated the relationship between PTM and phase separation of WTAP. We constructed WTAP 5ST-D and 5ST-A mutant to mimic WTAP phosphorylation and dephosphorylation status respectively, and figured out that dephosphorylated WTAP underwent LLPS. We also found that PPP4 was the main phosphatase to regulate WTAP dephosphorylation. To further investigation, we introduced the potent PPP4 inhibitor, fostriecin. Consistent with our findings in PPP4 deficient model, fostriecin treatment significantly inhibited the IFN-β-induced dephosphorylation of WTAP and disrupted its LLPS condensates, indicating that PPP4 was the key phosphatase recruited by IFN-β to regulate WTAP, confirming that PTM governs WTAP LLPS dynamics (Figure 2-figure supplement 1C and H). Furthermore, fostriecin-induced impairment of WTAP phosphorylation and phase separation correlated with reduced m⁶A modification level and elevated ISGs expression level (Figure 4C and Figure 4-figure supplement 1E). These findings together emphasized that dephosphorylation is required for WTAP LLPS.

As for the function of WTAP LLPS, we assumed that WTAP might undergo LLPS to sequester STAT1 together with m⁶A methyltransferase complex (MTC) for mediating m⁶A deposition on ISG mRNAs co-transcriptionally under IFN-β stimulation. Given that hex dissolved WTAP LLPS condensates without affecting dephosphorylation status (Figure 5-figure supplement 1H; Author response image 1), various experiments we performed previously actually confirmed the critical role of WTAP LLPS during m⁶A modification (Figure 4A), as well as the mechanism that WTAP phase separation enhanced the interaction between MTC and STAT1 (Figure 5E-F). Subsequent to reviewer’s comments, more experiments were conducted for further validation. We found the WTAP liquid condensates formed by wild type (WT) WTAP or WTAP 5ST-A mutant (which mimics dephosphorylated-WTAP) could be dissembled by hex, which also led to the impairment of the interaction between STAT1, METTL3 and WTAP (Figure 5-figure supplement 1E). In addition, in vitro experiments demonstrated that hex treatment significantly disrupted the interaction between recombinant GFP-STAT1 and mCherry-WTAP which expressed in the E. coli system (Figure 5F and Figure 5-figure supplement 1G). Notably, this prokaryotic expression system lacks endogenous phosphorylation machinery, resulting in non-phosphorylated mCherry-WTAP. For further validation, we performed the interaction of WTAP WT or 5ST-A with the promoter regions of ISG as well as the m⁶A modification of ISG mRNAs were attenuated by WTAP LLPS dissolution (Figure 4E and Figure 6E). These findings together revealed that WTAP LLPS were the critical mediators of the STAT1-MTC interactions, ISG promoter regions binding and the co-transcription m⁶A modification.

Collectively, our results demonstrated that IFN-β treatment recruited PPP4c to dephosphorylate WTAP, thereby driving the formation of WTAP liquid condensates which were essential for facilitating STAT1-MTC interaction and m⁶A deposition, subsequently mediated ISG expression. Since we revealed a strong association between PTM-regulated WTAP phase transition and its m⁶A modification function, WTAP LLPS was a novel and functionally distinct mechanism that must be reckoned with in this study.

Author response image 2.

Wild type (WT) WTAP or 5ST-A mutant-rescued WTAP^sgRNA THP-1-derived macrophages are stimulated with or without with 10 ng/mL IFN-β together followed with 2% or 5% 1,6-hexanediol (hex) and 20 μg/mL digitonin for 1 hour or left untreated. antibody and imaged using confocal microscope. Scale bars indicated 10 μm.

Reviewer #2 (Public review):

In this study, Cai and colleagues investigate how one component of the m⁶A methyltransferase complex, the WTAP protein, responds to IFNb stimulation. They find that viral infection or IFNb stimulation induces the transition of WTAP from aggregates to liquid droplets through dephosphorylation by PPP4. This process affects the m⁶A modification levels of ISG mRNAs and modulates their stability. In addition, the WTAP droplets interact with the transcription factor STAT1 to recruit the methyltransferase complex to ISG promoters and enhance m⁶A modification during transcription. The investigation dives into a previously unexplored area of how viral infection or IFNb stimulation affects m⁶A modification on ISGs. The observation that WTAP undergoes a phase transition is significant in our understanding of the mechanisms underlying m⁶A's function in immunity. However, there are still key gaps that should be addressed to fully accept the model presented.

Major points:

(1) More detailed analyses on the effects of WTAP sgRNA on the m⁶A modification of ISGs:

a. A comprehensive summary of the ISGs, including the percentage of ISGs that are m⁶A-modified. merip-isg percentage

b. The distribution of m⁶A modification across the ISGs. Topology

c. A comparison of the m⁶A modification distribution in ISGs with non-ISGs. Topology

In addition, since the authors propose a novel mechanism where the interaction between phosphorylated STAT1 and WTAP directs the MTC to the promoter regions of ISGs to facilitate co-transcriptional m⁶A modification, it is critical to analyze whether the m⁶A modification distribution holds true in the data.

We appreciate the reviewer’s summary of our manuscript and the constructive assessment. We have conducted the related analysis accordingly to present the m⁶A modification in ISGs in our model as reviewers suggested. Our results showed that about 64.29% of core ISGs were m⁶A modified under IFN-β stimulation (Figure 3-figure supplement 1B; Figure 3G), which was consistent with the similar percentage in previous studies[7,8]. The m⁶A distribution of the ISGs transcripts including IFIT1, IFIT2, OAS1 and OAS2 was validated (Figure 3-figure supplement 1H).

Generally, m⁶A deposition preferentially located in the vicinity of stop codon, while m⁶A modification was highly dynamic under different cellular condition. However, we compared the topology of m⁶A deposition of ISGs with non-ISGs, and found that m⁶A modification of ISG mRNAs showed higher preference of coding sequences (CDS) localization compared to global m⁶A deposition (Figure 3H). Similarly, various researches uncovered the m⁶A deposition regulated by co-transcriptionally m⁶A modification preferred to locate in the CDS [9-11]. Since our results of m⁶A modification distribution of ISGs was consistent with the co-transcriptional m⁶A modification characteristics, we believed that our hypothesis and results were correlated and consistent.

(2) Since a key part of the model includes the cytosol-localized STAT1 protein undergoing phosphorylation to translocate to the nucleus to mediate gene expression, the authors should focus on the interaction between phosphorylated STAT1 and WTAP in Figure 4, rather than the unphosphorylated STAT1. Only phosphorylated STAT1 localizes to the nucleus, so the presence of pSTAT1 in the immunoprecipitate is critical for establishing a functional link between STAT1 activation and its interaction with WTAP.

Thank you for the constructive comments. As we showed in Figure 4, we found the enhanced interaction between phase-separated WTAP and the nuclear-translocated STAT1 which were phosphorylated under IFN-β stimulation, indicating the importance of phosphorylation of STAT1. We repeated the immunoprecipitation experiments to clarify the function of pSTAT1 in WTAP interaction. Our results showed that IFN-β stimulation induced the interaction of WTAP with both pSTAT1 and STAT1 (Figure 5-figure supplement 1C), indicating that most of the WTAP-associated STAT1 was phosphorylated. Taken together, our data proved that LLPS WTAP bound with nuclear-translocated pSTAT1 under IFN-β stimulation.

(3) The authors should include pSTAT1 ChIP-seq and WTAP ChIP-seq on IFNb-treated samples in Figure 5 to allow for a comprehensive and unbiased genomic analysis for comparing the overlaps of peaks from both ChIP-seq datasets. These results should further support their hypothesis that WTAP interacts with pSTAT1 to enhance m⁶A modifications on ISGs.

Thank you for raising this thoughtful comment. In this study, MeRIP-seq and RNA-seq along with immunoprecipitation and immunofluorescence experiments supported that phase transition of WTAP enhanced its interaction to pSTAT1, thus mediating ISGs m⁶A modification and expression by enhancing its interaction with nuclear-translocated pSTAT1 during virus infection and IFN-β stimulation. However, how WTAP-mediated m⁶A modification was related to STAT1-mediated transcription remained unclear. Several researches have revealed the recruitment of m⁶A methyltransferase complex (MTC) to transcription start sites (TSS) of coding genes and R-loop structure by interacting with transcriptional factors STAT5B, SMAD2/3, DNA helicase DDX21, indicating the engagement of MTC mediated m⁶A modification on nascent transcripts at the very beginning of transcription [11-13]. These researches inspired us that phase-separated WTAP could be recruited to the ISG promoter regions via interacting with nuclear-translocated pSTAT1. To validate this mechanism, we have conducted ChIP-qPCR experiments targeting STAT1 and WTAP, revealed that IFN-β induced the comparable recruitment dynamics of both STAT1 and WTAP to ISG promoter regions (Figure 6A-B). Notably, STAT1 deficiency significantly abolished the bindings between WTAP and ISG promoter regions (Figure 6C). These findings established nuclear-translocated STAT1-dependent recruitment of phase separated WTAP and ISG promoter region, substantiated our hypothesis within the current study. We totally agree that ChIP-seq data will be more supportive to explore the mechanism in depth. We will continuously focus on the whole genome chromatin distribution of WTAP and explore more functional effect of transcriptional factor-dependent WTAP-promoter regions interaction in the future.

Minor points:

(1) Since IFNb is primarily known for modulating biological processes through gene transcription, it would be informative if the authors discussed the mechanism of how IFNb would induce the interaction between WTAP and PPP4.

Protein phosphatase 4 (PPP4) is a multi-subunit serine/threonine phosphatase complex that participates in diverse biologic process, including DDR, cell cycle progression, and apoptosis[14]. Several signaling pathway such as NF-κB and mTOR signaling, can be regulated by PPP4. Previous research showed that deficiency of PPP4 enhanced IFN-β downstream signaling and ISGs expression, which was consistent with our findings that knockdown of PPP4C impaired WTAP-mediated m⁶A modification, enhanced the ISGs expression[15,16]. Since there was no significant enhancement in PPP4 expression level during 0-3 hours of IFN-β stimulation in our results, we explored the PTM that may influence the protein-protein interaction, such as ubiquitination. Intriguingly, we found the ubiquitination level of PPP4 was enhanced after IFN-β stimulation, which may affect the interaction between PPP4 and WTAP (Author response image 3). Further investigation between PPP4 and WTAP will be conducted in our future study.

Author response image 3.

HEK 293T transfected with HA-ubiquitin (HA-Ub) and Flag-PPP4 were treated with 10 ng/mL IFN-β or left untreated. Whole cell lysate (WCL) was collected and immunoprecipitation (IP) experiment using anti-Flag antibody was performed, followed with immunoblot. Similar results were obtained for three independent biological experiments.

(2) The authors should include mCherry alone controls in Figure 1D to demonstrate that mCherry does not contribute to the phase separation of WTAP. Does mCherry have or lack a PLD?

According to the crystal structure of mCherry protein in protein structure database RCSB-PDB, mCherry protein presents as a β-barrel protein, with no IDRs or multivalent interaction domains including PLD, indicating that mCherry protein has no capability to undergo phase separation. This characteristic makes it a suitable protein to tag and trace the localization or expression levels of proteins, and a negative control for protein phase separation studies. As the reviewer suggested, we include mCherry alone controls in the revised version (Figure 1D).

(3) The authors should clarify the immunoprecipitation assays in the methods. For example, the labeling in Figure 2A suggests that antibodies against WTAP and pan-p were used for two immunoprecipitations. Is that accurate?

Due to the lack of phosphorylated-WTAP antibody, the detection of phosphorylation of WTAP was conducted by WTAP-antibody-mediated immunoprecipitation. We conducted immunoprecipitation to pull-down WTAP protein by WTAP antibody, then used antibody against phosphoserine/threonine/tyrosine (pan-p) to detect the phosphorylation level of WTAP. To avoid the misunderstanding, we have modified the description from pan-p to pWTAP (pan-p) in figures and revised the figure legends.

(4) The authors should include overall m⁶A modification levels quantified of GFP^sgRNA and WTAP^sgRNA cells, either by mass spectrometry (preferably) or dot blot.

We thank reviewer for raising these useful suggestions. As we showed in Figure 3F and J-K, the m⁶A modification event and degradation of ISG mRNAs were significantly depleted in WTAP^sgRNA cell lines, indicating that function of WTAP was deficient. Thus, we used the WTAP^sgRNA #2 cell line to perform most of our experiment. Furthermore, we also found 46.4% of global m⁶A modification was decreased in WTAP^sgRNA THP-1 cells than that of control cells with or without IFN-β stimulation (Author response image 4), further validating that level of m⁶A modification was significantly affected in WTAP^sgRNA cells. Taken together, our data confirmed that m⁶A methyltransferase activity was efficiently inhibited in our WTAP^sgRNA cells.

Author response image 4.

Control (GFP^sgRNA) and WTAP^sgRNA #2 THP-1-derived macrophages were treated with 10 ng/mL IFN-β for 4 hours. Global m⁶A level was detected and quantified through ELISA assays. All error bars, mean values ± SEM, P-values were determined by two-way ANOVA test independent biological experiments.

Reviewer #3 (Public review):

Summary:

This study presents a valuable finding on the mechanism used by WTAP to modulate the IFN-β stimulation. It describes the phase transition of WTAP driven by IFN-β-induced dephosphorylation. The evidence supporting the claims of the authors is solid, although major analysis and controls would strengthen the impact of the findings. Additionally, more attention to the figure design and to the text would help the reader to understand the major findings.

Strength:

The key finding is the revelation that WTAP undergoes phase separation during virus infection or IFN-β treatment. The authors conducted a series of precise experiments to uncover the mechanism behind WTAP phase separation and identified the regulatory role of 5 phosphorylation sites. They also succeeded in pinpointing the phosphatase involved.

Weaknesses:

However, as the authors acknowledge, it is already widely known in the field that IFN and viral infection regulate m⁶A mRNAs and ISGs. Therefore, a more detailed discussion could help the reader interpret the obtained findings in light of previous research.

We are grateful for the positive comments and the unbiased advice by the reviewer. To interpret our findings in previous research, we have revised the manuscript carefully and added more detailed discussion on ISG mRNAs m⁶A modification during virus infection or IFN stimulation.

It is well-known that protein concentration drives phase separation events. Similarly, previous studies and some of the figures presented by the authors show an increase in WTAP expression upon IFN treatment. The authors do not discuss the contribution of WTAP expression levels to the phase separation event observed upon IFN treatment. Similarly, METTL3 and METTL14, as well as other proteins of the MTC are upregulated upon IFN treatment. How does the MTC protein concentration contribute to the observed phase separation event?

We thank reviewer for pointing out the importance of the concentration dependent phase transition. Previously, Ge et al. discovered that expression level of WTAP was up-regulated during LPS stimulation, thereby promoting WTAP phase separation ability and m⁶A modification, indicating that WTAP concentration indeed affects the phase separation event. In our article, we have generated the phase diagram with different concentration of recombinant mCherry-WTAP in vitro (Figure 1-figure supplement 1C). For in cells experiments, we constructed the TRE-mCherry-WTAP HeLa cells in which the expression of mCherry-WTAP was induced by doxycycline (Dox) in a dose-dependent manner (Author response image 5A). We detected the LLPS of mCherry-WTAP at different concentrations by increasing the doses of Dox, and found that WTAP automatically underwent LLPS only in an artificially high expression level (Author response image 5B). However, the cells we used to detect the WTAP phase separation in our article was mCherry-WTAP-rescued HeLa cells, in which mCherry-WTAP was introduced in WTAP^sgRNA HeLa cells to stably express mCherry-WTAP. We had adjusted and verified the mCherry-WTAP expression level precisely to make the protein abundance similar to endogenous WTAP in wild type (WT) HeLa cells (Author response image 5C). We also repeated the IFN-β stimulation experiments and found no significant increase of WTAP protein level (Figure 5-figure supplement 1A). These findings indicated the phase separation of WTAP in our article was not artificially induced due to the extremely high protein expression level.

MTC protein expression level was crucial for m⁶A modification during virus infection event. Rubio et al. and Winkler et al. revealed that WTAP, METTL3 and METTL14 were up-regulated after 24 hours of HCMV infection[8,17]. Recently, Ge et al. proposed that WTAP protein was degraded after 12 hours of VSV and HSV stimulation5,18. However, these studies only focused on the virus infection event, how the MTC protein expression change after direct IFN-β stimulation was still unclear. Our study investigated the transition change of WTAP under IFNβ stimulation in a short time, we detected the expression level of MTC proteins within 6 hours of IFN-β treatment, and found no significant enhancement of WTAP, METTL3 or METTL14 protein level and mRNA level (Figure 5-figure supplement 1B and Figure 5-figure supplement 1A;). Our in vitro experiments showed that introducing CFP-METTL3 protein have no significant influence on WTAP phase separation (Figure 4H). Additionally, we performed in cells experiments and found that increased expression of METTL3 had no effect on WTAP phase separation event (Author response image 5D). Taken together, WTAP phase separation can be promoted by dramatically increased concentration of WTAP both in vitro and in cells, but the phase separation of WTAP under IFN-β stimulation in our study was not related with the expression level of MTC proteins.

Author response image 5.

(A) Immunoblot analysis of the expression of mCherry-WTAP in TRE-mCherry-WTAP HeLa cells treated with different doses of doxycycline (Dox). Protein level of mCherry-WTAP was quantified and normalized to β-actin of n=3 independent biological experiments. Results were obtained for three independent biological experiments. (B) Phase separation diagram of mCherry-WTAP in TRE-mCherry-WTAP HeLa cells treated with different doses of Dox were observed through confocal microscopy. Representative images for three independent biological experiments were shown in b while number of WTAP condensates that diameter over 0.4 μm of n=80 cells were counted and shown as medium with interquartile range. Dotted white lines indicated the location of nucleus. Scale bars indicated 10 μm. (C) Immunoblot analysis of the expression of endogenous WTAP in wildtype (WT) HeLa cells and mCherry-WTAP-rescued WTAP^sgRNA HeLa cells. (D) mCherry-WTAP-rescued HeLa cells were transfected with 0, 200 or 400 ng of Flag-METTL3, followed with 10 ng/mL IFN-β for 1 hour or left untreated (UT). Phase separation of mCherry-WTAP was observed through confocal microscopy. The number of WTAP condensates that diameter over 0.4 μm of n = 20 cells were counted through ImageJ and shown. Representative images of n=20 cells were shown. All error bars, mean values ± SD were determined by unpaired two-tailed Student’s t-test of n = 3 independent biological experiments in (A). For (A, C), similar results were obtained for three independent biological experiments.

How is PP4 related to the IFN signaling cascade?

Both reviewer #2 and reviewer #3 raised a similar point on the relationship between PPP4 and IFN signaling. In short, protein phosphatase 4 (PPP4) participates in diverse biologic process, including DDR, cell cycle progression and apoptosis14 and several signaling pathway. Previous research showed that deficiency of PPP4 enhanced IFN-β downstream signaling and ISGs expression, which was consistent with our findings that knockdown of PPP4C impaired WTAP-mediated m⁶A modification, and enhanced the ISGs expression[15,16]. Since there was no significant enhancement in PPP4C expression level during 0-3 hours of IFN-β stimulation in our results, we tried to explore the post-translation modification which may influence the protein-protein interaction, such as ubiquitination. Intriguingly, we found the ubiquitination level of PPP4 was enhanced after IFN-β stimulation, which may affect the interaction between PPP4 and WTAP (Author response image 4). Investigation between PPP4 and WTAP will be conducted in our further study (also see minor points 1 of reviewer#2).

In general, it is very confusing to talk about WTAP KO as WTAPgRNA.

As we describe above, all WTAP-deficient THP-1 cells were generated using the CRISPR-Cas9 system with WTAP-specific sgRNA, and used bulk cells instead of the monoclonal knockout cell for further experiments. Since monoclonal knockout cells were not obtained, we refer it as WTAP^sgRNA THP-1 cells rather than WTAP-KO THP-1 cells. We confirmed that WTAP expression was efficiently knocked down in WTAP^sgRNA THP-1 cells, and the m⁶A modification level was significantly impaired (Figure 3I, Figure 3-figure supplement 1A and Author response image 4), which was suitable for mechanism investigation.

Recommendations for the authors:

Reviewer #1 (Recommendations for the authors):

Related to the points raised in 'weaknesses' above, if the authors want to claim that this mechanism is reliant on WTAP phase-separated states, additional controls should be done to demonstrate this. Based on the available data it seems that it is just as likely that the phosphorylation state of WTAP is mediating interactions with other factors and/or altering its subcellular localization, which may or may not be related to phase separation.

We are grateful for the constructive suggestions. As we showed in , Figure 5-figure supplement 1H; Author response image 1 along with the explanation above, 5% hex dispersed the phase separation condensates of WTAP without affecting its phosphorylation status, proving the interaction between STAT1 and methylation complex impaired by hex was depended on WTAP LLPS but not its phosphorylation status (Figure 5E-H). To further confirmed the recruitment of WTAP LLPS on ISG promoter region, we performed the immunoprecipitation and ChIP-qPCR experiments of wild type (WT) WTAP, 5ST-D and 5ST-A rescued THP-1 cells. Our results uncovered the interaction between de-phosphorylated-mimic WTAP mutant and STAT1, and its binding ability with ISG promoter regions were depleted by hex without affecting its phosphorylation status (Author response image 2, Figure 5-figure supplement 1 F, Figure 6E). Taken together, we identified the de-phosphorylation event that regulated phase transition of WTAP during IFN-β stimulation, and proposed that LLPS of WTAP mediated by dephosphorylation was the key mechanism to bind with STAT1 and mediate the m⁶A modification on ISG mRNAs.

Reviewer #2 (Recommendations for the authors):

The author order is different for the main text and the supplementary file.

Thank you for pointing out this mistake. We have corrected it in our revised version.

Reviewer #3 (Recommendations for the authors):

Signaling molecules? Do you mean RNA or protein post-translational modification?

Thank you for pointing out this problem. In this sentence, we mean the post-translational modification of signaling proteins. We have corrected this mistake in our revised version.

Line 145: Do you mean Figure 1C?

We have corrected it in our revised version.

Line 214: Are the cells KO for WTAP? Do you mean CRISPR KO? In general, it is easier to present WTAPgRNA as WTAPKO.

Thank you for the constructive suggestion. As we explained above, in this project, all WTAP-deficient THP-1 cells were generated using the CRISPR-Cas9 system with WTAP-specific sgRNA, and used bulk cells instead of the monoclonal knockout cells. We confirmed that WTAP expression was efficiently knocked down in WTAP^sgRNA THP-1 cells, and the m⁶A modification level was significantly impaired (Figure 3I, Figure3-figure supplement 1A and Author response image 4). Since monoclonal knockout cells were not obtained, we refer it as WTAP^sgRNA THP-1 cells rather than WTAP-KO THP-1 cells.

Line 221: WTAP KO has no effect on the expression level of transcription factors?

Thank you for pointing out the uncritical expression. We have corrected this in our revised version.

Figure 3C: I would suggest removing the tracks and showing the expression levels in TPMs.

According to the reviewer’s suggestion, we have analyzed the results and showed the ISGs expression levels in fold change of TPMs (Figure 3D).

Figure 4C: It seems that the IP efficiency from METTL3 is lower in WTAP KO cells. That may impact the author's conclusions.

We have repeated the immunoprecipitation experiments of METTL3 and confirmed the immunoprecipitation (IP) efficiency from METTL3 had no difference between WTAP^sgRNA cells and the control cells (Figure 5C).

I would suggest performing an IP of WTAP with the 5StoA mutation to confirm the missing interaction with WTAP.

According to the reviewer’s suggestion, we investigated the interaction between STAT1 and WTAP in WT cells and WTAP 5ST-A-rescued THP-1 cells. Our results showed that interaction between STAT1, METTL3 and WTAP were enhanced with WTAP 5ST-A mutation, which was depleted by hex treatment (Figure 5-figure supplement 1E). Thus, the interaction of WTAP WT or 5ST-A with the promoter regions of ISG were attenuated by WTAP LLPS dissolution (Figure 6E). Taken together, the interaction between STAT1 and MTC were relied on LLPS of WTAP.

In the graphical abstract, it is confusing to represent WTAP as a line. All other proteins are presented as circles. It is easy to confuse WTAP protein with an RNA. Additionally, m⁶A is too big in size. It should be smaller than a protein.

We thank the reviewer for raising this suggestion. We have modified the graphical abstract to avoid the confusion in our revised version (Figure 6F).

References

(1) Wegmann, S., Eftekharzadeh, B., Tepper, K., Zoltowska, K.M., Bennett, R.E., Dujardin, S., Laskowski, P.R., MacKenzie, D., Kamath, T., Commins, C., et al. (2018). Tau protein liquid-liquid phase separation can initiate tau aggregation. The EMBO journal 37. 10.15252/embj.201798049.

(2) Lu, Y., Wu, T., Gutman, O., Lu, H., Zhou, Q., Henis, Y.I., and Luo, K. (2020). Phase separation of TAZ compartmentalizes the transcription machinery to promote gene expression. Nat Cell Biol 22, 453-464. 10.1038/s41556-020-0485-0.

(3) Zhang, H., Shao, S., Zeng, Y., Wang, X., Qin, Y., Ren, Q., Xiang, S., Wang, Y., Xiao, J., and Sun, Y. (2022). Reversible phase separation of HSF1 is required for an acute transcriptional response during heat shock. Nat Cell Biol 24, 340-352. 10.1038/s41556-022-00846-7.

(4) Duster, R., Kaltheuner, I.H., Schmitz, M., and Geyer, M. (2021). 1,6-Hexanediol, commonly used to dissolve liquid-liquid phase separated condensates, directly impairs kinase and phosphatase activities. J Biol Chem 296, 100260. 10.1016/j.jbc.2021.100260.

(5) Ge, Y., Chen, R., Ling, T., Liu, B., Huang, J., Cheng, Y., Lin, Y., Chen, H., Xie, X., Xia, G., et al. (2024). Elevated WTAP promotes hyperinflammation by increasing m⁶A modification in inflammatory disease models. J Clin Invest 134. 10.1172/JCI177932.

(6) Hou, S., Hu, J., Yu, Z., Li, D., Liu, C., and Zhang, Y. (2024). Machine learning predictor PSPire screens for phase-separating proteins lacking intrinsically disordered regions. Nat Commun 15, 2147. 10.1038/s41467-024-46445-y.

(7) McFadden, M.J., McIntyre, A.B.R., Mourelatos, H., Abell, N.S., Gokhale, N.S., Ipas, H., Xhemalce, B., Mason, C.E., and Horner, S.M. (2021). Post-transcriptional regulation of antiviral gene expression by N6-methyladenosine. Cell Rep 34, 108798. 10.1016/j.celrep.2021.108798.

(8) Winkler, R., Gillis, E., Lasman, L., Safra, M., Geula, S., Soyris, C., Nachshon, A., Tai-Schmiedel, J., Friedman, N., Le-Trilling, V.T.K., et al. (2019). m(6)A modification controls the innate immune response to infection by targeting type I interferons. Nat Immunol 20, 173-182. 10.1038/s41590-018-0275-z.

(9) Li, Y., Xia, L., Tan, K., Ye, X., Zuo, Z., Li, M., Xiao, R., Wang, Z., Liu, X., Deng, M., et al. (2020). N(6)-Methyladenosine co-transcriptionally directs the demethylation of histone H3K9me2. Nat Genet 52, 870-877. 10.1038/s41588-020-0677-3.

(10) Huang, H., Weng, H., Zhou, K., Wu, T., Zhao, B.S., Sun, M., Chen, Z., Deng, X., Xiao, G., Auer, F., et al. (2019). Histone H3 trimethylation at lysine 36 guides m(6)A RNA modification co-transcriptionally. Nature 567, 414-419. 10.1038/s41586-019-1016-7.

(11) Barbieri, I., Tzelepis, K., Pandolfini, L., Shi, J., Millan-Zambrano, G., Robson, S.C., Aspris, D., Migliori, V., Bannister, A.J., Han, N., et al. (2017). Promoter-bound METTL3 maintains myeloid leukaemia by m(6)A-dependent translation control. Nature 552, 126-131. 10.1038/nature24678.

(12) Hao, J.D., Liu, Q.L., Liu, M.X., Yang, X., Wang, L.M., Su, S.Y., Xiao, W., Zhang, M.Q., Zhang, Y.C., Zhang, L., et al. (2024). DDX21 mediates co-transcriptional RNA m(6)A modification to promote transcription termination and genome stability. Mol Cell 84, 1711-1726 e1711. 10.1016/j.molcel.2024.03.006.

(13) Bhattarai, P.Y., Kim, G., Lim, S.C., and Choi, H.S. (2024). METTL3-STAT5B interaction facilitates the co-transcriptional m(6)A modification of mRNA to promote breast tumorigenesis. Cancer Lett 603, 217215. 10.1016/j.canlet.2024.217215.

(14) Dong, M.Z., Ouyang, Y.C., Gao, S.C., Ma, X.S., Hou, Y., Schatten, H., Wang, Z.B., and Sun, Q.Y. (2022). PPP4C facilitates homologous recombination DNA repair by dephosphorylating PLK1 during early embryo development. Development 149. 10.1242/dev.200351.

(15) Zhan, Z., Cao, H., Xie, X., Yang, L., Zhang, P., Chen, Y., Fan, H., Liu, Z., and Liu, X. (2015). Phosphatase PP4 Negatively Regulates Type I IFN Production and Antiviral Innate Immunity by Dephosphorylating and Deactivating TBK1. J Immunol 195, 3849-3857. 10.4049/jimmunol.1403083.

(16) Raja, R., Wu, C., Bassoy, E.Y., Rubino, T.E., Jr., Utagawa, E.C., Magtibay, P.M., Butler, K.A., and Curtis, M. (2022). PP4 inhibition sensitizes ovarian cancer to NK cell-mediated cytotoxicity via STAT1 activation and inflammatory signaling. J Immunother Cancer 10. 10.1136/jitc-2022-005026.

(17) Rubio, R.M., Depledge, D.P., Bianco, C., Thompson, L., and Mohr, I. (2018). RNA m(6) A modification enzymes shape innate responses to DNA by regulating interferon beta. Genes Dev 32, 1472-1484. 10.1101/gad.319475.118.

(18) Ge, Y., Ling, T., Wang, Y., Jia, X., Xie, X., Chen, R., Chen, S., Yuan, S., and Xu, A. (2021). Degradation of WTAP blocks antiviral responses by reducing the m(6) A levels of IRF3 and IFNAR1 mRNA. EMBO Rep 22, e52101. 10.15252/embr.202052101.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Summary:

In this study, the authors engineer the endogenous left boundary of the Drosophila eve TAD, replacing the endogenous Nhomie boundary by either a neutral DNA, a wildtype Nhomie boundary, an inverted Nhomie boundary, or a second copy of the Homie boundary. They perform Micro-C on young embryos and conclude that endogenous Nhomie and Homie boundaries flanking eve pair with head-to-tail directionality to form a chromosomal stem loop. Abrogating the Nhomie boundary leads to ectopic activation of genes in the former neighboring TAD by eve embryonic stripe enhancers. Replacing Nhomie by an inverted version or by Homie (which pairs with itself head-to-head) transformed the stem loop into a circle loop. An important finding was that stem and circle loops differentially impact endogenous gene regulation both within the eve TAD and in the TADs bracketing eve. Intriguingly, an eve TAD with a circle loop configuration leads to ectopic activation of flanking genes by eve enhancers - indicating compromised regulatory boundary activity despite the presence of an eve TAD with intact left and right boundaries.

Strengths:

Overall, the results obtained are of high-quality and are meticulously discussed. This work advances our fundamental understanding of how 3D genome topologies affect enhancer-promoter communication.

Weaknesses:

Though convincingly demonstrated at eve, the generalizability of TAD formation by directional boundary pairing remains unclear, though the authors propose this mechanism could underly the formation of all TADs in Drosophila and possibly even in mammals. Strong and ample evidence has been obtained to date that cohesin-mediated chromosomal loop extrusion explains the formation of a large fraction of TADs in mammals. 

(1.1) The difficultly with most all of the studies on mammal TADs, cohesin and CTCF roadblocks is that the sequencing depth is not sufficient, and large bin sizes (>1 kb) are needed to visualize chromosome architecture.  The resulting contact profiles show TAD neighborhoods, not actual TADs.

The problem with these studies is illustrated by comparing the contact profiles of mammalian MicroC data sets at different bin sizes in Author response image 1.  In this figure, the darkness of the “pixels” in panels E, F, G and H was enhanced by reducing brightness in photoshop.

Author response image 1.

Mammalian MicroC profiles different bun sizes

Panels A and C show “TADs” using bin sizes typical of most mammalian studies (see Krietenstein et al. (2023) (Krietenstein et al. 2020)).  At this level of resolution, TADs, the “trees” that are the building blocks of chromosomes, are not visible.  Instead, what is seen are TAD neighborhoods or “forests”.  Each neighborhood consists of several dozen individual TADs.  The large bins in these panels also artificially accentuated TAD:TAD interactions, generating a series of “stripes” and “dots” that correspond to TADs bumping into each other and sequences getting crosslinked.  For example, in panel A there is prominent stripe on the edge of a “TAD” (blue arrow).  In panel C, this stripe resolves into a series of dots arranged as parallel, but interrupted “stripes” (green and blue arrows).  At the next level of resolution, it can be seen that the stripe marked by the blue arrow and magenta asterisk is generated by contacts between the left boundary of the TAD indicated by the magenta bar with sequences in a TAD (blue bar) ~180 kb way.  While dots and stripes are prominent features in contact profiles visualized with larger bin sizes (A and C), the actual TADs that are observed with a bin size of 200 bp (examples are underlined by black bars in panel G) are not bordered by stripes, nor are they topped by obvious dots.  The one possible exception is the dot that appears at the top of the volcano triangle underlined with magenta.

The chromosome 1 DNA segment from the MicroC data of Hseih et al. (2023) (Hsieh et al. 2020) shows a putative volcano triangle with a plume (indicated by a V in Author response image 1 panels D, F and H).  Sequences in the V TAD don’t crosslink with their immediate neighbors, and this gives a “plume” above the volcano triangle, as indicate by the light blue asterisk in panels D, F and H.  Interestingly the V TAD does contact two distant TADs, U on the left and W on the right. The U TAD is ~550 kb from V, and the region of contact is indicated by the black arrow.  The W TAD is ~585 kb from V, and the region of contact is indicated by the magenta arrow.  While the plume still seems to be visible with a bin size of 400 bp (light blue asterisk), it is hard to discern when the bin size is 200 bp, as there are not enough reads.

The evidence demonstrating that cohesin is required for TAD formation/maintenance is based on low resolution Hi-C data, and the effects that are observed are on TAD neighborhoods (forests) and not TADs (trees).  In fact, there is published evidence that cohesin is not required in mammals for TAD formation/maintenance.  In an experiment from Goel et al. 2023 the authors depleted the cohesin component Rad21 and then visualized the effects on TAD organization using the high resolution region capture MicroC (RCMC) protocol.  The MicroC contact map in this figure visualizes a ~250 kb DNA segment around the Ppm1pg locus at 250 bp resolution.  On the right side of the diagonal is the untreated control, while the left side shows the MicroC profile of the same region after Rad21 depletion.  The authors indicated that there was a 97% depletion of Rad21 in their experiment.  However, as is evident from a comparison of the experimental and control, loss of Rad21 has no apparent effect on the TAD organization of this mammalian DNA segment.

Several other features are worth noting.  First, unlike the MicroC experiments shown in Author response image 1, there are dots at the apex of the TADs in this chromosomal segment.  In the MicroC protocol, fixed chromatin is digested to mononucleosomes by extensive MNase digestion.  The resulting DNA fragments are then ligated, and dinucleosome-length fragments are isolated and sequenced. 

DNA sequences that are nucleosome free in chromatin (which would be promoters, enhancers, silencers and boundary elements) are typically digested to oligonucleotides in this procedure and won’t be recovered. This means that the dots shown here must correspond to mononucleosome-length elements that are MNase resistant.  This is also true for the dots in the MicroC contact profiles of the Drosophila Abd-B regulatory domain (see Fig. 2B in the paper).  Second, the TADs are connected to each other by 45o stripes (see blue and green arrowheads).  While it is not clear from this experiment whether the stipes are generated by an active mechanism (enzyme) or by some “passive” mechanism (e.g., sliding), the stripes in this chromosomal segment are not generated by cohesin, as they are unperturbed by Rad21 depletion.  Third, there are no volcano triangles with plumes in this chromosomal DNA segment.  Instead, the contact patterns (purple and green asterisks) between neighboring TADs closely resemble those seen for the Abd-B regulatory domains (compare Goel et al. 2023 with Fig. 2B in the paper).  This similarity suggests that the TADs in and around Ppm1g may be circle-loops, not stem-loops.  As volcano triangles with plumes also seem to be rare in the MicroC data sets of Krietenstein et al. (Krietenstein et al. 2020) and Hesih et al. (Hsieh et al. 2020) (with the caveat that these data sets are low resolution: see Author response image 1), it is possible that much of the mammalian genome is assembled into circle-loop TADs, a topology that can’t be generated by the cohesin loop extrusion (bolo tie clip) /CTCF roadblock model.

While Rad21 depletion has no apparent effect on TADs, it does appear to impact TAD neighborhoods.  This is in a supplemental figure in Goel et al. (Goel et al. 2023).  In this figure, TADs in the Ppm1g region of chromosome 5 are visualized with bin sizes of 5 kb and 1 kb.  A 1.2 Mb DNA segment is shown for the 5 kb bin size, while an 800 kb DNA segment is shown for the 1 kb bin size.  As can be seen from comparing the MicroC profiles in Author response image 2 with that in Goel et al. 2023, individual TADs are not visible.  Instead, the individual TADs are binned into large TAD “neighborhoods” that consist of several dozen or more TADs.

Unlike the individual TADs shown in Goel et al. 2023, the TAD neighborhoods in Author response image 2 are sensitive to Rad21 depletion.  The effects of Rad21 depletion can be seen by comparing the relative pixel density inside the blue lines before (above the diagonal) and after (below the diagonal) auxin-induced Rad21 degradation.  The reduction in pixel density is greatest for more distant TAD:TAD contacts (farthest from the diagonal).  By contrast, the TADs themselves are unaffected (Goel et al. 2023), as are contacts between individual TADs and their immediate neighbors.  In addition, contacts between partially overlapping TAD neighborhoods are also lost.  At this point it isn’t clear why contacts between distant TADs in the same neighborhood are lost when Rad21 is depleted; however, a plausible speculation is that it is related to the functioning of cohesin in holding newly replicated DNAs together until mitosis and whatever other role it might have in chromosome condensation.

Author response image 2.

Ppm1g full locus chr5

Moreover, given the unique specificity with which Nhomie and Homie are known to pair (and exhibit "homing" activity), it is conceivable that formation of the eve TAD by boundary pairing represents a phenomenon observed at exceptional loci rather than a universal rule of TAD formation. Indeed, characteristic Micro-C features of the eve TAD are only observed at a restricted number of loci in the fly genome…..

(1.2) The available evidence does not support the claim that nhomie and homie are “exceptional.”  To begin with, nhomie and homie rely on precisely the same set of factors that have been implicated in the functioning of other boundaries in the fly genome.  For example, homie requires (among other factors) the generic boundary protein Su(Hw) for insulation and long-distance interactions (Fujioka et al. 2024).  (This is also true of nhomie: unpublished data.)  The Su(Hw) protein (like other fly polydactyl zinc finger proteins) can engage in distant interactions.  This was first shown by Sigrist and Pirrotta (Sigrist and Pirrotta 1997), who found that the su(Hw) element from the gypsy transposon can mediate long-distance regulatory interactions (PRE dependent silencing) between transgenes inserted at different sites on homologous chromosomes (trans interactions) and at sites on different chromosomes.

The ability to mediate long-distance interactions is not unique to the su(Hw) element, or homie and nhomie.  Muller et al. (Muller et al. 1999) found that the Mcp boundary from the Drosophila BX-C is also able to engage in long-distance regulatory interactions—both PRE-dependent silencing of mini-white and enhancer activation of mini-white and yellow.  The functioning of the Mcp boundary depends upon two other generic insulator proteins, Pita and the fly CTCF homolog (Kyrchanova et al. 2017).  Like Su(Hw) both are polydactyl zinc finger proteins, and they resemble the mammalian CTCF protein in that their N-terminal domain mediates multimerization (Bonchuk et al. 2020; Zolotarev et al. 2016).  Figure 6 from Muller et el. 1999 shows PRE-dependent “pairing sensitive silencing” interactions between transgenes carrying a mini-white reporter, the Mcp and scs’ (Beaf dependent)(Hart et al. 1997) boundary elements, and a PRE closely linked to Mcp.  In this experiment flies homozygous for different transgene inserts were mated and the eye color was examined in their transheterozygous progeny.  As indicated in the figure, the strongest trans-silencing interactions were observed for inserts on the same chromosomal arm; however, transgenes inserted on the left arm of chromosome 3 can interact across the centromere with transgenes inserted on the right arm of chromosome 3. 

Figure 5C (left) from Muller et el. 1999 shows a trans-silencing interaction between w#11.102 at 84D and w#11.16 approximately 5.8 Mb away, at 87D.  Figure 5C (right) shows a trans-silencing interaction across the centromere between w#14.29 on the left arm of chromosome 3 at 78F and w#11.102 on the right arm of chromosome 3 at 84D. The eye color phenotype of mini-white-containing transgenes is usually additive: homozygyous inserts have twice as dark eye color as the corresponding hemizygous inserts.  Likewise, in flies trans-_heterozygous for _mini-white transgenes inserted at different sites, the eye color is equivalent to the sum of the two transgenes.  This is not true when mini-white transgenes are silenced by PREs.  In the combination shown in panel A, the t_rans-_heterozygous fly has a lighter eye color than either of the parents.  In the combination in panel B, the _trans-_heterozygous fly is slightly lighter than either parent.

As evident from the diagram in Figure 6 from Muller et el. 1999, all of the transgenes inserted on the 3rd chromosome that were tested were able to participate in long distance (>Mbs) regulatory interactions.  On the other hand, not all possible pairwise interactions are observed.  This would suggest that potential interactions depend upon the large scale (Mb) 3D folding of the 3rd chromosome.

When the scs boundary (Zw5 dependent) (Gaszner et al. 1999) was added to the transgene to give sMws’, it further enhanced the ability of distant transgenes to find each other and pair.  All eight of the sMws’ inserts that were tested were able to interact with at least one other sMws’ insert on a different chromosome and silence mini-white.  Vazquez et al. () subsequently tagged the sMws’ transgene with LacO sequences (ps0Mws’) and visualized pairing interactions in imaginal discs.  Trans-heterozygous combinations on the same chromosome were found paired in 94-99% of the disc nuclei, while a trans-heterozygous combination on different chromosomes was found paired in 96% of the nuclei (Table 3 from Vazquez et al. 2006).  Vazquez et al. also examined a combination of four transgenes inserted on the same chromosome (two at the same insertion site, and two at different insertion sites).  In this case, all four transgenes were clustered together in 94% of the nuclei (Table 3 from Vazquez et al. 2006).  Their studies also suggest that the distant transgenes remain paired for at least several hours.  A similar experiment was done by Li et al. (Li et al. 2011), except that the transgene contained only a single boundary, Mcp or Fab-7.  While pairing was still observed in trans-heterozygotes, the frequency was reduced without scs and scs’.

It is worth pointing out that there is no plausible mechanism in which cohesin could extrude a loop through hundreds of intervening TADs, across the centromere (ff#13.101_ßà_w#11.102: Figure 6 from Muller et el. 1999; w#14.29_ßà_w#11.02: Figure 6 from Muller et el. 1999 and 5) and come to a halt when it “encounters” Mcp containing transgenes on different homologs.  The same is true for Mcp-dependent pairing interactions in cis (Fig. 7 in Muller et al. (Muller et al. 1999)) or Mcp-dependent pairing interactions between transgenes inserted on different chromosomes (Fig. 8 in Muller et al. (Muller et al. 1999); Line 8 in Table 3 from Vazquez et al. 2006). 

These are not the only boundaries that can engage in long-distance pairing.  Mohana et al. (Mohana et al. 2023) identified nearly 60 meta-loops, many of which appear to be formed by the pairing of TAD boundary elements.  Two examples (at 200 bp resolution from 12-16 hr embryos) are shown in Author response image 3.

Author response image 3.

Metaloops on the 2nd and 3rd chromosomes: circle-loops and multiple stem-loops

One of these meta-loops (panel A) is generated by the pairing of two TAD boundaries on the 2nd chromosome.  The first boundary, blue, (indicated by blue arrow) is located at ~2,006, 500 bp between a small TAD containing the Nplp4 and CG15353 genes and a larger TAD containing 3 genes, CG33543, Obp22a and Npc2a.  Nplp4 encodes a neuropeptide.  The functions of CG15354 and CG33543 are unknown.  Obp22a encodes an odorant binding protein, while Npc2a encodes the Niemann-Pick type C-2a protein which is involved sterol homeostasis.  The other boundary (purple: indicated by purple arrow) is located between two TADs 2.8 Mb away at 4,794,250 bp.  The upstream TAD contains the fipi gene (CG15630) which has neuronal functions in male courtship, while the downstream TAD contains CG3294, which is thought to be a spliceosome component, and schlaff (slf) which encodes a chitin binding protein.  As illustrated in the accompanying diagram, the blue boundary pairs with the purple boundary in a head-to-head orientation, generating a ~2.8 Mb loop with a circle-loop topology.  As a result of this pairing, the multi-gene (CG33543, Obp22a and Npc2a) TAD upstream of the blue boundary interacts with the CG15630 TAD upstream of the purple boundary.  Conversely the small Nplp4:CG15353 TAD downstream of the blue boundary interacts with the CG3294:slf TAD downstream of the purple boundary.  Even if one imagined that the cohesin bolo tie clip was somehow able to extrude 2.8 Mb of chromatin and then know to stop when it encountered the blue and purple boundaries, it would’ve generated a stemloop, not a circle-loop.

The second meta-loop (panel B) is more complicated as it is generated by pairing interactions between four boundary elements.  The blue boundary (blue arrow) located ~4,801,800 bp (3L) separates a large TAD containing the RhoGEF64C gene from a small TAD containing CG7509, which encodes a predicted subunit of an extracellular carboxypeptidase.  As can be seen in the MicroC contact profile and the accompanying diagram, the blue boundary pairs with the purple boundary (purple arrow) which is located at ~7,013, 500 (3L) just upstream of the 2nd internal promoter (indicated by black arrowhead) of the Mp (Multiplexin) gene.  This pairing interaction is head-to-tail and generates a large stem-loop that spans ~2.2 Mb.  The stem-loop brings sequences upstream of the blue boundary and downstream of the purple boundary into contact (the strings below a bolo tie clip), just as was observed in the boundary bypass experiments of Muravyova et al. (Muravyova et al. 2001) and Kyrchanova et al. (Kyrchanova et al. 2008).  The physical interactions result in a box of contacts (right top) between sequences in the large RhoGEF64C TAD and sequences in a large TAD that contains an internal Mp promoter.  The second pairing interaction is between the brown boundary (brown arrow) and the green boundary (green arrow).  The brown boundary is located at ~4 805,600 bp (3L) and separates the TAD containing CG7590 from a large TAD containing CG1808 (predicted to encode an oxidoreductase) and the Dhc64C (Dynein heavy chain 64C) gene.  The green boundary is located at ~6,995,500 bp (3L), and it separates a TAD containing CG32388 and the biniou (bin) transcription factor from a TAD that contains the most distal promoter of the Mp (Multiplexin) gene (blue arrowhead).  As indicated in the diagram, the brown and green boundaries pair with each other head-to-tail, and this generates a small internal loop (and the final configuration would resemble a bolo tie with two tie clips).  This small internal loop brings the CG7590 TAD into contact with the TAD that extends from the distal Mp promoter to the 2nd internal Mp promoter.  The resulting contact profile is a rectangular box with diagonal endpoints corresponding to the paired blue:purple and brown:green boundaries.  The pairing of the brown:green boundaries also brings the TADs immediately downstream of the brown boundary and upstream of the green boundary into contact with each other, and this gives a rectangular box of interactions between the Dhc64C TAD, and sequences in the bin/CG3238 TAD.  This box is located on the lower left side of the contact map.

Since the bin and Mp meta-loops in Author response image 3B are stem-loops, they could have been generated by “sequential” cohesin loop extrusion events.  Besides the fact that cohesin extrusion of 2 Mb of chromatin and breaking through multiple intervening TAD boundaries challenges the imagination, there is no mechanism in the cohesion loop extrusion/CTCF roadblock model to explain why cohesion complex 1 would come to a halt at the purple boundary on one side and the blue boundary on the other, while cohesin complex 2 would instead stop when it hits the brown and green boundaries.  This highlights another problem with the cohesin loop extrusion/CTCF roadblock model, namely that the roadblocks are functionally autonomous: they have an intrinsic ability to block cohesin that is entirely independent of the intrinsic ability of other roadblocks in the neighborhood.  As a result, there is no mechanism for generating specificity in loop formation.  By contrast, boundary pairing interactions are by definition non-autonomous and depend on the ability of individual boundaries to pair with other boundaries: specificity is built into the model. The mechanism for pairing, and accordingly the basis for partner preferences/specificity, are reasonably well understood.  Probably the most common mechanism in flies is based on shared binding sites for architectural proteins that can form dimers or multimers (Bonchuk et al. 2021; Fedotova et al. 2017).  Flies have a large family of polydactyl zinc finger DNA binding proteins, and as noted above, many of these form dimers or multimers and also function as TAD boundary proteins.  This pairing principle was first discovered by Kyrchanova et al. (Kyrchanova et al. 2008).  This paper also showed that orientation-dependent pairing interactions is a common feature of endogenous fly boundaries.  Another mechanism for pairing is specific protein:protein interactions between different DNA binding factors (Blanton et al. 2003).  Yet a third mechanism would be proteins that bridge different DNA binding proteins together.  The boundaries that use these different mechanisms (BX-C boundaries, scs, scs’) depend upon the same sorts of proteins that are used by homie and nhomie.  Likewise, these same set of factors reappear in one combination or another in most other TAD boundaries.  As for the orientation of pairing interactions, this is most likely determined by the order of binding sites for chromosome architectural proteins in the partner boundaries.

…and many TADs lack focal 3D interactions between their boundaries.

(1.3) The idea that flies differ from mammals in that they “lack” focal 3D interactions is simply mistaken.  One of the problems with drawing this distinction is that most all of the “focal 3D interactions” seen mammalian Hi-C experiments are a consequence of binning large DNA segments in low resolution restriction enzyme-dependent experiments.  This is even true in the two “high” resolution MicroC experiments that have been published (Hsieh et al. 2020; Krietenstein et al. 2020).  As illustrated above in Author response image 1, most of the “focal 3D interactions” (the dots at the apex of TAD triangles) seen with large bin sizes (1 kb and greater) disappear when the bin size is 200 bp and TADs rather than TAD neighborhoods are being visualized.

As described in point #1.1, in the MicroC protocol, fixed chromatin is first digested to mononucloesomes by extensive MNase digestion, processed/biotinylated, and ligated to give dinucleosome-length fragments, which are then sequenced.  Regions of chromatin that are nucleosome free (promoters, enhancers, silencers, boundary elements) will typically be reduced to oligonucleotides in this procedure and will not be recovered when dinucleosome-length fragments are sequenced.  The loss of sequences from typical paired boundary elements is illustrated by the lar meta-loop shown in Author response image 4 (at 200 bp resolution).  Panels A and B show the contact profiles generated when the blue boundary (which separates two TADs that span  the Lar (Leukocyteantigen-related-like) transcription unit interacts with the purple boundary (which separates two TADs in a gene poor region ~620 kb away).  The blue and purple boundaries pair with each other head-to-head, and this pairing orientation generates yet another circle-loop.  In the circle-loop topology, sequences in the TADs upstream of both boundaries come into contact with each other, and this gives the small dark rectangular box to the upper left of the paired boundaries (Author response image 4A).  (Note that this small box corresponds to the two small TADs upstream of the blue and purple boundaries, respectively. See panel B.)  Sequences in the TADs downstream of the two boundaries also come into contact with each other, and this gives the large box to the lower right of the paired boundaries.  While this meta-loop is clearly generated by pairing interactions between the blue and purple boundaries, the interacting sequences are degraded in the MicroC protocol, and sequences corresponding to the blue and purple boundaries aren’t recovered.  This can be seen in panel B (red arrow and red arrowheads).  When a different Hi-C procedure is used (dHS-C) that captures nucleosome-free regions of chromatin that are physically linked to each other (Author response image 4C & D), the sequences in the interacting blue and purple boundaries are recovered and generate a prominent “dot” at their physical intersection (blue arrow in panel D).

Author response image 4.

Lar metaloop. Panels A & bB: MicroC. Panels C & D: dHS-C

While sequences corresponding to the blue and purple boundaries are lost in the MicroC procedure, there is at least one class of elements that engage in physical pairing interactions whose sequences are (comparatively) resistant to MNase digestion.  This class of elements includes many PREs ((Kyrchanova et al. 2018); unpublished data), the boundary bypass elements in the Abd-B region of BX-C (Kyrchanova et al. 2023; Kyrchanova et al. 2019a; Kyrchanova et al. 2019b; Postika et al. 2018), and “tethering” elements (Batut et al. 2022; Li et al. 2023).  In all of the cases tested, these elements are bound in nuclear extracts by a large (>1000 kD) GAGA factor-containing multiprotein complex called LBC.  LBC also binds to the hsp70 and eve promoters (unpublished data).  Indirect end-labeling experiments (Galloni et al. 1993; Samal et al. 1981; Udvardy and Schedl 1984) indicate that the LBC protects a ~120-180 bp DNA segment from MNase digestion.  It is likely that this is the reason why LBC-bound sequences can be recovered in MicroC experiments as dots when they are physically linked to each other.  One such example (based on the ChIP signatures of the paired elements) is indicated by the green arrow in panel B and D of Author response image 4.  Note that there are no dots corresponding to these two LBC elements within either of the TADs immediately downstream of the blue and purple boundaries.  Instead the sequences corresponding to the two LBC elements are only recovered when the two elements pair with each other over a distance of ~620 kb.  The fact that these two elements pair with each other is consistent with other findings which indicate that, like classical boundaries, LBC elements exhibit partner preferences.  In fact, LBC elements can sometimes function as TAD boundaries.  For example, the Fab-7 boundary has two LBC elements, and full Fab-7 boundary function can be reconstituted with just these two elements (Kyrchanova et al. 2018).

Reviewer #2 (Public Review):

"Chromatin Structure II: Stem-loops and circle-loops" by Ke*, Fujioka*, Schedl, and Jaynes reports a set of experiments and subsequent analyses focusing on the role of Drosophila boundary elements in shaping 3D genome structure and regulating gene expression. The authors primarily focus on the region of the fly genome containing the even skipped (eve) gene; eve is expressed in a canonical spatial pattern in fly embryos and its locus is flanked by the well-characterized neighbor of homie (nhomie) and homie boundary elements. The main focus of investigation is the orientation dependence of these boundary elements, which had been observed previously using reporter assays. In this study, the authors use Crispr/Cas9 editing followed by recombination-mediated cassette exchange to create a series of recombinant fly lines in which the nhomie boundary element is either replaced with exongenous sequence from phage 𝝀, an inversion of nhomie, or a copy of homie that has the same orientation as the endogenous homie sequence. The nhomie sequence is also regenerated in its native orientation to control for effects introduced by the transgenesis process.

The authors then perform high-resolution Micro-C to analyze 3D structure and couple this with fluorescent and colorimetric RNA in situ hybridization experiments to measure the expression of eve and nearby genes during different stages of fly development. The major findings of these experiments are that total loss of boundary sequence (replacement with 𝝀 DNA) results in major 3D structure changes and the most prominent observed gene changes, while inversion of the nhomie boundary or replacement with homie resulted in more modest effects in terms of 3D structure and gene expression changes and a distinct pattern of gene expression change from the 𝝀 DNA replacement. As the samples in which the nhomie boundary is inverted or replaced with homie have similar Micro-C profiles at the eve locus and show similar patterns of a spurious gene activation relative to the control, the observed effects appear to be driven by the relative orientation of the nhomie and homie boundary elements to one another.

Collectively, the findings reported in the manuscript are of broad interest to the 3D genome field. Although extensive work has gone into characterizing the patterns of 3D genome organization in a whole host of species, the underlying mechanisms that structure genomes and their functional consequences are still poorly understood. The perhaps best understood system, mechanistically, is the coordinated action of CTCF with the cohesin complex, which in vertebrates appears to shape 3D contact maps through a loop extrusion-pausing mechanism that relies on orientation-dependent sequence elements found at the boundaries of interacting chromatin loops.

(2.1) The notion that mammalian genome is shaped in 3D by the coordinate action of cohesin and CTCF has achieved the status of dogma in the field of chromosome structure in vertebrates.  However, as we have pointed out in #1.1, the evidence supporting this dogma is far from convincing.  To begin with, it is based on low resolution Hi-C experiments that rely on large bin sizes to visualize so-called “TADs.”  In fact, the notion that cohesin/CTCF are responsible on their own for shaping the mammalian 3D genome appears to be a result of mistaking a series of forests for the actual trees that populate each of the forests.

As illustrated in Author response image 1 above, the “TADs” that are visualized in these low resolution data sets are not TADs at all, but rather TAD neighborhoods consisting of several dozen or more individual TADs.  Moreover, the “interesting” features that are evident at low resolution (>1 kb)—the dots and stripes—largely disappear at resolutions appropriate for visualizing individual TADs (~200 bp).

In Goel et al. 2023, we presented data from one of the key experiments in Goel et al. (Goel et al. 2023).  In this experiment,  the authors used RCMC to generate high resolution (~250 bp) MicroC contact maps before and after Rad21 depletion.  Contrary to dogma, Rad21 depletion has absolutely no effect on TADs in a ~250 kb DNA segment—and these TADs look very much like the TADs we observe in the Drosophila genome, in particular in the Abd-B region of BX-C that is thought to be assembled into a series of circle-loops (see Fig. 2B).

While Goel et al. (Goel et al. 2023) observed no effect of Rad21 depletion on TADs, they found that loss of Rad21 disturbs long-distance (but not short-distance) contacts in large TAD neighborhoods when their RCMC data set is visualized using bin sizes of 5 kb and I kb.  This is shown in Author response image 2.  The significance of this finding is, however, uncertain.  It could mean that the 3D organization of large TAD neighborhoods have a special requirement for cohesin activity.  On the other hand, since cohesin functions to hold sister chromosomes together after replication until they separate during mitosis (and might also participate in mitotic condensation), it is also possible that the loss of long-range contacts in large TAD neighborhoods when Rad21 is depleted is simply a reflection of this particular activity.  Further studies will be required to address these possibilities.

As for CTCF: a careful inspection of the ChIP data in Goel et al. 2023 indicates that CTCF is not found at each and every TAD boundary.  In fact, the notion that CTCF is the be-all and end-all of TAD boundaries in mammals is truly hard to fathom.  For one, the demands for specificity in TAD formation (and in regulatory interactions) are likely much greater than those in flies, and specificity can’t be generated by a single DNA binding protein.  For another, several dozen chromosomal architectural proteins have already been identified in flies.  This means that (unlike what is thought to be true in mammals) it is possible to use a combinatorial mechanism to generate specificity in, for example, the long distance interactions in RFig 6 and 7.  As noted in #2.1 above, many of the known chromosomal architectural proteins in flies are polydactyl zinc finger proteins (just like CTCF).  There are some 200 different polydactyl zinc finger proteins in flies, and the function of only a hand full of these is known at present.  However, it seems likely that a reasonable fraction of this class of DNA binding proteins will ultimately turn out to have an architectural function of some type (Bonchuk et al. 2021; Fedotova et al. 2017).  The number of different polydactyl zinc finger protein genes in mammals is nearly 3 times that of flies.  It is really possible that of these, only CTCF is involved in shaping the 3D structure of the mammalian genome?

Despite having a CTCF paralog and cohesin, the Drosophila genome does not appear to be structure by loop extrusion-pausing. The identification of orientation-dependent elements with pronounced structural effects on genome folding thus may shed light on alternative mechanisms used to regulated genome structure, which in turn may yield insights into the significance of particular folding patterns.

(2.2) Here we would like to draw the reviewer’s and reader’s attention to Author response image 3, which shows that orientation-dependent pairing interactions have a significant impact on physical interactions between different sequences.  We would also refer the reader to two other publications.  One of these is Kyrchanova et al. (Kyrchanova et al. 2008), which was the first to demonstrate that orientation of pairing interactions matters.  The second is Fujioka et al. (Fujioka et al. 2016), which describes experiments indicating that nhomie and homie pair with each other head-to-tail and with themselves head-to-head.

On the whole, this study is comprehensive and represents a useful contribution to the 3D genome field. The transgenic lines and Micro-C datasets generated in the course of the work will be valuable resources for the research community. Moreover, the manuscript, while dense in places, is generally clearly written and comprehensive in its description of the work. However, I have a number of comments and critiques of the manuscript, mainly centering on the framing of the experiments and presentation of the Micro-C results and on manner in which the data are analyzed and reported. They are as follows:

Major Points:

(1) The authors motivate much of the introduction and results with hypothetical "stem loop" and "circle loop" models of chromosome confirmation, which they argue are reflected in the Micro-C data and help to explain the observed ISH patterns. While such structures may possibly form, the support for these specific models vs. the many alternatives is not in any way justified. For instance, no consideration is given to important biophysical properties such as persistence length, packing/scaling, and conformational entropy. As the biophysical properties of chromatin are a very trafficked topic both in terms of experimentation and computational modeling and generally considered in the analysis of chromosome conformation data, the study would be strengthened by acknowledgement of this body of work and more direct integration of its findings.

(2.3) The reviewer is not correct in claiming that “stem-loops” and “circle-loops” are “hypothetical.”  There is ample evidence that both types of loops are present in eukaryotic genomes, and that loop conformation has significant readouts in terms of not only the physical properties of TADs but also their functional properties.  Here we would draw the reviewer’s attention to Author response image 3 and Author response image 4 for examples of loops formed by the orientation-dependent pairing of yet other TAD boundary elements.  As evident from the MicroC data in these figures, circle-loops and stem-loops have readily distinguishable contact patterns.  The experiments in Fujioka et al. (Fujioka et al. 2016) demonstrate that homie and nhomie pair with each other head-to-tail, while they pair with themselves head-to-head.  The accompany paper (Bing et al. 2024) also provides evidence that loop topology is reflected both in the pattern of activation of reporters and in the MicroC contact profiles.  We would also mention again Kyrchanova et al. (Kyrchanova et al. 2008), who were the first to report orientation-dependent pairing of endogenous fly boundaries.

At this juncture it would premature to try to incorporate computational modeling of chromosome conformation in our studies.  The reason is that the experimental foundations that would be essential for building accurate models are lacking.  As should be evident from RFigs. 1-3 above, studies on mammalian chromosomes are simply not of high enough resolution to draw firm conclusions about chromosome conformation: in most studies only the forests are visible.  While the situation is better in flies, there are still too many unknown.  As just one example, it would be important to know the orientation of the boundary pairing interactions that generate each TAD.  While it is possible to infer loop topology from how TADs interact with their neighbors (a plume versus clouds), a conclusive identification of stem- and circle-loops will require a method to unambiguously determine whether a TAD boundary pairs with its neighbor head-to-head or headto-tail.

(2) Similar to Point 1, while there is a fair amount of discussion of how the observed results are or are not consistent with loop extrusion, there is no discussion of the biophysical forces that are thought to underly compartmentalization such as block-polymer co-segregation and their potential influence. I found this absence surprising, as it is generally accepted that A/B compartmentalization essentially can explain the contact maps observed in Drosophila and other non-vertebrate eukaryotes (Rowley, ..., Corces 2017; PMID 28826674). The manuscript would be strengthened by consideration of this phenomenon.

(2.4) Compartments in mammals have typically been identified and characterized using lowresolution data sets, and these studies have relied on visualizing compartments using quite large bin sizes (>>1 kb).  Our experiments have nothing to do with the large-scale compartments seen in these Hi-C experiments.  Instead, we are studying the properties of individual TADs: how TADs are formed, the relationship between TAD topology and boundary:boundary pairing, and the impact of TAD topology on interactions between TADs in the immediate neighborhood.  There is no evidence to date that these large compartments or “block polymer co-segregation” have a) any impact on the properties of individual boundary elements, b) have a role in determining which boundary elements actually come together to form a given TAD, c) impact the orientation of the interactions between boundaries that generate the TAD or d) determine how TADs tend to interact with their immediate neighbors.  

In more recent publications (c.f., Harris et al. 2023) compartments have shrunk in size and instead of being units of several hundred kb, the median length of the “compartmental” unit in mammalian cells is about12 kb. This is not too much different from the size of fly TADs.  However, the available evidence does not support the idea that block polymer co-segregation/co-repulsion drive the TAD:TAD interactions seen in MicroC experiments.  For example, according to this “micro-compartment” model, the specific patterns of interaction between TADs in the CG3294 meta-loop in Author response image 3 would be driven by block polymer co-segregation and co-repulsion. In this model, the TAD upstream of the blue boundary (which contains CG33543, the odorant binding protein gene Obp22a and the Npc2a gene which encodes a protein involved in sterol homeostasis) would share the same chromatin state/biophysical properties as the TAD upstream of the purple boundary, which has the fipi gene. While it is true that CG33543, Obp22a and also the fipi gene are not expressed in embryos, Npc2a is expressed at high levels during embryogenesis, yet it is part of the TAD that interacts with the fipi TAD.  The TAD downstream of the blue boundary contains CG15353 and Nplp4 and it interacts with the TAD downstream of the purple boundary which contains CG3294 and slf.  CG15353 and Nplp4 are not expressed in the embryo and as such should share a compartment with a TAD that is also silent. However, slf is expressed at a high level in 1216 hr embryos, while CG3294 is expressed at a low level.  In neither case would one conclude that the TADs upstream and downstream of the blue and purple boundaries, respectively, interact because of shared chromatin/biophysical states that drive block polymer co-segregation corepulsion. 

One might also consider several gedanken experiments involving the long-range interactions that generate the CG3294 meta-loop in Author response image 3.    According to the micro-compartment model the patchwork pattern of crosslinking evident in the CG3294 meta-loop arises because the interacting  TADs share the same biochemical/biophysical properties, and this drives block polymer cosegregation and co-repulsion.  If this model is correct, then this patchwork pattern of TAD:TAD interactions would remain unchanged if we were to delete the blue or the purple boundary.  However, given what we know about how boundaries can find and pair with distant boundaries (c.f., Figure 6 from Muller et el. 1999 and the discussion in #1.2), the result of these gedanken experiments seem clear: the patchwork pattern shown in Author response image 3A will disappear.  What would happen if we inverted the blue or the purple boundary? Would the TAD containing CG33543, Obp22a and Npc2a still interact with fipi as would be expected from the compartment model?  Or would the pattern of interactions flip so that the CG33543, Obp22a and Npc2a TAD interacts with the TAD containing CG3294 and slf?  Again we can anticipate the results based on previous studies: the interacting TADs will switch when the CG3294 meta-loop is converted into a stem-loop.  If this happened, the only explanation possible in the compartment model is that the chromatin states change when the boundary is inverted so that TAD upstream of blue boundary now shares the same chromatin state as the TAD downstream of the purple boundary, while the TAD downstream of the blue boundary shares same state as the TAD upstream of the purple boundary.  However, there is no evidence that boundary orientation per se can induce a complete switch in “chromatin states” as would be required in the compartment model. 

While we have not done these experimental manipulations with the CG3294 meta-loop, an equivalent experiment was done in Bing et al. (Bing et al. 2024).  However, instead of deleting a boundary element, we inserted a homie boundary element together with two reporters (gfp and LacZ) 142 kb away from the eve TAD.  The result of this gedanken “reverse boundary deletion” experiment is shown in Author response image 5.  Panel A shows the MicroC contact profile in the region spanning the transgene insertion site and the eve TAD in wild type (read “deletion”) NC14 embryos.  Panel B shows the MicroC contact profile from 12-16 hr embryos carrying the homie dual reporter transgene inserted at -142 kb.  Prior to the “deletion”, the homie element in the transgene pairs with nhomie and homie in the eve TAD and this generates a “mini-metaloop.”  In this particular insert, the homie boundary in the transgene (red arrow) is “pointing” in the opposite orientation from the homie boundary in the eve TAD (red arrow).  In this orientation, the pairing of the transgene homie with eve nhomie/homie brings the LacZ reporter into contact with sequences in the eve TAD.  Since a mini-metaloop is formed by homie_à _nhomie/homie pairing, sequences in TADs upstream and downstream of the transgene insert interact with sequences in TADs close to the eve TAD (Author response image 5B).  Taken together these interactions correspond to the interaction patchwork that is typically seen in “compartments” (see boxed region and inset).  If this patchwork is driven as per the model, by block polymer co-segregation and co-repulsion, then it should still be present when the transgene is deleted.  However, panel A shows that the interactions linking the transgene and the sequences in TADs next to the transgene to eve and TADs next to eve disappear when the homie boundary (plus transgene) is “deleted” in wild type flies.

Author response image 5.

Boundary deletion and compartments

A second experiment would be to invert the homie boundary so that instead of pointing away from eve it points towards eve.  Again, if the compartmental patchwork is driven by block polymer co-segregation and co-repulsion, inverting the homie boundary in the transgene should have no effect on the compartmental contact profile.  Inspection of Fig. 7 in Bing et al. (Bing et al. 2024) will show that this prediction doesn’t hold either.  When homie is inverted, sequences in the eve TAD interact with the gfp reporter not the LacZ reporter.  In addition, there are corresponding changes in how sequences in TADs to either side of eve interact with sequences to either side of the transgene insert.  

Yet another “test” of compartments generated by block polymer co-segregation/co-repulsion is provided by the plume above the eve volcano triangle.  According to the compartment model, sequences in TADs flanking the eve locus form the plume above the eve volcano triangle because their chromatin shares properties that drive block polymer co-segregation.  These same properties result in repulsive interactions with chromatin in the eve TAD, and this would explain why the eve TAD doesn’t crosslink with its neighbors.  If the distinctive chromatin properties of eve and the neighboring TADs drive block polymer co-segregation and co-repulsion, then inverting the nhomie boundary or introducing homie in the forward orientation should have absolutely no effect on the physical interactions between chromatin in the eve TAD and chromatin in the neighboring TADs.  However, Figures 4 and 6 in this paper indicate that boundary pairing orientation, not block polymer co-segregation/co-repulsion, is responsible for forming the plume above the eve TAD. Other findings also appear to be inconsistent with the compartment model. (A) The plume topping the eve volcano triangle is present in NC14 embryos when eve is broadly expressed (and potentially active throughout the embryo).  It is also present in 12-16 hr embryos when eve is only expressed in a very small subset of cells and is subject to PcG silencing everywhere else in the embryo.  B) According to the compartment model the precise patchwork pattern of physical interactions should depend upon the transcriptional program/chromatin state that is characteristic of a particular developmental stage or cell type.  As cell fate decisions are just being made during NC14 one might expect that most nuclei will share similar chromatin states throughout much of the genome.  This would not be true for 12-16 hr embryos.  At this stage the compartmental patchwork would be generated by a complex mixture of interactions in cells that have quite different transcriptional programs and chromatin states.  In this case, the patchwork pattern would be expected to become fuzzy as a given chromosomal segment would be in compartment A in one group of cells and in compartment B in another.   Unlike 12-16 hr embryos,  larval wing discs would be much more homogeneous and likely give a distinct and relatively well resolved compartmental pattern. We’ve examined the compartment patchwork of the same chromosomal segments in NC14 embryos, 12-16 hr embryos and larval wing disc cells.  While there are some differences (e.g., changes in some of the BX-C TADs in the wing disc sample) the compartmental patchwork patterns are surprisingly similar in all three cases. Nor is there any “fuzziness” in the compartmental patterns evident in 12-16 hr embryos, despite the fact that there are many different cell types at this stage of development.  C) TAD interactions with their neighbors and compartmental patchworks are substantially suppressed in salivary gland polytene chromosomes.  This would suggest that features of chromosome structure might be the driving force behind many of the “compartmental” interactions as opposed to distinct biochemical/biophysical of properties of small chromosomal segments that drive polymer co- segregation/co-repulsion.  

(3) The contact maps presented in the study represent many cells and distinct cell types. It is clear from single-cell Hi-C and multiplexed FISH experiments that chromosome conformation is highly variable even within populations of the same cell, let alone between cell types, with structures such as TADs being entirely absent at the single cell level and only appearing upon pseudobulking. It is difficult to square these observations with the models of relatively static structures depicted here. The authors should provide commentary on this point.

(2.5) As should be evident from Author response image 1, single-cell Hi-C experiments would not provide useful information about the physical organization of individual TADs, TAD boundaries or how individual TADs interact with their immediate neighbors.  In addition, since they capture only a very small fraction of the possible contacts within and between TADs, we suspect that these single-cell studies aren’t likely to be useful for making solid conclusions about TAD neighborhoods like those shown in Author response image 1 panels A, B, C and D, or Author response image 2.  While it might be possible to discern relatively stable contacts between pairs of insulators in single cells with the right experimental protocol, the stabilities/dynamics of these interactions may be better judged by the length of time that physical interactions are seen to persist in live imaging studies such as Chen et al. (2018), Vazquez et al. (2006) and Li et al. (2011).

The in situ FISH data we’ve seen also seems problematic in that probe hybridization results in a significant decondensation of chromatin.  For two probe sets complementary to adjacent ~1.2 kb DNA sequences, the measured center-to-center distance that we’ve seen was ~110 nM.  This is about 1/3rd the length that is expected for a 1.2 kb naked DNA fragment, and about 1.7 times larger than that expected for a beads-on-a-string nucleosome array (~60 nM).  However, chromatin is thought to be compacted into a 30 nM fiber, which is estimated to reduce the length of DNA by at least another ~6 fold.  If this estimate is correct, FISH hybridization would appear to result in a ~10 fold decompaction of chromatin.  A decompaction of this magnitude would necessarily be followed by a significant distortion in the actual conformation of chromatin loops.

(4) The analysis of the Micro-C data appears to be largely qualitative. Key information about the number of reads sequenced, reaps mapped, and data quality are not presented. No quantitative framework for identifying features such as the "plumes" is described. The study and its findings would be strengthened by a more rigorous analysis of these rich datasets, including the use of systematic thresholds for calling patterns of organization in the data.

Additional information on the number of reads and data quality have been included in the methods section. 

(5) Related to Point 4, the lack of quantitative details about the Micro-C data make it difficult to evaluate if the changes observed are due to biological or technical factors. It is essential that the authors provide quantitative means of controlling for factors like sampling depth, normalization, and data quality between the samples.

In our view the changes in the MicroC contact patterns for the eve locus and its neighbors when the nhomie boundary is manipulated are not only clear cut and unambiguous but are also readily evident in the Figs that are presented in the manuscript.  If the reviewer believes that there aren’t significant differences between the MicroC contact patterns for the four different nhomie replacements, it seems certain that they would also remain unconvinced by a quantitative analysis.

The reviewer also suggests that biological and/or technical differences between the four samples could account for the observed changes in the MicroC patterns for the eve TAD and its neighbors.  If this were the case, then similar changes in MicroC patterns should be observed elsewhere in the genome.  Since much of the genome is analyzed in these MicroC experiments there is an abundance of internal controls for each experimental manipulation of the nhomie boundary.  For two of the nhomie replacements, nhomie reverse and homie forward, the plume above the eve volcano triangle is replaced by clouds surrounding the eve volcano triangle.  If these changes in the eve MicroC contact patterns are due to significant technical (or biological) factors, we should observe precisely the same sorts of changes in TADs elsewhere in the genome that are volcano triangles with plumes.   Author response image 6 shows the MicroC contact pattern for several genes in the Antennapedia complex.  The deformed gene is included in a TAD which, like eve, is a volcano triangle topped by a plume.  A comparison of the deformed MicroC contact patterns for nhomie forward (panel B) with the MicroC patterns for nhomie reverse (panel C) and homie forward (panel D) indicates that while there are clearly technical differences between the samples, these differences do not result in the conversion of the deformed plume into clouds as is observed for the eve TAD.  The MicroC patterns elsewhere in Antennapedia complex are also very similar in all four samples.  Likewise, comparisons of regions elsewhere in the fly genome indicate that the basic contact patterns are similar in all four samples.   So while there are technical differences which are reflected in the relative pixel density in the TAD triangles and the LDC domains, these differences do not result in converting plumes into clouds nor do the alter the basic patterns of TAD triangles and LDC domains.  As for biological differences— the embryos in each sample are at roughly the same developmental stage and were collected and processed using the same procedures. Thus, the biological factors that could reasonably be expected to impact the organization of specific TADs (e.g., cell type specific differences) are not going to impact the patterns we see in our experiments. 

Author response image 6.

(6) The ISH effects reported are modest, especially in the case of the HCR. The details provided for how the imaging data were acquired and analyzed are minimal, which makes evaluating them challenging. It would strengthen the study to provide much more detail about the acquisition and analysis and to include depiction of intermediates in the analysis process, e.g. the showing segmentation of stripes.

The imaging analysis is presented in Fig. 5 is just standard confocal microscopy.  Individual embryos were visualized and scored.  An embryo in which stripes could be readily detected was scored as ‘positive’ while an embryo in which stripes couldn’t be detected was scored as ‘negative.’   

Recommendations for the authors:

Editor comments:

It was noted that the Jaynes lab previously published extensive genetic evidence to support the stem loop and circle loop models of Homie-Nhomie interactions (Fujioka 2016 Plos Genetics) that were more convincing than the Micro-C data presented here in proof of their prior model. Maybe the authors could more clearly summarize their prior genetic results to further try to convince the reader about the validity of their model.

Reviewer #1 (Recommendations For The Authors):

Below, I list specific comments to further improve the manuscript for publication. Most importantly, I recommend the authors tone down their proposal that boundary pairing is a universal TAD forming mechanism.

(1) The title is cryptic.

(2) The second sentence in the abstract is an overstatement: "In flies, TADs are formed by physical interactions between neighboring boundaries". Hi-C and Micro-C studies have not provided evidence that most TADs in Drosophila show focal interactions between their bracketing boundaries. The authors rely too strongly on prior studies that used artificial reporter transgenes to show that multimerized insulator protein binding sites or some endogenous fly boundaries can mediate boundary bypass, as evidence that endogenous boundaries pair.

Please see responses #1.1 and #1.3 and figures Author response image 1 and Author response image 3.  Note that using dHS-C, most TADs that we’ve looked at so far are topped by a “dot” at their apex.

(3) Line 64: the references do not cite the stated "studies dating back to the '90's'".

The papers cited for that sentence are reviews which discussed the earlier findings.  The relevant publications are cited at the appropriate places in the same paragraph.  

(4) Line 93: "On the other hand, while boundaries have partner preferences, they are also promiscuous in their ability to establish functional interactions with other boundaries." It was unclear what is meant here.

Boundaries that a) share binding sites for proteins that multimerized, b) have binding sites for proteins that interact with each other, or c) have binding sites for proteins that can be bridged by a third protein can potentially pair with each other.  However, while these mechanisms enable promiscuous pairing interactions, they will also generate partner preferences (through a greater number of a, b and/or c).

(5) It could be interesting to discuss the fact that it remains unclear whether Nhomie and Homie pair in cis or in trans, given that homologous chromosomes are paired in Drosophila.

The studies in Fujioka et al. (Fujioka et al. 2016) show that nhomie and homie can pair both in cis and in trans.  Given the results described in #1.2, we imagine that they are paired in both cis and trans in our experiments.

(6) Line 321: Could the authors further explain why they think that "the nhomie reverse circle-loop also differs from the nhomie deletion (λ DNA) in that there is not such an obvious preference for which eve enhancers activate expression"?

The likely explanation is that the topology/folding of the altered TADs impacts the probability of interactions between the various eve enhancers and the promoters of the flanking genes.  

(7) The manuscript would benefit from shortening the long Discussion by avoiding repeating points described previously in the Results.

(8) Line 495: "If, as seems likely, a significant fraction of the TADs genome-wide are circle loops, this would effectively exclude cohesin-based loop extrusion as a general mechanism for TAD formation in flies". The evidence provided in this manuscript appears insufficient to discard ample evidence from multiple laboratories that TADs form by compartmentalization or loop extrusion. Multiple laboratories have, for example, demonstrated that cohesin depletion disrupts a large fraction of mammalian TADs. 

Points made here and in #9 have been responded to in #1.1, #2.1 and #2.4 above.  We would suggest that the evidence for loop extrusion falls short of compelling (as it is based on the analysis of TAD neighborhoods, not TADs—that is forests, not trees) and given the results reported in Goel et al. (in particular Fig. 4 and Sup Fig. 8) is clearly suspect. This is not to mention the fact that cohesin loop-extrusion can’t generate circle-loops TADs, yet circle-loops clearly exist.  Likewise, as discussed in #2.4, it is not clear to us that the shared chromatin states, polymer co-segregation and co-repulsion account for the compartmental patchwork patterns of TAD;TAD interactions. The results from the  experimental manipulations in this paper and the accompanying paper, together with studies by others (e.g., Kyrchanova et al. (Kyrchanova et al. 2008), Mohana et al. (Mohana et al. 2023) would also seem to be at odds with the model for compartments as currently formulated.  

The unique properties of Nhomie and Homie, namely the remarkable specificity with which they physically pair over large distances (Fujioka et al. 2016) may rather suggest that boundary pairing is a phenomenon restricted to special loci. Moreover, it has not yet been demonstrated that Nhomie or Homie are also able to pair with the TAD boundaries on their left or right, respectively.

Points made here were discussed in detail in #1.2.  As described in detail in #1.2, It is not the case that nhomie and homie are in “unique” or “special.”  Other fly boundaries can do the same things.  As for whether nhomie and homie pair with their neighbors:  We haven’t done transgene experiments (e.g., testing by transvection or boundary bypass).  Likewise, in MicroC experiments there are no obvious dots at the apex of the neighboring TADs that would correspond to nhomie pairing with the neighboring boundary to the left and homie pairing with the neighboring boundary to the right. However, this is to be expected. As we discussed in in #1.3 above, only MNase resistant elements will generate dots in standard MicroC experiments.  On the other hand, when boundary:boundary interactions are analyzed by dHS-C (c.f., Author response image 4), there are dots at the apex of both neighboring TADs.  This would be direct evidence that nhomie pairs with the neighboring boundary to the left and homie pairs with the neighboring boundary to the right.

(9) The comment in point 8 also applies to the concluding 2 sentences (lines 519-524) of the Discussion.

See response to 8 above. Otherwise, the concluding sentences are completely accurate. Validation of the cohesin loop extrusion/CTCF roadblock model will required demonstrating a) that all TADs are either stem-loops or unanchored loops and b) that TAD endpoints are always marked by CTCF. 

The likely presence of circle-loops and evidence that TAD boundaries that don’t have CTCF (c.f.,Goel et al. 2023) already suggests that this model can’t (either fully or not all) account for TAD formation in mammals. 

(10) Figs. 3 and 6: It would be helpful to add the WT screenshot in the same figure, for direct comparison.

It is easy enough to scroll between Figs-especially since nhomie forward looks just like WT.

(11) Fig. 6: It would be helpful to show a cartoon view of a circle loop to the right of the Micro-C screenshot, as was done in Fig. 3.

Good idea.   Added to the Fig.

(12) Fig. 5: It would be helpful to standardize the labelling of the different genotypes throughout the figures and panels ("inverted" versus "reverse" versus an arrow indicating the direction).

Fixed.

Reviewer #2 (Recommendations For The Authors):

Minor Points:

(1) The Micro-C data does not appear to be deposited in an appropriate repository. It would be beneficial to the community to make these data available in this way.

This has been done.

(2) Readers not familiar with Drosophila development would benefit from a gentle introduction to the stages analyzed and some brief discussion on how the phenomenon of somatic homolog pairing might influence the study, if at all.

We included a rough description the stages that were analyzed for both the in situs and MicroC. We thought that an actual description of what is going on at each of the stages wasn’t necessary as the process of development is not a focus of this manuscript.  In other studies, we’ve found that there are only minor differences in MicroC patterns between the blastoderm stage and stage 12-16 embryos.  While these minor differences are clearly interesting, we didn’t discuss them in the text.   In all of experiments chromosomes are likely to be paired.  In NC14 embryos (the stage for visualizing eve stripes and the MicroC contact profiles in Fig. 2) replication of euchromatic sequences is thought to be quite rapid.  While homolog pairing is incomplete at this stage, sister chromosomes are paired.  In stage 12-16 embryos, homologs will be paired and if the cells are arrested in G2, then sister chromosome will also be paired.  So in all of experiments, chromosomes (sisters and/or homologs) are paired. However, since we don’t have examples of unpaired chromosomes, our experiments don’t provide any info on how chromosome pairing might impact MicroC/expression patterns.

(3) "P > 0.01" appears several times. I believe the authors mean to report "P < 0.01".

Fixed.  

References for Response

Batut PJ, Bing XY, Sisco Z, Raimundo J, Levo M, Levine MS. 2022. Genome organization controls transcriptional dynamics during development. Science. 375(6580):566-570.

Bing X, Ke W, Fujioka M, Kurbidaeva A, Levitt S, Levine M, Schedl P, Jaynes JB. 2024. Chromosome structure i: Loop extrusion or boundary:Boundary pairing? eLife.

Blanton J, Gaszner M, Schedl P. 2003. Protein:Protein interactions and the pairing of boundary elements in vivo. Genes Dev. 17(5):664-675.

Bonchuk A, Boyko K, Fedotova A, Nikolaeva A, Lushchekina S, Khrustaleva A, Popov V, Georgiev P. 2021. Structural basis of diversity and homodimerization specificity of zincfinger-associated domains in drosophila. Nucleic Acids Res. 49(4):2375-2389.

Bonchuk A, Kamalyan S, Mariasina S, Boyko K, Popov V, Maksimenko O, Georgiev P. 2020. Nterminal domain of the architectural protein ctcf has similar structural organization and ability to self-association in bilaterian organisms. Sci Rep. 10(1):2677.

Chen H, Levo M, Barinov L, Fujioka M, Jaynes JB, Gregor T. 2018. Dynamic interplay between enhancer–promoter topology and gene activity. Nat Genet. 50(9):1296.

Fedotova AA, Bonchuk AN, Mogila VA, Georgiev PG. 2017. C2h2 zinc finger proteins: The largest but poorly explored family of higher eukaryotic transcription factors. Acta Naturae. 9(2):47-58.

Fujioka M, Ke W, Schedl P, Jaynes JB. 2024. The homie insulator has sub-elements with different insulating and long-range pairing properties. bioRxiv. 2024.02.01.578481.

Fujioka M, Mistry H, Schedl P, Jaynes JB. 2016. Determinants of chromosome architecture: Insulator pairing in cis and in trans. PLoS Genet. 12(2):e1005889.

Galloni M, Gyurkovics H, Schedl P, Karch F. 1993. The bluetail transposon: Evidence for independent cis‐regulatory domains and domain boundaries in the bithorax complex. The EMBO Journal. 12(3):1087-1097.

Gaszner M, Vazquez J, Schedl P. 1999. The zw5 protein, a component of the scs chromatin domain boundary, is able to block enhancer-promoter interaction. Genes Dev. 13(16):2098-2107.

Goel VY, Huseyin MK, Hansen AS. 2023. Region capture micro-c reveals coalescence of enhancers and promoters into nested microcompartments. Nat Genet. 55(6):1048-1056.

Harris HL, Gu H, Olshansky M, Wang A, Farabella I, Eliaz Y, Kalluchi A, Krishna A, Jacobs M, Cauer G et al. 2023. Chromatin alternates between a and b compartments at kilobase scale for subgenic organization. Nat Commun. 14(1):3303.

Hart CM, Zhao K, Laemmli UK. 1997. The scs' boundary element: Characterization of boundary element-associated factors. Mol Cell Biol. 17(2):999-1009.

Hsieh TS, Cattoglio C, Slobodyanyuk E, Hansen AS, Rando OJ, Tjian R, Darzacq X. 2020. Resolving the 3d landscape of transcription-linked mammalian chromatin folding. Mol Cell. 78(3):539-553.e538.

Krietenstein N, Abraham S, Venev SV, Abdennur N, Gibcus J, Hsieh TS, Parsi KM, Yang L, Maehr R, Mirny LA et al. 2020. Ultrastructural details of mammalian chromosome architecture. Mol Cell. 78(3):554-565.e557.

Kyrchanova O, Chetverina D, Maksimenko O, Kullyev A, Georgiev P. 2008. Orientation-dependent interaction between drosophila insulators is a property of this class of regulatory elements. Nucleic Acids Res. 36(22):7019-7028.

Kyrchanova O, Ibragimov A, Postika N, Georgiev P, Schedl P. 2023. Boundary bypass activity in the abdominal-b region of the drosophila bithorax complex is position dependent and regulated. Open Biol. 13(8):230035.

Kyrchanova O, Kurbidaeva A, Sabirov M, Postika N, Wolle D, Aoki T, Maksimenko O, Mogila V, Schedl P, Georgiev P. 2018. The bithorax complex iab-7 polycomb response element has a novel role in the functioning of the fab-7 chromatin boundary. PLoS Genet. 14(8):e1007442. Kyrchanova O, Sabirov M, Mogila V, Kurbidaeva A, Postika N, Maksimenko O, Schedl P, Georgiev P. 2019a. Complete reconstitution of bypass and blocking functions in a minimal artificial fab-7 insulator from drosophila bithorax complex. Proceedings of the National Academy of Sciences.201907190.

Kyrchanova O, Wolle D, Sabirov M, Kurbidaeva A, Aoki T, Maksimenko O, Kyrchanova M, Georgiev P, Schedl P. 2019b. Distinct elements confer the blocking and bypass functions of the bithorax fab-8 boundary. Genetics.genetics. 302694.302019.

Kyrchanova O, Zolotarev N, Mogila V, Maksimenko O, Schedl P, Georgiev P. 2017. Architectural protein pita cooperates with dctcf in organization of functional boundaries in bithorax complex. Development. 144(14):2663-2672.

Li H-B, Muller M, Bahechar IA, Kyrchanova O, Ohno K, Georgiev P, Pirrotta V. 2011. Insulators, not polycomb response elements, are required for long-range interactions between polycomb targets in drosophila melanogaster. Mol Cell Biol. 31(4):616-625.

Li X, Tang X, Bing X, Catalano C, Li T, Dolsten G, Wu C, Levine M. 2023. Gaga-associated factor fosters loop formation in the drosophila genome. Mol Cell. 83(9):1519-1526.e1514.

Mohana G, Dorier J, Li X, Mouginot M, Smith RC, Malek H, Leleu M, Rodriguez D, Khadka J, Rosa P et al. 2023. Chromosome-level organization of the regulatory genome in the drosophila nervous system. Cell. 186(18):3826-3844.e3826.

Muller M, Hagstrom K, Gyurkovics H, Pirrotta V, Schedl P. 1999. The mcp element from the drosophila melanogaster bithorax complex mediates long-distance regulatory interactions. Genetics. 153(3):1333-1356.

Muravyova E, Golovnin A, Gracheva E, Parshikov A, Belenkaya T, Pirrotta V, Georgiev P. 2001. Loss of insulator activity by paired su(hw) chromatin insulators. Science. 291(5503):495498.

Postika N, Metzler M, Affolter M, Müller M, Schedl P, Georgiev P, Kyrchanova O. 2018. Boundaries mediate long-distance interactions between enhancers and promoters in the drosophila bithorax complex. PLoS Genet. 14(12):e1007702.

Samal B, Worcel A, Louis C, Schedl P. 1981. Chromatin structure of the histone genes of d. Melanogaster. Cell. 23(2):401-409.

Sigrist CJ, Pirrotta V. 1997. Chromatin insulator elements block the silencing of a target gene by the drosophila polycomb response element (pre) but allow trans interactions between pres on different chromosomes. Genetics. 147(1):209-221.

Udvardy A, Schedl P. 1984. Chromatin organization of the 87a7 heat shock locus of drosophila melanogaster. J Mol Biol. 172(4):385-403.

Vazquez J, Muller M, Pirrotta V, Sedat JW. 2006. The mcp element mediates stable long-range chromosome-chromosome interactions in drosophila. Molecular Biology of the Cell. 17(5):2158-2165.

Zolotarev N, Fedotova A, Kyrchanova O, Bonchuk A, Penin AA, Lando AS, Eliseeva IA, Kulakovskiy IV, Maksimenko O, Georgiev P. 2016. Architectural proteins pita, zw5,and zipic contain homodimerization domain and support specific long-range interactions in drosophila. Nucleic Acids Res. 44(15):7228-7241.

Author response:

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

Reviewer #1 (Public review):

From the Reviewing Editor:

Four reviewers have assessed your manuscript on valence and salience signaling in the central amygdala. There was universal agreement that the question being asked by the experiment is important. There was consensus that the neural population being examined (GABA neurons) was important and the circular shift method for identifying task-responsive neurons was rigorous. Indeed, observing valenced outcome signaling in GABA neurons would considerably increase the role the central amygdala in valence. However, each reviewer brought up significant concerns about the design, analysis and interpretation of the results. Overall, these concerns limit the conclusions that can be drawn from the results. Addressing the concerns (described below) would work towards better answering the question at the outset of the experiment: how does the central amygdala represent salience vs valence.

A weakness noted by all reviewers was the use of the terms 'valence' and 'salience' as well as the experimental design used to reveal these signals. The two outcomes used emphasized non-overlapping sensory modalities and produced unrelated behavioral responses. Within each modality there are no manipulations that would scale either the value of the valenced outcomes or the intensity of the salient outcomes. While the food outcomes were presented many times (20 times per session over 10 sessions of appetitive conditioning) the shock outcomes were presented many fewer times (10 times in a single session). The large difference in presentations is likely to further distinguish the two outcomes. Collectively, these experimental design decisions meant that any observed differences in central amygdala GABA neuron responding are unlikely to reflect valence, but likely to reflect one or more of the above features.

We appreciate the reviewers’ comments regarding the experimental design. When assessing fear versus reward, we chose stimuli that elicit known behavioral responses, freezing versus consumption. The use of stimuli of the same modality is unlikely to elicit easily definable fear or reward responses or to be precisely matched for sensory intensity. For example, sweet or bitter tastes can be used, but even these activate different taste receptors and vary in the duration of the activation of taste-specific signaling (e.g. how long the taste lingers in the mouth). The approach we employed is similar to that of Yang et al., 2023 (doi: 10.1038/s41586-023-05910-2) that used water reward and shock to characterize the response profiles of somatostatin neurons of the central amygdala. Similar to what was reported by Yang and colleagues we observed that the majority of CeA GABA neurons responded selectively to one unconditioned stimulus (~52%). We observed that 15% of neurons responded in the same direction, either activated or inhibited, by the food or shock US. These were defined as salience based on the definitions of Lin and Nicolelis, 2008 (doi: 10.1016/j.neuron.2008.04.031) in which basal forebrain neurons responded similarly to reward or punishment irrespective of valence. The designation of valence encoding based opposite responses to the food or shock is straightforward (~10% of cells); however, we agree that the designation of modality-specific encoding neurons as valence encoding is less straightforward.

A second weakness noted by a majority of reviewers was a lack of cue-responsive unit and a lack of exploration of the diversity of response types, and the relationship cue and outcome firing. The lack of large numbers of neurons increasing firing to one or both cues is particularly surprising given the critical contribution of central amygdala GABA neurons to the acquisition of conditioned fear (which the authors measured) as well as to conditioned orienting (which the authors did not measure). Regression-like analyses would be a straightforward means of identifying neurons varying their firing in accordance with these or other behaviors. It was also noted that appetitive behavior was not measured in a rigorous way. Instead of measuring time near hopper, measures of licking would have been better. Further, measures of orienting behaviors such as startle were missing.

The authors also missed an opportunity for clustering-like analyses which could have been used to reveal neurons uniquely signaling cues, outcomes or combinations of cues and outcomes. If the authors calcium imaging approach is not able to detect expected central amygdala cue responding, might it be missing other critical aspects of responding?

As stated in the manuscript, we were surprised by the relatively low number of cue responsive cells; however, when using a less stringent statistical method (Figure 5 - Supplement 2), we observed 13% of neurons responded to the food associated cue and 23% responded to the shock associated cue. The differences are therefore likely a reflection of the rigor of the statistical measure to define the responsive units. The number of CS responsive units is less than reported in the CeAl by Ciocchi et al., 2010 (doi: 10.1038/nature09559 ) who observed 30% activated by the CS and 25% inhibited, but is not that dissimilar from the results of Duvarci et al., 2011 (doi: 10.1523/JNEUROSCI.4985-10.2011 ) who observed 11% activated in the CeAl and 25% inhibited by the CS. These numbers are also consistent with previous single cell calcium imaging of cell types in the CeA. For example, Yang et al., 2023 (doi: 10.1038/s41586-023-05910-2) observed that 13% of somatostatin neurons responded to a reward CS and 8% responded to a shock CS. Yu et al., 2017 (doi: 10.1038/s41593-017-0009-9) observed 26.5% of PKCdelta neurons responded to the shock CS. It should also be noted that our analysis was not restricted to the CeAl. Finally, Food learning was assessed in an operant chamber in freely moving mice with reward pellet delivery. Because liquids were not used for the reward US, licking is not a metric that can be used.

All reviewers point out that the evidence for salience encoding is even more limited than the evidence for valence. Although the specific concern for each reviewer varied, they all centered on an oversimplistic definition of salience. Salience ought to scale with the absolute value and intensity of the stimulus. Salience cannot simply be responding in the same direction. Further, even though the authors observed subsets of central amygdala neurons increasing or decreasing activity to both outcomes - the outcomes can readily be distinguished based on the temporal profile of responding.

We thank the reviewers for their comments relating to the definition of salience and valence encoding by central amygdala neurons. We have addressed each of the concerns below.

Additional concerns are raised by each reviewer. Our consensus is that this study sought to answer an important question - whether central amygdala signal salience or valence in cue-outcome learning. However, the experimental design, analyses, and interpretations do not permit a rigorous and definitive answer to that question. Such an answer would require additional experiments whose designs would address the significant concerns described here. Fully addressing the concerns of each reviewer would result in a re-evaluation of the findings. For example, experimental design better revealing valence and salience, and analyses describing diversity of neuronal responding and relationship to behavior would likely make the results Important or even Fundamental.

We appreciate the reviewers’ comments and have addressed each concern below.

Reviewer #2 (Public review):

In this article, Kong and authors sought to determine the encoding properties of central amygdala (CeA) neurons in response to oppositely valenced stimuli and cues predicting those stimuli. The amygdala and its subregional components have historically been understood to be regions that encode associative information, including valence stimuli. The authors performed calcium imaging of GABA-ergic CeA neurons in freely-moving mice conditioned in Pavlovian appetitive and fear paradigms, and showed that CeA neurons are responsive to both appetitive and aversive unconditioned and conditioned stimuli. They used a variant of a previously published 'circular shifting' technique (Harris, 2021), which allowed them to delineate between excited/non-responsive/inhibited neurons. While there is considerable overlap of CeA neurons responding to both unconditioned stimuli (in this case, food and shock, deemed "salience-encoding" neurons), there are considerably fewer CeA neurons that respond to both conditioned stimuli that predict the food and shock. The authors finally demonstrated that there are no differences in the order of Pavlovian paradigms (fear - shock vs. shock - fear), which is an interesting result, and convincingly presented given their counterbalanced experimental design.

In total, I find the presented study useful in understanding the dynamics of CeA neurons during a Pavlovian learning paradigm. There are many strengths of this study, including the important question and clear presentation, the circular shifting analysis was convincing to me, and the manuscript was well written. We hope the authors will find our comments constructive if they choose to revise their manuscript.

While the experiments and data are of value, I do not agree with the authors interpretation of their data, and take issue with the way they used the terms "salience" and "valence" (and would encourage them to check out Namburi et al., NPP, 2016) regarding the operational definitions of salience and valence which differ from my reading of the literature. To be fair, a recent study from another group that reports experiments/findings which are very similar to the ones in the present study (Yang et al., 2023, describing valence coding in the CeA using a similar approach) also uses the terms valence and salience in a rather liberal way that I would also have issues with (see below). Either new experiments or revised claims would be needed here, and more balanced discussion on this topic would be nice to see, and I felt that there were some aspects of novelty in this study that could be better highlighted (see below).

One noteworthy point of alarm is that it seems as if two data panels including heatmaps are duplicated (perhaps that panel G of Figure 5-figure supplement 2 is a cut and paste error? It is duplicated from panel E and does not match the associated histogram).

We thank the reviewer for their insightful comments and assessment of the manuscript.

Major concerns:

(1) The authors wish to make claims about salience and valence. This is my biggest gripe, so I will start here.

(1a) Valence scales for positive and negative stimuli and as stated in Namburi et al., NPP, 2016 where we operationalize "valence" as having different responses for positive and negative values and no response for stimuli that are not motivational significant (neutral cues that do not predict an outcome). The threshold for claiming salience, which we define as scaling with the absolute value of the stimulus, and not responding to a neutral stimulus (Namburi et al., NPP, 2016; Tye, Neuron, 2018; Li et al., Nature, 2022) would require the lack of response to a neutral cue.

We appreciate the reviewer’s comment on the definitions of salience and valence and agree that there is not a consistent classification of these response types in the field. As stated above, we used the designation of salience encoding if the cells respond in the same direction to different stimuli regardless of the valence of the stimulus similar to what was described previously (Lin and Nicolelis, 2008, doi: 10.1016/j.neuron.2008.04.031). Similar definitions of salience have also been reported elsewhere (for examples see: Stephenson-Jones et al., 2020, doi: 10.1016/j.neuron.2019.12.006,  Zhu et al., 2018 doi: 10.1126/science.aat0481, and  Comoli et al., 2003, doi: 10.1038/nn1113P). Per the suggestion of the reviewer, we longitudinally tracked cells on the first day of Pavlovian reward conditioning the fear conditioning day. Although there were considerably fewer head entries on the first day of reward conditioning, we were able to identify 10 cells that were activated by both the food US and shock US. We compared the responses to the first five head entries and last head entries and the first 5 shocks and last five shocks. Consistent with what has been reported for salience encoding neurons in the basal forebrain (Lin and Nicolelis, 2008, doi: 10.1016/j.neuron.2008.04.031), we observed that the responses were highest when the US was most unexpected and decreased in later trials.

Author response image 1.

(1b) The other major issue is that the authors choose to make claims about the neural responses to the USs rather than the CSs. However, being shocked and receiving sucrose also would have very different sensorimotor representations, and any differences in responses could be attributed to those confounds rather than valence or salience. They could make claims regarding salience or valence with respect to the differences in the CSs but they should restrict analysis to the period prior to the US delivery.

Perhaps the reviewer missed this, but analysis of valence and salience encoding to the different CSs are presented in Figure 5G, Figure 5 -Supplement 1 C-D, and Figure 5 -Supplement 2 N-O. Analysis of CS responsiveness to CSFood and CSShock were analyzed during the conditioning sessions Figure 3E-F, Figure 4B-C, Figure 5 – Supplement 2J-O and Figure 5 – Supplement 3K-L, and during recall probe tests for both CSFood and CSShock, Figure 5 – Supplement 1C-J.

(1c) The third obstacle to using the terms "salience" or "valence" is the lack of scaling, which is perhaps a bigger ask. At minimum either the scaling or the neutral cue would be needed to make claims about valence or salience encoding. Perhaps the authors disagree - that is fine. But they should at least acknowledge that there is literature that would say otherwise.<br /> (1d) In order to make claims about valence, the authors must take into account the sensory confound of the modality of the US (also mentioned in Namburi et al., 2016). The claim that these CeA neurons are indeed valence-encoding (based on their responses to the unconditioned stimuli) is confounded by the fact that the appetitive US (food) is a gustatory stimulus while the aversive US (shock) is a tactile stimulus.

We provided the same analysis for the US and CS. The US responses were larger and more prevalent, but similar types of encoding were observed for the CS. We agree that the food reward and the shock are very different sensory modalities. As stated above, the use of stimuli of the same modality is unlikely to elicit easily definable fear or reward responses or to be precisely matched for sensory intensity. We agree that the definition of cells that respond to only one stimulus is difficult to define in terms of valence encoding, as opposed to being specific for the sensory modality and without scaling of the stimulus it is difficult to fully address this issue. It should be noted however, that if the cells in the CeA were exclusively tuned to stimuli of different sensory modalities, we would expect to see a similar number of cells responding to the CS tones (auditory) as respond to the food (taste) and shock (somatosensory) but we do not. Of the cells tracked longitudinally 80% responded to the USs, with 65% of cells responding to food (activated or inhibited) and 44% responding to shock (activated or inhibited).

(2) Much of the central findings in this manuscript have been previously described in the literature. Yang et al., 2023 for instance shows that the CeA encodes salience (as demonstrated by the scaled responses to the increased value of unconditioned stimuli, Figure 1 j-m), and that learning amplifies responsiveness to unconditioned stimuli (Figure 2). It is nice to see a reproduction of the finding that learning amplifies CeA responses, though one study is in SST::Cre and this one in VGAT::cre - perhaps highlighting this difference could maximize the collective utility for the scientific community?

We agree that the analysis performed here is similar to what was conducted by Yang et al., 2023. With the major difference being the types of neurons sampled. Yang et al., imaged only somatostatin neurons were as we recorded all GABAergic cell types within the CeA. Moreover, because we imaged from 10 mice, we sampled neurons that ostensibly covered the entire dorsal to ventral extent of the CeA (Figure 1 – Supplement 1). Remarkably, we found that the vast majority of CeA neurons (80%) are responsive to food or shock. Within this 80% there are 8 distinct response profiles consistent with the heterogeneity of cell types within the CeA based on connectivity, electrophysiological properties, and gene expression. Moreover, we did not find any spatial distinction between food or shock responsive cells, with the responsive cell types being intermingled throughout the dorsal to ventral axis (Figure 5 – Supplement 3).

(3) There is at least one instance of copy-paste error in the figures that raised alarm. In the supplementary information (Figure 5- figure supplement 2 E;G), the heat maps for food-responsive neurons and shock-responsive neurons are identical. While this almost certainly is a clerical error, the authors would benefit from carefully reviewing each figure to ensure that no data is incorrectly duplicated.

We thank the reviewer for catching this error. It has been corrected.

(4) The authors describe experiments to compare shock and reward learning; however, there are temporal differences in what they compare in Figure 5. The authors compare the 10th day of reward learning with the 1st day of fear conditioning, which effectively represent different points of learning and retrieval. At the end of reward conditioning, animals are utilizing a learned association to the cue, which demonstrates retrieval. On the day of fear conditioning, animals are still learning the cue at the beginning of the session, but they are not necessarily retrieving an association to a learned cue. The authors would benefit from recording at a later timepoint (to be consistent with reward learning- 10 days after fear conditioning), to more accurately compare these two timepoints. Or perhaps, it might be easier to just make the comparison between Day 1 of reward learning and Day 1 of fear learning, since they must already have these data.

We agree that there are temporal differences between the food and shock US deliveries. This is likely a reflection of the fact that the shock delivery is passive and easily resolved based on the time of the US delivery, whereas the food responses are variable because they are dependent upon the consumption of the sucrose pellet. Because of these differences the kinetics of the responses cannot be accurately compared. This is why we restricted our analysis to whether the cells were food or shock responsive. Aside from reporting the temporal differences in the signals did not draw major conclusions about the differences in kinetics. In our experimental design we counterbalanced the animals that received fear conditioning firs then food conditioning, or food conditioning then fear conditioning to ensure that order effects did not influence the outcome of the study. It is widely known that Pavlovian fear conditioning can facilitate the acquisition of conditioned stimulus responses with just a single day of conditioning. In contrast, Pavlovian reward conditioning generally progresses more slowly. Because of this we restricted our analysis to the last day of reward conditioning to the first and only day of fear conditioning. However, as stated above, we compared the responses of neurons defined as salience during day 1 of reward conditioning and fear conditioning. As would be predicted based on previous definitions of salience encoding (Lin and Nicolelis, 2008, doi: 10.1016/j.neuron.2008.04.031), we observed that the responses were highest when the US was most unexpected

(5) The authors make a claim of valence encoding in their title and throughout the paper, which is not possible to make given their experimental design. However, they would greatly benefit from actually using a decoder to demonstrate their encoding claim (decoding performance for shock-food versus shuffled labels) and simply make claims about decoding food-predictive cues and shock-predictive cues. Interestingly, it seems like relatively few CeA neurons actually show differential responses to the food and shock CSs, and that is interesting in itself.

As stated above, valence and salience encoding were defined similar to what has been previously reported (Li et al., 2019, doi: 10.7554/eLife.41223; Yang et al., 2023, doi: 10.1038/s41586-023-05910-2; Huang et al., 2024, doi: 10.1038/s41586-024-07819; Lin and Nicolelis, 2008, doi: 10.1016/j.neuron.2008.04.031; Stephenson-Jones et al., 2020, doi: 10.1016/j.neuron.2019.12.006; Zhu et al., 2018, doi: 10.1126/science.aat0481; and Comoli et al., 2003, doi: 10.1038/nn1113P). Interestingly, many of these studies did not vary the US intensity.

Reviewer #3 (Public review):

Summary:

In their manuscript entitled Kong and colleagues investigate the role of distinct populations of neurons in the central amygdala (CeA) in encoding valence and salience during both appetitive and aversive conditioning. The study expands on the work of Yang et al. (2023), which specifically focused on somatostatin (SST) neurons of the CeA. Thus, this study broadens the scope to other neuronal subtypes, demonstrating that CeA neurons in general are predominantly tuned to valence representations rather than salience.

We thank the reviewer for their insightful comments and assessment of the manuscript.

Strengths:

One of the key strengths of the study is its rigorous quantitative approach based on the "circular-shift method", which carefully assesses correlations between neural activity and behavior-related variables. The authors' findings that neuronal responses to the unconditioned stimulus (US) change with learning are consistent with previous studies (Yang et al., 2023). They also show that the encoding of positive and negative valence is not influenced by prior training order, indicating that prior experience does not affect how these neurons process valence.

Weaknesses:

However, there are limitations to the analysis, including the lack of population-based analyses, such as clustering approaches. The authors do not employ hierarchical clustering or other methods to extract meaning from the diversity of neuronal responses they recorded. Clustering-based approaches could provide deeper insights into how different subpopulations of neurons contribute to emotional processing. Without these methods, the study may miss patterns of functional specialization within the neuronal populations that could be crucial for understanding how valence and salience are encoded at the population level.

We appreciate the reviewer’s comments regarding clustering-based approaches. In order to classify cells as responsive to the US or CS we chose to develop a statistically rigorous method for classifying cell response types. Using this approach, we were able to define cell responses to the US and CS. Importantly, we identified 8 distinct response types to the USs. It is not clear how additional clustering analysis would improve cell classifications.

Furthermore, while salience encoding is inferred based on responses to stimuli of opposite valence, the study does not test whether these neuronal responses scale with stimulus intensity-a hallmark of classical salience encoding. This limits the conclusions that can be drawn about salience encoding specifically.

As stated above, we used salience classifications similar to those previously described (Lin and Nicolelis, 2008, doi: 10.1016/j.neuron.2008.04.031; Stephenson-Jones et al., 2020, doi: 10.1016/j.neuron.2019.12.006; Zhu et al., 2018, doi: 10.1126/science.aat0481; and Comoli et al., 2003, doi: 10.1038/nn1113P). We agree that varying the stimulus intensity would provide a more rigorous assessment of salience encoding; however, several of the studies mentioned above classify cells as salience encoding without varying stimulus intensity. Additionally, the inclusion of recordings with varying US intensities on top of the Pavlovian reward and fear conditioning would further decrease the number of cells that can be longitudinally tracked and would likely decrease the number of cells that could be classified.

In sum, while the study makes valuable contributions to our understanding of CeA function, the lack of clustering-based population analyses and the absence of intensity scaling in the assessment of salience encoding are notable limitations.

Reviewer #4 (Public review):

Summary:

The authors have performed endoscopic calcium recordings of individual CeA neuron responses to food and shock, as well as to cues predicting food and shock. They claim that a majority of neurons encode valence, with a substantial minority encoding salience.

Strengths:

The use of endoscopic imaging is valuable, as it provides the ability to resolve signals from single cells, while also being able to track these cells across time. The recordings appear well-executed, and employ a sophisticated circular shifting analysis to avoid statistical errors caused by correlations between neighboring image pixels.

Weaknesses:

My main critique is that the authors didn't fully test whether neurons encode valence. While it is true that they found CeA neurons responding to stimuli that have positive or negative value, this by itself doesn't indicate that valence is the primary driver of neural activity. For example, they report that a majority of CeA neurons respond selectively to either the positive or negative US, and that this is evidence for "type I" valence encoding. However, it could also be the case that these neurons simply discriminate between motivationally relevant stimuli in a manner unrelated to valence per se. A simple test of this would be to check if neural responses generalize across more than one type of appetitive or aversive stimulus, but this was not done. The closest the authors came was to note that a small number of neurons respond to CS cues, of which some respond to the corresponding US in the same direction. This is relegated to the supplemental figures (3 and 4), and it is not noted whether the the same-direction CS-US neurons are also valence-encoding with respect to different USs. For example, are the neurons excited by CS-food and US-food also inhibited by shock? If so, that would go a long way toward classifying at least a few neurons as truly encoding valence in a generalizable way.

As stated above, valence and salience encoding were defined similar to what has been previously reported (Li et al., 2019, doi: 10.7554/eLife.41223; Yang et al., 2023, doi: 10.1038/s41586-023-05910-2; Huang et al., 2024, doi: 10.1038/s41586-024-07819; Lin and Nicolelis, 2008, doi: 10.1016/j.neuron.2008.04.031; Stephenson-Jones et al., 2020, doi: 10.1016/j.neuron.2019.12.006; Zhu et al., 2018, doi: 10.1126/science.aat0481; and Comoli et al., 2003, doi: 10.1038/nn1113P). As reported in Figure 5 and Figure 5 – Supplement 3, ~29% of CeA neurons responded to both food and shock USs (15% in the same direction and 13.5% in the opposite direction). In contrast, only 6 of 303 cells responded to both the CSfood and CSshock, all in the same direction.

A second and related critique is that, although the authors correctly point out that definitions of salience and valence are sometimes confused in the existing literature, they then go on themselves to use the terms very loosely. For example, the authors define these terms in such a way that every neuron that responds to at least one stimulus is either salience or valence-encoding. This seems far too broad, as it makes essentially unfalsifiable their assertion that the CeA encodes some mixture of salience and valence. I already noted above that simply having different responses to food and shock does not qualify as valence-encoding. It also seems to me that having same-direction responses to these two stimuli similarly does not quality a neuron as encoding salience. Many authors define salience as being related to the ability of a stimulus to attract attention (which is itself a complex topic). However, the current paper does not acknowledge whether they are using this, or any other definition of salience, nor is this explicitly tested, e.g. by comparing neural response magnitudes to any measure of attention.

As stated in response to reviewer 2, we longitudinally tracked cells on the first day of Pavlovian reward conditioning the fear conditioning day. Although there were considerably fewer head entries on the first day of reward conditioning, we were able to identify 10 cells that were activated by both the food US and shock US. We compared the responses to the first five head entries and last head entries and the first 5 shocks and last five shocks. Consistent with what has been reported for salience encoding neurons in the basal forebrain (Lin and Nicolelis, 2008, doi: 10.1016/j.neuron.2008.04.031), we observed that the responses were highest when the US was most unexpected and decreased in later trials.

The impression I get from the authors' data is that CeA neurons respond to motivationally relevant stimuli, but in a way that is possibly more complex than what the authors currently imply. At the same time, they appear to have collected a large and high-quality dataset that could profitably be made available for additional analyses by themselves and/or others.

Lastly, the use of 10 daily sessions of training with 20 trials each seems rather low to me. In our hands, Pavlovian training in mice requires considerably more trials in order to effectively elicit responses to the CS. I wonder if the relatively sparse training might explain the relative lack of CS responses?

It is possible that learning would have occurred more quickly if we had used greater than 20 trials per session. However, we routinely used 20-25 trials for Pavlovian reward conditioning (doi: 10.1073/pnas.1007827107; doi: 10.1523/JNEUROSCI.5532-12.2013; doi: 10.1016/j.neuron.2013.07.044; and doi: 10.1016/j.neuron.2019.11.024).

Author response:

The following is the authors’ response to the original reviews

eLife Assessment

This useful study integrates experimental methods from materials science with psychophysical methods to investigate how frictional stabilities influence tactile surface discrimination. The authors argue that force fluctuations arising from transitions between frictional sliding conditions facilitate the discrimination of surfaces with similar friction coefficients. However, the reliance on friction data obtained from an artificial finger, together with the ambiguous correlative analyses relating these measurements to human psychophysics, renders the findings incomplete.

Our main goal with this paper was to show that the most common metric, i.e. average friction coefficient—widely used in tactile perception and device design – is fundamentally unsound, and to offer a secondary parameter that is compatible with the fact that human motion is unconstrained, leading to dynamic interfacial mechanics.

We understand the Reviewers wanted, through biomechanical measurements, to demonstrate that humans using instabilities. This is seemingly reasonable, but in individual responses, we explain the significant challenges and fundamental unknowns to those experiments. We believe this paper sets forth an important step to approach this problem. At the same time, we have made several changes in the discussion, conclusion, and title to clarify that our study is correlative between mechanical characterization and human testing.

In short, there are still several fundamental unknowns that prevented us from basing the study around biomechanical measurements: (1) a decision-making model would need to be created, but it is unknown if tactile decision making follows other models, (2) it is further unknown what constitutes “tactile evidence”, though at our manuscript’s conclusion, we propose that friction instabilities are better suited for to be tactile evidence than the averaging of friction coefficients from a narrow range of human exploration (3) in the design of samples, from a friction mechanics and materials perspective, it is not at this point, possible to pre-program surfaces a priori to deliver friction instabilities and instead must be experimentally determined – especially when attempting to achieve this in controlled surfaces that do not create other overriding tactile cues, like macroscopic bumps or large differences in surface roughness. (4) Given that the basis for tactile percepts, like which object feels “rougher” or “smoother” is not sufficiently established, it is necessary to use a 3-alternative forced choice task which avoids asking objects along a preset perceptual dimension – a challenge recognized by Reviewer 3. However, this would bring in issues of memory in the decision-making model. (5) The prior points are compounded by the fact that, we believe, tactile exploration must be performed in an unconstrained manner, i.e., without an apparatus generating motion onto a stationary finger. Work by Liu et al. (IEEE ToH, 2024) showed that recreating friction obtained during free exploration onto a stationary finger was uninterpretable by the participants, hinting at the importance of efference copies.[1] We believe that many of the above-mentioned issues constitutes a significant advance in knowledge and would require discussion and dissemination with the community.

Our changes to the manuscript

Page 1 & SI Page 1, Title

“Alternatives to Friction Coefficient: Fine Touch Perception Correlates with Frictional Instabilities”

Reviewer 1 (Public review):

Summary:

In this paper, Derkaloustian et. al look at the important topic of what affects fine touch perception. The observations that there may be some level of correlation with instabilities are intriguing. They attempted to characterize different materials by counting the frequency (occurrence #, not of vibration) of instabilities at various speeds and forces of a PDMS slab pulled lengthwise over the material. They then had humans make the same vertical motion to discriminate between these samples. They correlated the % correct in discrimination with differences in frequency of steady sliding over the design space as well as other traditional parameters such as friction coefficient and roughness. The authors pose an interesting hypothesis and make an interesting observation about the occurrences of instability regimes in different materials while in contact with PDMS, which is interesting for the community to see in the publication. It should be noted that the finger is complex, however, and there are many factors that may be quite oversimplified with the use of the PDMS finger, and the consideration and discounting of other parameters are not fully discussed in the main text or SI. Most importantly, however, the conclusions as stated do not align with the primary summary of the data in Figure 2.

Strengths:

The strength of this paper is in its intriguing hypothesis and important observation that instabilities may contribute to what humans are detecting as differences in these apparently similar samples.

We thank Reviewer 1 for their time on the manuscript, recognizing the approach we took, and offering constructive feedback. We believe that our conclusions, in fact, are supported by the primary summary of the data in Fig. 2 but we believe that our use of R² could have led to misinterpretation. The trend with friction coefficient and percent correct was indeed statistically significant but was spurious because the slope was negative. In the revision, we add clarifying comments throughout, change from R² to r as to highlight the negative trend, and adjust the figures to better focus on friction coefficient.

Finally, we added a new section to discuss the tradeoffs between using a real human finger versus a mock finger, and which situations may warrant the use of one or the other. In short, for our goal of characterizing surfaces to be used in tactile experiments, we believe a mock finger is more sustainable and practical than using real humans because human fingers are unique per participant, humans move their fingers at constantly changing pressures and velocities, and friction generated during free exploring human cannot be satisfactorily replicated by moving a sample onto a stationary finger. But, we do not disagree that for other types of experiments, characterizing a human participant directly may be more advantageous.

Weaknesses:

Comment 1

The most important weakness is that the findings do not support the statements of findings made in the abstract. Of specific note in this regard is the primary correlation in Figure 2B between SS (steady sliding) and percent correct discrimination. Of specific note in this regard is the primary correlation in Figure 2B between SS (steady sliding) and percent correct discrimination. While the statistical test shows significance (and is interesting!), the R-squared value is 0.38, while the R-squared value for the "Friction Coefficient vs. Percent Correct" plot has an R-squared of 0.6 and a p-value of < 0.01 (including Figure 2B). This suggests that the results do not support the claim in the abstract: "We found that participant accuracy in tactile discrimination was most strongly correlated with formations of steady sliding, and response times were negatively correlated with stiction spikes. Conversely, traditional metrics like surface roughness or average friction coefficient did not predict tactile discriminability."

We disagree that the trend with friction coefficient suggests the results do not support the claim because the correlation was found to be negative. However, we could have made the comparison more apparent and expanded on this point, given its novelty.

While the R² value corresponding to the “Friction Coefficient vs. Percent Correct” plot is notably higher, our results show that the slope is negative, which would be statistically spurious. This is because a negative correlation between percent correct (accuracy in discriminating surfaces) and difference in friction coefficient means that the more similar two surfaces are (by friction coefficient), the easier it would be for people to tell them apart. That is, it incorrectly concludes that two identical surfaces would be much easier to tell apart than two surfaces with greatly different friction coefficients.

This is counterintuitive to nearly all existing results, but we believe our samples were well-positioned to uncover this trend by minimizing variability, by controlling multiple physical parameters in the samples, and that the friction coefficient — typically calculated in the field as an average friction coefficient — ignores all the dynamic changes in forces present in elastic systems undergoing mesoscale friction, i.e., human touch, as seen in Fig. 1 in a mock finger and Fig. 3 in a real finger. By demonstrating this statistically spurious trend, we believe this strongly supports our premise that an alternative to friction coefficient is needed in the design of tactile psychophysics and haptic interfaces.

We believe that this could have been misinterpreted, so we took several steps to improve clarity, given the importance of this finding: we separated the panel on friction coefficient to its own panel, we changed from R² to r throughout, and we added clarifying text. We also added a small section focusing on this spurious trend.

Our changes to the manuscript

Page 1, Abstract

“In fact, the typical method of averaging friction coefficients led to a spurious correlation which erroneously suggests that distinct objects should feel identical and identical objects should feel distinct.”

Page 7

“As Fig. 1 was constructed from friction measurements, we can also calculate an average friction coefficient, µ, by averaging the friction coefficient obtained at each of the 16 combinations of masses and velocities (Table 1). This calculation is a standard approach in tactile studies for summarizing friction measurements, or in some cases, surfaces are never characterized at multiple masses and velocities. However, summarizing friction data in this manner has been considered as conceptually questionable by others from a mechanics perspective.[3] Fig. 1 shows that the type of instabilities and friction forces encountered on a single surface can vary widely depending on the conditions. As a result, large variations in the friction coefficient are expected, depending on the mass and velocity — even though measurements originate from the same surface. This variability in friction coefficient can be seen with the large interquartile range of friction coefficients, which shows that the variation in friction coefficient across a single surface is similar, or even larger, than the differences in average friction coefficient across two different surfaces. The observation that friction coefficients vary so widely on a single surface calls into question the approach of analyzing how humans may perceive two different objects based on their average friction coefficients.”

Page 9, Fig. 2 Caption

“D) GLMM of accuracy vs. difference in average friction coefficient , showing a negative correlation. E) GLMMs of accuracy vs. other commonly used material properties or parameters: ΔAverage roughness R_a, ΔHurst exponent H, and ΔWater contact angle hysteresis (º) (N = 10 participants_, _n = 600 total trials).”

Page 9

“Considering all instabilities individually, we found that only steady sliding was a positive, statistically significant predictor. (r \= 0.62, p < 0.05, shown in Fig. 2B).”

Page 10

“To compare the value of looking at frictional instabilities, we also performed GLMM fits on common approaches in the field, like a friction coefficient or material property typically used in tactile discrimination, shown in Fig. 2D-E. Interestingly, in Fig. 2D, we observed a spurious, negative correlation between friction coefficient (typically and often problematically simplified as across all tested conditions) and accuracy (r = -0.64, p < 0.01); that is, the more different the surfaces are by friction coefficient, the less people can tell them apart. This spurious correlation would be the opposite of intuition, and further calls into question the common practice of using friction coefficients in touch-related studies. Interestingly, this spurious correlation was also found by Gueorguiev et al.[21] The alternative, two-term model which includes adhesive contact area for friction coefficient[32] was even less predictive (see Fig. S6A of SI). We believe such a correlation could not have been uncovered previously as our samples are minimal in their physical variations. Yet, the dynamic changes in force even within a single sample are not considered, despite being a key feature of mesoscale friction during human touch.

We investigate different material properties in Fig. 2E. Differences in average roughness R_a (or other parameters, like root mean square roughness R_rms (Fig. S6A of SI) did not show a statistically significant correlation to accuracy. Though roughness is a popular parameter, correlating any roughness parameter to human performance here could be moot: the limit of detecting roughness differences has previously been defined as 13 nm on structured surfaces[36] and much higher for randomly rough surfaces,[49] all of which are magnitudes larger than the roughness differences between our surfaces. The differences in contact angle hysteresis – as an approximation of the adhesion contributions[50] – do not present any statistically significant effects on performance.”

Page 11-12

“Despite the correlative nature of this study, we still obtained high correlations compared to existing biomechanical studies[4,19,21], which we speculate is because instabilities are an important predictive phenomenon for models of human touch. We believe that biomechanical studies, including more sophisticated techniques, like spatially resolved force maps from digital image correlation[5,42] may yield stronger correlations and results if they analyze data based on instabilities.

Added References

(2) Khamis, H. et al. Friction sensing mechanisms for perception and motor control: passive touch without sliding may not provide perceivable frictional information. J. Neurophysiol. 125, 809– 823 (2021).

(6) Olczak, D., Sukumar, V. & Pruszynski, J. A. Edge orientation perception during active touch. J. Neurophysiol. 120, 2423–2429 (2018).

Comment 2, Part 1

Along the same lines, other parameters that were considered such as the "Percent Correct vs. Difference in Sp" and "Percent Correct vs. Difference in SFW" were not plotted for consideration in the SI. It would be helpful to compare these results with the other three metrics in order to fully understand the relationships.

We have added these plots to the SI. We note that we had checked these relationships and discussed them briefly, but did not include the plot. The plots show that the type of instability was not as helpful as its presence or absence.

Our changes to the manuscript

Page 9

“Furthermore, a model accounting for slow frictional waves alone specifically shows a significant, negative effect on performance (p < 0.01, Fig. S5 of SI), suggesting that in these samples and task, the type of instability was not as important.”

“Fig. S5. GLMM fits of participant accuracy vs. the differences in instability incidence for individual instability types. Left: accuracy vs. differences in formation of slow frictional waves (SFW) between pairs. P1 and P5 have the same x-axis value and are shifted for clarity. Right: accuracy vs. differences in formation of stiction spikes (Sp).”

SI Page 4

“and no correlation between accuracy and stiction spikes (Fig. S5).”

Comment 2, Part 2

Other parameters such as stiction magnitude and differences in friction coefficient over the test space could also be important and interesting.

We agree these are interesting and have thought about them. We are aware that others, like Gueorguiev et al., have studied stiction magnitudes, and though there was a correlation, the physical differences in surface roughness (glass versus PMMA) investigated made it unclear if these could be generalized further.[3] We are unsure how to proceed here with a satisfactory analysis of stiction magnitude, given that stiction spikes are not always generated. In fact, Fig. 1 shows that for many velocities and pressures, stiction spikes are not formed. In ongoing work, however, we are always cognizant that if stiction spikes are a dominant factor, then a secondary analysis on their magnitude would be important. We offer some speculation on why stiction spikes may be overrepresented in the literature:

(1) They are prone to being created if the finger was loaded for a long time onto a surface prior to movement, thus creating adhesion by contact aging which is unlike active human exploration. We avoid this by discarding the first pull in our measurements, which is a standard practice in mechanical characterization if contact aging needs to be avoided.

(2) The ranges of velocities and pressures explored by others were small.

(3) In an effort to generate strong tactile stimuli, highly adhesive or rough surfaces are used.

(4) Stiction spikes are visually distinctive on a plot, but we are unaware of any mechanistic reason that mechanoreceptors would be particularly sensitive to this low frequency event over other signals.

We interpret “difference in friction coefficient over the test space” to be, for a single surface, like C4, to find the highest average friction for a condition of single velocity and mass and subtract that from the lowest average friction for a condition of single velocity and mass. We calculated the difference in friction coefficient in the typical manner of the field, by averaging all data collected at all velocities and masses and assigning a single value for all of a surface, like C4. We had performed this, and have the data, but we are wary of overinterpreting secondary and tertiary metrics because they do not have any fundamental basis in traditional tribology, and this value, if used by humans, would suggest that they rapidly explore a large parameter space to find a “maximum” and “minimum” friction. Furthermore, the range in friction across the test space, after averaging, can be smaller than the range of friction experienced at different masses and velocities on a single surface. We have tabulated and newly included these values (the interquartile range of friction coefficients of different masses and velocities per surface) in Table 1.

Fig. 2D shows a GLMM fit between percent correct responses across our pairs and the differences in friction coefficient for each pair, where we see a spurious negative correlation. As we had the data of all average friction coefficients for each condition for a given material, we also looked at the difference in maximum and minimum friction coefficients. For our tested pairs, these differences also lined up on a statistically significant, negative GLMM fit (r = -0.86, p < 0.005). However, the values for a given surface can vary drastically, with an interquartile range of 1.20 to 2.09 on a single surface. We fit participant accuracy to the differences in these IQRs across pairs. This also led to a negative GLMM fit (r = -0.65, p < 0.05). However, we are hesitant to add this plot to the manuscript for the reasons stated previously.

Comment 3, Part 1

Beyond this fundamental concern, there is a weakness in the representativeness of the PDMS finger, the vertical motion, and the speed of sliding to real human exploration.

Overall, this is a continuous debate that we think offers two solutions, and we are not advocating for an “either-or” case. There is always a tradeoff between using a synthetic model of a finger versus a real human finger, and there is a place for both models. That is, while our mock finger will be “better” the more similar it is to a human finger, it is not our goal to fully replace a human finger. Rather our goal is to provide a consistent method of characterizing surfaces that is sufficiently similar to human touch as to be a useful and predictive tool.

The usefulness of the mock finger is in isolating the features of each surface that is independent of human variability, i.e., instabilities that form without changing loading conditions between sliding motions or even within one sliding motion. Of course, with this method, we still require confirmation of these features still forming during human exploration, which we show in Fig. 3. We believe that this method of characterizing surfaces at the mesoscale will ultimately lead to more successful human studies on tactile perception. Currently, and as shown in the paper, characterizing surfaces through traditional techniques, such as a commercial tribometer (friction coefficient, using a steel or hard metal ball), roughness (via atomic force microscopy or some other metrology), surface energy are less or not at all predictive. Thus, we believe this mock finger is better than the current state-of-the-art characterizing surfaces (we are also aware of a commercial mock finger company, but we were unable to purchase or obtain an evaluation model).

One of the main – and severe – limitations of using a human finger is that all fingers are different, meaning any study focusing on a particular user may not apply to others or be recreated easily by other researchers. We do not think it is feasible to set a standard for replication around a real human finger as that participant may no longer be available, or willing to travel the world as a “standard”. Furthermore, the method in which a person changes their pressures and velocities is different. We note that this is a challenge unique to touch perception – how an object is touched changes the friction generated, and thus the tactile stimulus generated, whereas a standardized stimulus is more straightforward for light or sound.

However, we do emphasize that we have strongly considered the balance between feasibility and ecological validity in the design of a mock finger. We have a mock finger, with the three components of stiffness of a human finger (more below). Furthermore, we have also successfully used this mock finger in correlations with human psychophysics in previous work, where findings from our mechanical experiments were more predictive of human performance[4–7] than other available methods.

Our changes to the manuscript Added (Page 2-3)

“Mock finger as a characterization tool

We use a mechanical setup with a PDMS (poly(dimethylsiloxane)) mock finger to derive tactile predictors as opposed to direct biomechanical measurements on human participants. While there is a tradeoff in selecting a synthetic finger over a real human finger to modeling human touch, human fingers themselves are also highly variable[23] both in their physical shape and their use during human motion. Our goal is to design a consistent method of characterization of samples that can be easily accessed by other researchers and does not rely on a standard established around single human participant. We believe that sufficient replication of surface, bulk properties, and contact geometry results in characterization that isolates consistent features of surfaces that are not derived from human-to-human variability. We have used this approach to successfully correlate human results with mock finger characterization previously.[8,9,24]

The major component of a human finger, by volume, is soft tissue (~56%),[25] resulting in an effective modulus close to 100 kPa.[26,27] In order to achieve this same softness, we crosslink PDMS in a 1×1×5 cm mold at a 30:1 elastomer:crosslinker ratio. In addition, two more features in the human finger impart significant mechanical differences. Human fingers have a bone at the fingertip, the distal phalanx,[26–28, 8–10]which we mimic with an acrylic “bone” within our PDMS network. The stratum corneum, the stiffer, glassier outer layer of skin,[29] is replicated with the surface of the mock finger glassified, or further crosslinked, after 8 hours of UV-Ozone treatment.30 This treatment also modifies the surface properties of the native PDMS to align with those of a human finger more closely: it minimizes the viscoelastic tack at the surface, resulting in a comparable non-sticky surface. Stabilizing after one day after treatment, the mock finger surface obtains a moderate hydrophilicity (~60º), as is typically observed for a real finger.[11,31]

The initial contact area formed before a friction trace is collected is a rectangle of 1×1 cm. While this shape is not entirely representative of a human finger with curves and ridges, human fingers flatten out enough to reduce the effects of curvature with even very light pressures.[31–33] This implies that for most realistic finger pressures, the contact area is largely load-independent, which is more accurately replicated with a rectangular mock finger.

Lastly, we consider the role of fingerprint ridges. A key finding of our previous work is that while fingerprints enhanced frictional dynamics at certain conditions, key features were still maintained with a flat finger.[11] Furthermore, for some loading conditions, the more amplified signals could also result in more similar friction traces for different surfaces. We have observed good agreement between these friction traces and human experiments.[8,9,22,34]”

Page 3-4, Materials and Methods

“Mock Finger Preparation

Friction forces across all six surfaces were measured using a custom apparatus with a polydimethylsiloxane (PDMS, Dow Sylgard 184) mock finger that mimics a human finger’s mechanical properties and contact mechanics while exploring a surface relatively closely.[8,9] PDMS and crosslinker were combined in a 30:1 ratio to achieve a stiffness of 100 kPa comparable to a real finger, then degassed in a vacuum desiccator for 30 minutes. We are aware that the manufacturer recommended crosslinking ratio for Sylgard 184 is 10:1 due to potential uncrosslinked liquid residues,[35] but further crosslinking concentrated at the surface prevents this. The prepared PDMS was then poured into a 1×1×5 cm mold also containing an acrylic 3D-printed “bone” to attach applied masses on top of the “fingertip” area contacting a surface during friction testing. After crosslinking in the mold at 60ºC for 1 hour, the finger was treated with UV-Ozone for 8 hours out of the mold to minimize viscoelastic tack.

Mechanical Testing

A custom device using our PDMS mock finger was used to collect macroscopic friction force traces replicating human exploration.[8,9] After placing a sample surface on a stage, the finger was lowered at a slight angle such that an initial 1×1 cm rectangle of “fingertip” contact area could be established. We considered a broad range of applied masses (M \= 0, 25, 75, and 100 g) added onto the deadweight of the finger (6 g) observed during a tactile discrimination task. The other side of the sensor was connected to a motorized stage (V-508 PIMag Precision Linear Stage, Physikinstrumente) to control both displacement (4 mm across all conditions) and sliding velocity (v \= 5, 10, 25, and 45 mm s^-1). Forces were measured at all 16 combinations of mass and velocity via a 250 g Futek force sensor (k \= 13.9 kN m^-1) threaded to the bone, and recorded at an average sampling rate of 550 Hz with a Keithley 7510 DMM digitized multimeter. Force traces were collected in sets of 4 slides, discarding the first due to contact aging. Because some mass-velocity combinations were near the boundaries of instability phase transitions, not all force traces at these given conditions exhibited similar profiles. Thus, three sets were collected on fresh spots for each condition to observe enough occurrences of multiple instabilities, at a total of nine traces per combination for each surface.”

Added References

(23) Infante, V. H. P. et al. The role of skin hydration, skin deformability, and age in tactile friction and perception of materials. Sci. Rep. 15, 9935 (2025).

(24) Nolin, A., Lo, C.-Y., Kayser, L. V. & Dhong, C. B. Transparent and Electrically Switchable Thin Film Tactile Actuators Based on Molecular Orientation. Preprint at https://doi.org/10.48550/arXiv.2411.07968 (2024).

(25) Murai, M., Lau, H.-K., Pereira, B. P. & Pho, R. W. H. A cadaver study on volume and surface area of the fingertip. J. Hand Surg. 22, 935–941 (1997).

(26) Abdouni, A. et al. Biophysical properties of the human finger for touch comprehension: influences of ageing and gender. R. Soc. Open Sci. (2017) doi:10.1098/rsos.170321.

(27) Cornuault, P.-H., Carpentier, L., Bueno, M.-A., Cote, J.-M. & Monteil, G. Influence of physico-chemical, mechanical and morphological fingerpad properties on the frictional distinction of sticky/slippery surfaces. J. R. Soc. Interface (2015) doi:10.1098/rsif.2015.0495.

(28) Qian, K. et al. Mechanical properties vary for different regions of the finger extensor apparatus. J. Biomech. 47, 3094–3099 (2014).

(29) Yuan, Y. & Verma, R. Measuring microelastic properties of stratum corneum. Colloids Surf. B Biointerfaces 48, 6–12 (2006).

(30) Fu, Y.-J. et al. Effect of UV-Ozone Treatment on Poly(dimethylsiloxane) Membranes: Surface Characterization and Gas Separation Performance. Langmuir 26, 4392–4399 (2010).

Comment 3, Part 2

The real finger has multiple layers with different moduli. In fact, the stratum corneum cells, which are the outer layer at the interface and determine the friction, have a much higher modulus than PDMS. The real finger has multiple layers with different moduli. In fact, the stratum corneum cells, which are the outer layer at the interface and determine the friction, have a much higher modulus than PDMS.

We have approximated the softness of the finger with 100 kPa crosslinked PDMS, which is close to what has been reported for the bulk of a human fingertip.[9,10] However, as mentioned in the Materials and Methods, there are two additional features of the mock finger that impart greater strength. The PDMS surrounds a rigid, acrylic bone comparable to the distal phalanx, which provides an additional layer of higher modulus.[8] Additionally, the 8-hour UV-Ozone treatment decreases the viscoelastic tack of the pristine PDMS by glassifying, or further crosslinking the surface of the finger,[12] therefore imparting greater stiffness at the surface similar to the contributions of the stratum corneum, along with a similar surface energy.[13] This technique is widely used in wearables,[14] soft robotics,[15] and microfluidics[16] to induce both these material changes. Additionally, the finger is used at least a day after UV-Ozone treatment is completed to generate a stable surface that is moderately hydrophilic, similar to the outermost layer of human skin.[17]

Comment 3, Part 3

In addition, the slanted position of the finger can cause non-uniform pressures across the finger. Both can contribute to making the PDMS finger have much more stick-slip than a real finger.

To ensure that there is minimal contribution from the slanted position of the finger, an initial contact area of 1×1 cm is established before sliding and recording friction measurements. As the PDMS finger is a soft object, the portion in contact with a surface flattens and the contact area remains largely unchanged during sliding. Any additional stick-slip after this alignment step is caused by contact aging at the interface, but the first trace we collect is always discarded to only consider stick-slip events caused by surface chemistry. We recognize that it is difficult to completely control the pressure distribution due to the planar interface, but this is also expected when humans freely explore a surface.

Comment 3, Part 4

In fact, if you look at the regime maps, there is very little space that has steady sliding. This does not represent well human exploration of surfaces. We do not tend to use a force and velocity that will cause extensive stick-slip (frequent regions of 100% stick-slip) and, in fact, the speeds used in the study are on the slow side, which also contributes to more stick-slip. At higher speeds and lower forces, all of the materials had steady sliding regions.”

We are not aware of published studies that extensively show that humans avoid stickslip regimes. In fact, we are aware familiar with literature where stiction spike formation is suppressed – a recent paper by AliAbbasi, Basdogan et. al. investigates electroadhesion and friction with NaCl solution-infused interfaces, resulting in significantly steadier forces.[18] We also directly showed evidence of instability formation that we observed during human exploration in Fig. 3B-C. These dynamic events are common, despite the lack of control of normal forces and sliding velocities. We also note that Reviewer 1, Comment 2, Part 2 was suggesting that we further explore possible trends from parameterizing the stiction spike.

We note that many studies have often not gone at the velocities and masses required for stiction spikes – even though these masses and velocities would be routinely seen in free exploration – this is usually due to constraints of their equipment.[19] Sliding events during human free exploration of surfaces can exceed 100 mm/s for rapid touches. However, for the surfaces investigated here, we observe that large regions of stick-slip can emerge at velocities as low as 5 mm/s depending on the applied load. The incidence of steady sliding appears more dependent on the applied mass, with almost no steady sliding observed at or above 75 g. Indeed, the force categorization along our transition zones is the main point of the paper.

Comment 3, Part 5

Further, on these very smooth surfaces, the friction and stiction are more complex and cannot dismiss considerations such as finger material property change with sweat pore occlusion and sweat capillary forces. Also, the vertical motion of both the PDMS finger and the instructed human subjects is not the motion that humans typically use to discriminate between surfaces.

We did not describe the task sufficiently. Humans were only given the instruction to slide their finger along a single axis from top to bottom of a sample, not vertical as in azimuthal to gravity. We have updated our wording in the manuscript to reflect this.

Page 4

“Participants could touch for as long as they wanted, but were asked to only use their dominant index fingers along a single axis to better mimic the conditions for instability formation during mechanical testing with the mock finger.”

Page 11

“The participant was then asked to explore each sample simultaneously, and ran over each surface in strokes along a single axis until the participant could decide which of the two had “more friction”.”

Comment 3, Part 6

Finally, fingerprints may not affect the shape and size of the contact area, but they certainly do affect the dynamic response and detection of vibrations.”

We are aware of the nuance. Our previous work on the role of fingerprints on friction experienced by a PDMS mock finger showed enhanced signals with the incorporation of ridges on the finger and used a rate-and-state model of a heterogenous, elastic body to find corresponding trends (though there is no existing model of friction that can accurately model experiments on mesoscale friction).[11] The key conclusion was that a flat finger still preserved key dynamic features, and the presence of stronger or more vibrations could result in more similar forces for different surfaces depending on the sliding conditions.

This is also in the context that we are seeking to provide a reasonable and experimentally accessible method to characterize surfaces, which will always be better as we get closer in replicating a true human finger. But our goal here was to replicate the finger sufficiently for use in human studies. We believe the more appropriate metric of success is if the mock finger is more successful than replacing traditional characterization experiments, like friction coefficient, roughness, surface energy, etc.

Comment 4

This all leads to the critical question, why are friction, normal force, and velocity not measured during the measured human exploration and in a systematic study using the real human finger? The authors posed an extremely interesting hypothesis that humans may alter their speed to feel the instability transition regions. This is something that could be measured with a real finger but is not likely to be correlated accurately enough to match regime boundaries with such a simplified artificial finger.

We are excited that our manuscript offers a tractable manner to test the hypothesis that tactile decision-making models use friction instabilities as evidence. However, we lay out the challenges and barriers, and how the scope of this paper will lead us in that direction. We also clarify that our goals are to provide a method to characterize samples to better design tactile interfaces in haptics or in psychophysical experiments and raise awareness that the common methods of sample characterization in touch by an average friction coefficient or roughness is fundamentally unsound. Throughout the paper, we have made changes to reflect that our study, at this point, is only correlative.

As discussed in the summary, and with additional detail here, to further support our findings through observation on humans would require answering:

(1) Which one, or combination of, of the multiple swipes that people make responsible for a tactile decision? (There is a need for a decision-making model)

(2) Establish what is, or may be, tactile evidence.

(3) Establish tactile decision-making models are similar or different than existing decision-making models.

(4) Design a task that does not require the use of subjective tactile descriptors, like “which one feels rougher”, which we have seen causes confusion in participants, which will likely require accounting for memory effects.

We elaborate these points below:

To successfully perform this experiment, we note that freely exploring humans make multiple strokes on a surface. Therefore, we would need to construct a decision-making model. It has not yet been demonstrated whether tactile decision making follows visual decision making, but perhaps to start, we can assume it does. Then, in the design of our decision-making paradigm, we immediately run into the problem: What is tactile evidence?

From Fig. 3C, we already can see that identifying evidence is challenging. Prior to this manuscript, people may have chosen the average force, or the highest force. Or we may choose the average friction force. Then, after deciding on the evidence, we need to find a method to manipulate the evidence, i.e., create samples or a machine that causes high friction, etc. We show that during the course of human touch, due to the dynamic nature of friction, the average can change a large amount and sample design becomes a central barrier to experiments. Others may suggest immobilizing the finger and applying a known force, but given how much friction changes with human exploration, there is no known method to make a machine recreate temporally and spatially varying friction forces during sliding onto a stationary finger. Finally, perhaps most importantly, in addition to mechanical challenges, a study by Liu, Colgate et al. showed that even if they recorded the friction (2D) of a finger exploring a surface and then replicated the same friction forces onto a finger, the participant could not determine which surface the replayed friction force was supposed to represent.[1] This supports that the efference copy is important, that the forces in response to expected motion are important to determine friction. Finally, there is no known method to design instabilities a priori. They must be found through experiments. Especially since if we were to introduce, say a bump or a trough, then we bring in confounding variables to how participants tell surfaces apart.

Furthermore, even if we had some consistent method to create tactile “evidence”, the paradigm also deserves some consideration. In our experience, the 3-AFC task we perform is important because the vocabulary for touch has not been established. That is, in 3-AFC, by asking to determine which one sample is unlike the others, we do not have to ask the participant questions like “which one is rougher” or “which one has less friction”. In contrast, 2-AFC, which is better for decision-making models because it does not include memory, requires the asking of a perceptual question like: “which one is rougher?”. In our ongoing work, taking two silane coatings, we found that participants could easily identify which surface is unlike the others above chance in a 3-AFC, but participants, even within their own trials, could not consistently identify one silane as perceptually “rougher” by 2-AFC. To us, this calls into question the validity of tactile descriptors, but is beyond the scope of this manuscript.

This is not our only goal, but in the context of human exploration, in this manuscript here, we believed it was important to identify a mechanical parameter that was consistent with how humans explore surfaces, but was also a parameter that could characterize to some consistent property of a surface – irrespective of whether a human was touching it. We thought that designing human decision-making models and paradigms around the friction coefficient would not be successful.

Given the scope of these challenges, we do not think it would be possible to establish these conceptual sequences in a single manuscript. However, we think that our manuscript brings an important step forward to approach this problem.

Reviewer 2 (Public review):

Summary:

In this paper, the authors want to test the hypothesis that frictional instabilities rather than friction are the main drivers for discriminating flat surfaces of different sub-nanometric roughness profiles.

They first produced flat surfaces with 6 different coatings giving them unique and various properties in terms of roughness (picometer scale), contact angles (from hydrophilic to hydrophobic), friction coefficient (as measured against a mock finger), and Hurst exponent.

Then, they used those surfaces in two different experiments. In the first experiment, they used a mock finger (PDMS of 100kPA molded into a fingertip shape) and slid it over the surfaces at different normal forces and speeds. They categorized the sliding behavior as steady sliding, sticking spikes, and slow frictional waves by visual inspection, and show that the surfaces have different behaviors depending on normal force and speed. In a second experiment, participants (10) were asked to discriminate pairs of those surfaces. It is found that each of those pairs could be reliably discriminated by most participants.

Finally, the participant's discrimination performance is correlated with differences in the physical attributes observed against the mock finger. The authors found a positive correlation between participants' performances and differences in the count of steady sliding against the mock finger and a negative correlation between participants' reaction time and differences in the count of stiction spikes against the mock finger. They interpret those correlations as evidence that participants use those differences to discriminate the surfaces.

Strengths:

The created surfaces are very interesting as they are flat at the nanometer scale, yet have different physical attributes and can be reliably discriminated.

We thank Reviewer 2 for their notes on our manuscript. The responses below address the reviewer’s comments and recommendations for revised work.

Weaknesses:

Comment 1

In my opinion, the data presented in the paper do not support the conclusions. The conclusions are based on a correlation between results obtained on the mock finger and results obtained with human participants but there is no evidence that the human participants' fingertips will behave similarly to the mock finger during the experiment. Figure 3 gives a hint that the 3 sliding behaviors can be observed in a real finger, but does not prove that the human finger will behave as the mock finger, i.e., there is no evidence that the phase maps in Figure 1C are similar for human fingers and across different people that can have very different stiffness and moisture levels.

We have made changes throughout the manuscript to acknowledge that our findings are correlative, clarifying this throughout, and incorporating into the discussion how our work may enable biomechanical measurements and tactile decision making models.

The mechanical characterization conducted with the mock finger seeks to extract significant features of friction traces of a set of surfaces to use as predictors of tactile discriminability. The goal is to find a consistent method to characterize surfaces for use in tactile experiments that can be replicated by others and used prior to any human experiments. However, in the overall response and in a response to a similar comment by Reviewer 1 (recreated below), we also explain why we believe experiments on humans to establish this fact is not yet reasonable.

First, we discuss the mock finger. The PDMS finger is treated to have comparable surface and bulk properties to a human finger. We have approximated the softness of the finger with 100 kPa crosslinked PDMS, which is close to what has been reported for the bulk of a human fingertip.[9,10] However, as mentioned in the Materials and Methods, there are two additional features of the mock finger that impart greater strength. The PDMS surrounds a rigid, acrylic bone comparable to the distal phalanx, which provides an additional layer of higher modulus.[8] Additionally, the 8-hour UV-Ozone treatment decreases the viscoelastic tack of the pristine PDMS by glassifying, or further crosslinking the surface of the finger,[12] therefore imparting greater stiffness at the surface similar to the contributions of the stratum corneum, along with a similar surface energy.[13] Additionally, the finger is used at least a day after UV-Ozone treatment is completed in order for the surface to return to moderate hydrophilicity, similar to the outermost layer of human skin.[17] We also discuss the shape of the contact formed. To ensure that there is minimal contribution from the slanted position of the finger, an initial contact area of 1×1 cm is established before sliding and recording friction measurements. As the PDMS finger is a soft object, the portion in contact with a surface flattens and the contact area remains largely unchanged during sliding. Any additional stick-slip after this alignment step is caused by contact aging at the interface, but the first trace we collect is always discarded to only consider stick-slip events caused by surface chemistry. We recognize that it is difficult to completely control the pressure distribution due to the planar interface, but this is also expected when humans freely explore a surface. Finally, we consider flat vs. fingerprinted fingers. Our previous work on the role of fingerprints on friction experienced by a PDMS mock finger showed enhanced signals with the incorporation of ridges on the finger and used a rate-and-state model of a heterogenous, elastic body to find corresponding trends.[11] The key conclusion was that a flat finger still preserved key dynamic features, and the presence of stronger or more vibrations could result in more similar forces for different surfaces depending on the sliding conditions. We note that we have subsequently used this flat mock finger in correlations with human psychophysics in previous work, where findings from our mechanical experiments were predictive of human performance.[4–7] We have added these details to the manuscript.

With this adequately similar mock finger, we collected friction traces at controlled conditions of normal force and velocity in order to extract the signals unique to each material that are not caused by the influence of human variability. For example, we observe the smallest regions of steady sliding on our phase maps (Fig. 1C) for short-chain alkylsilanes C4 and C5, while the increased intermolecular forces of other silanes increase the incidence of steady sliding. We have also previously shown that comparisons of similarly collected mechanical data is predictive of human performance, using the crosscorrelations between signals of two different materials.[4–7] While different participants produce different raw signals, we see that broad categories of stick-slip, i.e. instabilities, can be extracted (Fig. 3B-C) and used as a cue in a tactile discrimination task. As mentioned above, we have provided an additional section about the usefulness of our mock finger, as well as its structure, in the main manuscript.

Second, we lay out the challenges and barriers to demonstrating this in humans in the manner requested by the reviewer, and how the scope of this paper will lead us in that direction. We also clarify that our goals are to provide a method to characterize samples to better design tactile interfaces in haptics or in psychophysical experiments and raise awareness that the common methods of sample characterization in touch by an average friction coefficient or roughness is fundamentally unsound.

As discussed in the summary, and with additional detail here, to further support our findings through observation on humans would require answering:

(1) Which one, or combination of, of the multiple swipes that people make responsible for a tactile decision?

(2) Establish what is, or may be, tactile evidence.

(3) Establish tactile decision-making models are similar or different than existing decision-making models.

(4) Test the hypothesis, in these models, that friction instabilities are evidence, and not some other unknown metric.

(5) Design a task that does not require the use of subjective tactile descriptors, like “which one feels rougher”, which we see cause confusion in participants, which will likely require accounting for memory effects.

We elaborate these points below:

To successfully perform this experiment, we note that freely exploring humans make multiple strokes on a surface. Therefore, we would need to construct a decision-making model. It has not yet been demonstrated whether tactile decision making follows visual decision making, but perhaps to start, we can assume it does. Then, in the design of our decision-making paradigm, we immediately run into the problem: What is tactile evidence?

From Fig. 3C, we already can see that identifying evidence is challenging. Prior to this manuscript, people may have chosen the average force, or the highest force. Or we may choose the average friction force. Then, after deciding on the evidence, we need to find a method to manipulate the evidence, i.e., create samples or a machine that causes high friction, etc. We show that during the course of human touch, due to the dynamic nature of friction, the average can change a large amount and sample design becomes a central barrier to experiments. Others may suggest immobilizing the finger and applying a known force, but given how much friction changes with human exploration, there is no known method to make a machine recreate temporally and spatially varying friction forces during sliding onto a stationary finger. Finally, perhaps most importantly, in addition to mechanical challenges, a study by Liu, Colgate, et al. showed that even if they recorded the friction (2D) of a finger exploring a surface and then replicated the same friction forces onto a finger, the participant could not determine which surface the replayed friction force was supposed to represent.[1] This supports that the efference copy is important, that the forces in response to expected motion are important to determine friction. Finally, there is no known method to design instabilities a priori. They must be found through experiments, especially since if we were to introduce, say a bump or a trough, then we bring in confounding variables to how participants tell surfaces apart.

Furthermore, even if we had some consistent method to create tactile “evidence”, the paradigm also deserves some consideration. In our experience, the 3-AFC task we perform is important because the vocabulary for touch has not been established. That is, in 3-AFC, by asking to determine which one sample is unlike the others, we do not have to ask the participant questions like “which one is rougher” or “which one has less friction”. In contrast, 2-AFC, which is better for decision-making models because it does not include memory, requires the asking of a perceptual question like: “which one is rougher?”. In our ongoing work, taking two silane coatings, we found that participants could easily identify which surface is unlike the others above chance in a 3-AFC, but participants, even within their own trials, could not consistently identify one silane as perceptually “rougher” by 2-AFC. To us, this calls into question the validity of tactile descriptors, but is beyond the scope of the current manuscript.

This is not our only goal, but in the context of human exploration, in this manuscript here, we believed it was important to identify a mechanical parameter that was consistent with how humans explore surfaces, but was also a parameter that could characterize to some consistent property of a surface – irrespective of whether a human was touching it. We thought that designing human decision-making models and paradigms around the friction coefficient would not be successful.

Given the scope of these challenges, we do not think it would be possible to establish this conceptual sequence in a single manuscript.

See Reviewer 1, comment 3part 3 for changes to the manuscript

Comment 2

I believe that the authors collected the contact forces during the psychophysics experiments, so this shortcoming could be solved if the authors use the actual data, and show that the participant responses can be better predicted by the occurrence of frictional instabilities than by the usual metrics on a trial by trial basis, or at least on a subject by subject basis. I.e. Poor performers should show fewer signs of differences in the sliding behaviors than good performers.

To fully implement this, a decision-making model is necessary because, as a counter example, a participant could have generated 10 swipes of SFW and 1 swipe of a Sp, but the Sp may have been the most important event for making a tactile decision. This type of scenario is not compatible with the analysis suggested — and similar counterpoints can be made for other types of seemingly straightforward analysis.

While we are interested and actively working on this, the study here is critical to establish types of evidence for a future decision-making model. We know humans change their friction constantly during real exploration, so it is unclear which of these constantly changing values we should input into the decision making model, and the future challenges we anticipate are explained in Weaknesses, Comment 1.

Comment 3

The sample size (10) is very small.

We recognize that, with all factors being equal, this sample size is on the smaller end. However, we emphasize the degree of control of samples is far above typical, with minimal variations in sample properties such as surface roughness, and every sample for every trial was pristine. Furthermore, the sample preparation (> 300 individual wafers were used) became a factor. Although not typically appropriate, and thus not included in the manuscript, a post-hoc power analysis for our 100 trials of our pair that was closest to chance, P4, (53%, closest to chance at 33%) showed a power of 98.2%, suggesting that the study was appropriately powered.

Reviewer 2 (Recommendations for the authors):

Comment 1

Differences in SS and Sp (Table 2) are NOT physical or mechanical differences but are obtained by counting differences in the number of occurrences of each sliding behavior. It is rather a weird choice.

We disagree that differences in SS and Sp are not physical or mechanical, as these are well-established phenomena in the soft matter and tribology literature.[20–22] These are known as “mechanical instabilities” and generated due to the effects of two physical phenomena: the elasticity of the finger (which is constant in our mechanical testing) and the friction forces present (which change per sample type). The motivation behind using these different shapes is that the instabilities, in some conditions, can be invariant to external factors like velocity. This would be quite advantageous for human exploration because, unlike friction coefficient, which changes with nearly any factor, including velocity and mass, the instabilities being invariant to velocity would mean that we are accurately characterizing a unique identifier of the surface even though velocity may be variable.

This “weird choice” is the central innovation of this paper. This choice was necessary because we demonstrated that the common usage of friction coefficient is fundamentally flawed: we see that friction coefficient suggests that surface which are more different would feel more similar – indeed the most distinctive surfaces would be two surfaces that are identical, which is clearly spurious. Furthermore, Table 1 now includes the range of friction generated on a surface, the range of friction coefficients of a single surface is large – of order the differences in friction between two surfaces. This is expected in soft sliding systems and emphasizes our issue with the use of average friction coefficient in psychophysical design. One potential explanation for why we were able to see this is effect is because our surfaces have similar (< 0.6 nm variability) roughness, removing potential confounding factors from large scale roughness, and this type of low roughness control has not been widely used in tactile studies to the best of our knowledge.

Comment 2

Figures 2B-C: why are the x-data different than Table 2?

The x-data in Fig. 2B-C are the absolute differences in the number of occurrences measured for a given instability type or material property out of 144 pulls. Modeling the human participant results in our GLMMs required the independent variables to be in this form rather than percentages. We initially chose to list percent differences in Table 2 to highlight the ranges of differences instead of an absolute value, but have added both for clarity.

Our changes to the manuscript

Page 7

“To determine if humans can detect these three different instabilities, we selected six pairs of surfaces to create a broad range of potential instabilities present across all three types. These are summarized in Table 2, where the first column for each instability is the difference in occurrence of that instability formed between each pair, and the second is the percent difference.”

“Thus, when comparing C4 versus C4-APTMS, they have a difference in steady sliding of 20 out of a maximum 144 pulls, for a |ΔSS| of 13.9%. The absolute value is taken to compare total differences present, as the psychophysical task does not distinguish between sample order.”

Comment 3

We constructed a set of coated surfaces with physical differences which were imperceptible by touch but created different types of instabilities based on how quickly a finger is slid and how hard a human finger is pressed during sliding." Yet, in your experiment, participants could discriminate them, so this is incoherent.

To clarify the point, macroscopic objects can differ in physical shape and in chemical composition. What we meant was that the physical differences, i.e., roughness, were below a limit (Skedung et al.) that participants, without a coating, would not be able to tell these apart.[23] Therefore, the reason people could tell our surfaces apart was due to the chemical composition of the surface, and not any differences in roughness or physical effects like film stiffness (due to the molecular-scale thinness of the surface coatings, they are mechanically negligible). However, we concede that at the molecular scale, the traditional macroscopic distinction between physical and chemical is blurred.

We have made minor revisions to the wording in the abstract. We clarify that the surface coatings had physical differences in roughness that were smaller than 0.6 nm, which based purely on roughness, would not be expected to be distinguishable to participants. Therefore, the reason participants can tell these surfaces apart is due to differences in friction generated by chemical composition, and we were able to minimize contributions from physical differences in the sample our study.

Our changes to the manuscript

Page 1, Abstract

“Here, we constructed a set of coated surfaces with minimal physical differences that by themselves, are not perceptible to people, but instead, due to modification in surface chemistry, the surfaces created different types of instabilities based on how quickly a finger is slid and how hard a human finger is pressed during sliding.”

“In one experiment, we used a mechanical mock finger to quantify and classify differences in instability formation from different coated surfaces. In a second experiment, participants perform a discrimination task using the same coated surfaces. Using the data from these two experiments, we found that human discrimination response times were faster with surfaces where the mock finger produced more stiction spikes and discrimination accuracy was higher where the mock finger produced more steady sliding. Conversely, traditional metrics like surface roughness or average friction coefficient did not relate to tactile discriminability. In fact, the typical method of averaging friction coefficients led to a spurious correlation which erroneously suggests that distinct objects should feel identical and identical objects should feel distinct—similar to findings by others. Friction instabilities may offer a more predictive and tractable framework of fine touch perception than friction coefficients, which would accelerate the design of tactile interfaces.”

Reviewer 3 (Public review):

Strengths

The paper describes a new perspective on friction perception, with the hypothesis that humans are sensitive to the instabilities of the surface rather than the coefficient of friction. The paper is very well written and with a comprehensive literature survey.

One of the central tools used by the author to characterize the frictional behavior is the frictional instabilities maps. With these maps, it becomes clear that two different surfaces can have both similar and different behavior depending on the normal force and the speed of exploration. It puts forward that friction is a complicated phenomenon, especially for soft materials.

The psychophysics study is centered around an odd-one-out protocol, which has the advantage of avoiding any external reference to what would mean friction or texture for example. The comparisons are made only based on the texture being similar or not.

The results show a significant relationship between the distance between frictional maps and the success rate in discriminating two kinds of surface.

We thank Reviewer 3 for their notes and interesting discussion points on our manuscript. Below, we address the reviewer’s feedback and comments on related works.

Weaknesses:

Comment 1

The main weakness of the paper comes from the fact that the frictional maps and the extensive psychophysics study are not made at the same time, nor with the same finger. The frictional maps are produced with an artificial finger made out of PDMS which is a poor substitute for the complex tribological properties of skin.

A similar comment was made by Reviewers 1 and 2. We agree in part and have made changes throughout that our study is correlative, but presents an important step forward to these biomechanical measurements and corresponding decision making models.

We are not claiming that our PDMS fingers are superior to real fingers, but rather, we cannot establish standards in the field by using real human fingers that vary between subjects and researchers. We believe the mock finger we designed is a reasonable mimic of the human finger by matching surface energy, heterogeneous mechanical structure, and the ability to test multiple physiologically relevant pressures and sliding velocities.

We achieve a heterogeneous mechanical structure with the 3 primary components of stiffness of a human finger. The effective modulus of ~100 kPa, from soft tissue,[9,10] is obtained with a 30:1 ratio of PDMS to crosslinker. The PDMS also surrounds a rigid, acrylic bone comparable to the distal phalanx, which provides an additional layer of higher modulus.[8] Additionally, the 8-hour UV-Ozone treatment decreases the viscoelastic tack of the pristine PDMS by glassifying, or further crosslinking the surface of the finger,[12] therefore imparting greater stiffness at the surface similar to the contributions of the stratum corneum, along with a similar surface energy.[13] The finger is used at least a day after UV-Ozone treatment is completed in order for the surface to return to moderate hydrophilicity, similar to the outermost layer of human skin.[17] We also discuss the shape of the contact formed. To ensure that there is minimal contribution from the slanted position of the finger, an initial contact area of 1×1 cm is established before sliding and recording friction measurements. As the PDMS finger is a soft object, the portion in contact with a surface flattens and the contact area remains largely unchanged during sliding. We recognize that it is difficult to completely control the pressure distribution due to the planar interface, but this variation is also expected when humans freely explore a surface. Finally, we consider flat vs. fingerprinted fingers. Our previous work on the role of fingerprints on friction experienced by a PDMS mock finger showed enhanced signals with the incorporation of ridges on the finger and used a rate-andstate model of a heterogenous, elastic body to find corresponding trends.[11] The key conclusion was that a flat finger still preserved key dynamic features, and the presence of stronger or more vibrations could result in more similar forces for different surfaces depending on the sliding conditions. We note that we have subsequently used the controlled mechanical data collected with this flat mock finger in correlations with human psychophysics in previous work, where findings from our mechanical experiments were predictive of human performance.[4–7] Ultimately, we see from our prior work and here that, despite the drawbacks of our mock finger, it outperforms other standard characterization technique in providing information about the mesoscale that correlates to tactile perception. We have added these details to the manuscript.

We also note that an intermediate option, replicating real fingers, even in a mold, may also inadvertently limit trends from characterization to a specific finger. One of the main – and severe – limitations of using a human finger is that all fingers are different, meaning any study focusing on a particular user may not apply to others or be recreated easily by other researchers. We cannot set a standard for replication around a real human finger as that participant may no longer be available, or willing to travel the world as a “standard”. Furthermore, the method in which a single person changes their pressures and velocities as they touch a surface is highly variable. We also note that in the Summary Response, we noted that a study by Colgate et al. (IEEE ToH 2024) demonstrated that efference copies may be important, and thus constraining a human finger and replaying the forces recorded during free exploration will not lead to the participant identifying a surface with any consistency. Thus, it is important to allow humans to freely explore surfaces, but creates nearly limitless variability in friction forces.

This is also against the backdrop that we are seeking to provide a method to characterize surfaces. Indeed, the more features we replicate in the mock finger to a human finger, the more likely it is that the mechanical data will correlate to human performance. However, we have used this technique several times to achieve stronger correlations to human data than other available techniques. We believe the metric of success should be in comparison to the available characterization technique, rather than a 1:1 reconstruction of forces of an arbitrary human finger. Indeed, a 1:1 reconstruction of forces of an arbitrary human finger would be limited to the finger of a single individual, perhaps even to that individual on a given day.

See Reviewer1 weaknesses, comment 2 part 2 for changes to the manuscript

Comment 2

The evidence would have been much stronger if the measurement of the interaction was done during the psychophysical experiment. In addition, because of the protocol, the correlation is based on aggregates rather than on individual interactions.

We agree that this would have helped further establish our argument, but in the overall statement and in other reviewer responses, we describe the significant challenges to establishing this.

To fully implement this, a decision-making model is necessary because, as a counter example, a participant could have generated 10 swipes of SFW and 1 swipe of a Sp, but the Sp may have been the most important event for making a tactile decision. We also clarify that our goals are to provide a method to characterize samples to better design tactile interfaces in haptics or in psychophysical experiments.

As discussed in the summary, and expanded on here, in our view, to develop a decision-making model, the challenges are as follows:

(1) Which one, or combination of, of the multiple swipes that people make responsible for a tactile decision?

(2) Establish what is, or may be, tactile evidence.

(3) Establish tactile decision-making models are similar or different than existing decision-making models.

(4) Test the hypothesis, in these models, that friction instabilities are evidence, and not some other unknown metric.

(5) Design a task that does not require the use of subjective tactile descriptors, like “which one feels rougher”, which we see cause confusion in participants, which will likely require accounting for memory effects.

(6) Design samples that vary in the amount of evidence generated, but this evidence cannot be controlled directly. Rather, the samples indirectly vary evidence by how likely it is for a human to generate different types of friction instabilities during standard exploration.

We elaborate these points below:

To successfully perform this experiment, we note that freely exploring humans make multiple strokes on a surface. Therefore, we would need to construct a decision-making model. It has not yet been demonstrated whether tactile decision making follows visual decision making, but perhaps to start, we can assume it does. Then, in the design of our decision-making paradigm, we immediately run into the problem: What is tactile evidence?

From Fig. 3C, we already can see that identifying evidence is challenging. Prior to this manuscript, people may have chosen the average force, or the highest force. Or we may choose the average friction force. Then, after deciding on the evidence, we need to find a method to manipulate the evidence, i.e., create samples or a machine that causes high friction, etc. We show that during the course of human touch, due to the dynamic nature of friction, the average can change a large amount and sample design becomes a central barrier to experiments. Others may suggest to immobilize the finger and applying a known force, but given how much friction changes with human exploration, there is no known method to make a machine recreate temporally and spatially varying friction forces during sliding onto a stationary finger. Finally, perhaps most importantly, in addition to mechanical challenges, a study by Liu, Colgate et al. showed that even if they recorded the friction (2D) of a finger exploring a surface and then replicated the same friction forces onto a finger, the participant could not determine which surface the replayed friction force was supposed to represent.[1] This supports that the efference copy is important, that the forces in response to expected motion are important to determine friction. Finally, there is no known method to design instabilities a priori. They must be found through experiments, especially since if we were to introduce, say a bump or a trough, then we bring in confounding variables to how participants tell surfaces apart.

Furthermore, even if we had some consistent method to create tactile “evidence”, the paradigm also deserves some consideration. In our experience, the 3-AFC task we perform is important because the vocabulary for touch has not been established. That is, in 3-AFC, by asking to determine which one sample is unlike the others, we do not have to ask the participant questions like “which one is rougher” or “which one has less friction”. In contrast, 2-AFC, which is better for decision-making models because it does not include memory, requires the asking of a perceptual question like: “which one is rougher?”. In our ongoing work, taking two silane coatings, we found that participants could easily identify which surface is unlike the others above chance in a 3-AFC, but participants, even within their own trials, could not consistently identify one silane as perceptually “rougher” by 2-AFC. To us, this calls into question the validity of tactile descriptors, but is beyond the scope of the current manuscript.

This is not our only goal, but in the context of human exploration, in this manuscript here, we believed it was important to identify a mechanical parameter that was consistent with how humans explore surfaces, but was also a parameter that could characterize to some consistent property of a surface – irrespective of whether a human was touching it. We thought that designing human decision-making models and paradigms around the friction coefficient would not be successful.

Given the scope of these challenges, we do not think it would be possible to establish this conceptual sequence in a single manuscript.

Comment 3

The authors compensate with a third experiment where they used a 2AFC protocol and an online force measurement. But the results of this third study, fail to convince the relation.

With this experiment, our central goal was to demonstrate that the instabilities we have identified with the PDMS finger also occur with a human finger. Several instances of SS, Sp, and SFW were recorded with this setup as a participant touched surfaces in real time.

Comment 4

No map of the real finger interaction is shown, bringing doubt to the validity of the frictional map for something as variable as human fingers.

Real fingers change constantly during exploration, and friction is state-dependent, meaning that the friction will depend on how the person was moving the moment prior. Therefore, a map is only valid for a single human movement – even if participants all were instructed to take a single swipe and start from zero motion, humans are unable to maintain constant velocities and pressures. Clearly, this is not sustainable for any analysis, and these drawbacks apply to any measured parameter, whether instabilities suggested here, or friction coefficients used throughout. We believe the difficulty of this approach emphasizes why a standard map of characterization of a surface by a mock finger, even with its drawbacks, is a viable path forward.

Reviewer 3 (Recommendations for the authors):

Comment 1

It would be interesting to comment on a potential connection between the frictional instability maps and Schalamack waves.

Schallamach waves are a subset of slow frictional waves (SFW). Schallamach waves are very specifically defined in the field. They occur when pockets of air that form between a soft sliding object and rigid surface which then propagate rear-to-front (retrograde waves) relative to motion of the sliding motion and form buckles due to adhesive pinning. Wrinkles then form at the detached portion of the soft material, until the interface reattaches and the process repeats.[24] There is typically a high burden of proof to establish a Schallamach wave over a more general slow frictional wave. We note that it would be exceedingly difficult to design samples that can reliably create subsets of SFW, but we are aware that this may be an interesting question at a future point in our work.

Comment 2

The force sensors look very compliant, and given the dynamic nature of the signal, it is important to characterize the frequency response of the system to make sure that the fluctuations are not amplified.

Thank you for noticing. We mistyped the sensor spring constant as 13.9 N m^-1 instead of kN m^-1. However, below we show how the instabilities are derived from the mechanics at the interface due to the compliance of the finger. The “springs” of the force sensor and PDMS finger are connected in parallel. Since k_sensor = 13.9 kN m^-1, the spring constant of the system overall reflects the compliance of the finger, and highlights the oscillations arising solely from stick-slip. A sample calculation is shown below.

Author response image 1.

Fitting a line to the initial slope of the force trace for C6 gives the equation y = 25.679x – 0.2149. The slope here represents force data over time data, and is divided by the velocity (25 mm/s) to determine the spring constant of the system k_total == 1027.16 N/m. This value is lower than k_sensor = 13.9 kN/m, indicating that the “springs” representing the force sensor and PDMS finger are connected in parallel:

. The finger is the compliant component of the system, with k_finger = 1.11 kN/m, and of course, real human fingers are also compliant so this matches our goals with the design of the mock finger.

Our changes to the manuscript

(Page 4) (k = 13.9 kN m¹)

Comment 3

The authors should discuss about the stochastic nature of friction: - Wiertlewski, Hudin, Hayward, IEEE WHC 2011 Greenspon, McLellan, Lieber, Bensmaia, JRSI 2020.

We believe that, given the references, this comment on “stochastic” refers to the macroscopically-observable fluctuations (i.e., the mechanical “noise” which is not due to instrument noise) in friction arising from the discordant network of stick-slip phenomena occurring throughout the contact zone, and not the stochastic nature of nanoscale friction that occurs thermal fluctuations nor due to statistical distributions in bond breaking associated with soft contact.

We first note that our small-scale fluctuations do not arise from a periodic surface texture that dominates in the frequency regime. However, even on our comparatively smooth surfaces, we do expect fluctuations due to nanoscale variation in contact, generation of stick-slip across at microscale length scales that occur either concurrently or discordantly across the contact zone, and the nonlinear dependence of friction to nearly any variation in state and composition.[11]

Perhaps the most relevant to the manuscript is that a major advantage of analysis by friction is that it sidesteps these ever-present microscale fluctuations, leading to more clearly defined classifiers or categories during analysis. Wiertlewski et. al. showed repeated measurements in their systems ultimately gave rise to consistent frequencies[25] (we think their system was in a steady sliding regime and the patterning gave rise to underlying macroscopic waves). These consistent frequencies, at least in soft systems and absent obvious macroscopic patterned features, would be expected to arise from the instability categories and we see them throughout.

Comment 4

It is stated that "we observed a spurious, negative correlation between friction coefficient and accuracy".

What makes you qualify that correlation as spurious?

We mean this as in the statistical definition of “spurious”.

This correlation would indicate that by the metric of friction coefficient, more different surfaces are perceived more similarly. Thus, two very different surfaces, like Teflon and sandpaper, by friction coefficient would be expected to feel very similar. Two nearly identical surfaces would be expected to feel very different – but of course, humans cannot consistently distinguish two identical surfaces. This finding is counterintuitive and refutes that friction coefficient is a reliable classifier of surfaces by touch. We do not think it is productive to determine a mechanism for a spurious correlation, but perhaps one reason we were able to observe this is because our study, to the best of our knowledge, is unique for having samples that are controlled in their physical differences in roughness and surface features.

See response to Reviewer 1 weaknesses, comment 1 for changes to the manuscript

Comment 5

The authors should comment on the influence of friction on perceptual invariance. Despite inducing radially different frictional behavior for various conditions, these surfaces are stably perceived. Maybe this is a sign that humans extract a different metric?

We agree – we are excited that frictional instabilities may offer a more stable perceptual cue because they are not prone to fluctuations (as discussed in Comment 3) and instability formation, in many conditions, is invariant to applied pressures and velocities – thus forming large zones where a human may reasonable encounter a given instability.

Raw friction is highly prone to variation during human exploration (in alignment with Recommendations for the authors, Comment 3), but ongoing work seeks to explain tactile constancy, or the ability to identify objects despite these large changes in force. Very recently published work by Fehlberg et. al. identified the role of modulating finger speed and normal force in amplifying the differences in friction coefficient between materials in order to identify them,[26] and we postulate that their work may be streamlined and consistent with the idea of friction instabilities, though we have not had a chance to discuss this in-depth with the authors yet.

We think that the instability maps show a viable path forward to how surfaces are stably perceived, and instabilities themselves show a potential mechanism: mathematically, instabilities for given conditions can be invariant to velocity or mass, creating zones where a certain instability is encountered. This reduces the immense variability of friction to a smaller, more stable classification of surfaces (e.g., a 30% SS surface or a 60% SS surface). A given surface will typically produce the same instability at a specific condition (we found some boundaries of experimental parameters are very condition sensitive, but many conditions are not), whereas a single friction trace which is highly prone to variation is not a stable metric.

Added Reference

(53) M. Fehlberg, E. Monfort, S. Saikumar, K. Drewing and R. Bennewitz, IEEE Trans. Haptics, 2024, 17, 957–963.

References

(1) Liu, Z., Kim, J.-T., Rogers, J. A., Klatzky, R. L. & Colgate, J. E. Realism of Tactile Texture Playback: A Combination of Stretch and Vibration. IEEE Trans. Haptics 17, 441–450 (2024).

(2) Waters, I., Alazmani, A. & Culmer, P. Engineering Incipient Slip Into Surgical Graspers to Enhance Grasp Performance. IEEE Transactions on Medical Robotics and Bionics 2, 541–544 (2020).

(3) Gueorguiev, D., Bochereau, S., Mouraux, A., Hayward, V. & Thonnard, J.-L. Touch uses frictional cues to discriminate flat materials. Sci Rep 6, 25553 (2016).

(4) Carpenter, C. W. et al. Human ability to discriminate surface chemistry by touch. Mater. Horiz. 5, 70– 77 (2018).

(5) Nolin, A. et al. Predicting human touch sensitivity to single atom substitutions in surface monolayers for molecular control in tactile interfaces. Soft Matter 17, 5050–5060 (2021).

(6) Nolin, A. et al. Controlling fine touch sensations with polymer tacticity and crystallinity. Soft Matter 18, 3928–3940 (2022).

(7) Swain, Z. et al. Self-Assembled Thin Films as Alternative Surface Textures in Assistive Aids with Users Who are Blind. J. Mater. Chem. B (2024) doi:10.1039/D4TB01646G.

(8) Qian, K. et al. Mechanical properties vary for different regions of the finger extensor apparatus. J Biomech 47, 3094–3099 (2014).

(9) Abdouni, A. et al. Biophysical properties of the human finger for touch comprehension: influences of ageing and gender. Royal Society Open Science (2017) doi:10.1098/rsos.170321.

(10) Cornuault, P.-H., Carpentier, L., Bueno, M.-A., Cote, J.-M. & Monteil, G. Influence of physicochemical, mechanical and morphological fingerpad properties on the frictional distinction of sticky/slippery surfaces. Journal of The Royal Society Interface (2015) doi:10.1098/rsif.2015.0495.

(11) Dhong, C. et al. Role of fingerprint-inspired relief structures in elastomeric slabs for detecting frictional differences arising from surface monolayers. Soft Matter 14, 7483–7491 (2018).

(12) Fu, Y.-J. et al. Effect of UV-Ozone Treatment on Poly(dimethylsiloxane) Membranes: Surface Characterization and Gas Separation Performance. Langmuir 26, 4392–4399 (2010).

(13) Yuan, Y. & Verma, R. Measuring microelastic properties of stratum corneum. Colloids Surf B Biointerfaces 48, 6–12 (2006).

(14) Yu, G. et al. A wearable pressure sensor based on ultra-violet/ozone microstructured carbon nanotube/polydimethylsiloxane arrays for electronic skins. Nanotechnology 29, 115502 (2018).

(15) Zheng, L. et al. Dual-Stimulus Smart Actuator and Robot Hand Based on a Vapor-Responsive PDMS Film and Triboelectric Nanogenerator. ACS Appl. Mater. Interfaces 11, 42504–42511 (2019).

(16) Ma, K., Rivera, J., Hirasaki, G. J. & Biswal, S. L. Wettability control and patterning of PDMS using UV–ozone and water immersion. Journal of Colloid and Interface Science 363, 371–378 (2011).

(17) Mavon, A. et al. Sebum and stratum corneum lipids increase human skin surface free energy as determined from contact angle measurements: A study on two anatomical sites. Colloids and Surfaces B: Biointerfaces 8, 147–155 (1997).

(18) AliAbbasi, E. et al. Effect of Finger Moisture on Tactile Perception of Electroadhesion. IEEE Trans. Haptics 17, 841–849 (2024).

(19) Corniani, G. et al. Sub-surface deformation of individual fingerprint ridges during tactile interactions.

eLife 13, (2024).

(20) Israelachvili, J. N. Intermolecular and Surface Forces. (Academic Press, 2011).

(21) Das, S. et al. Stick–slip friction of gecko-mimetic flaps on smooth and rough surfaces. J R Soc Interface 12, 20141346 (2015).

(22) Persson, B. N. J., Albohr, O., Creton, C. & Peveri, V. Contact area between a viscoelastic solid and a hard, randomly rough, substrate. The Journal of Chemical Physics 120, 8779–8793 (2004).

(23) Skedung, L. et al. Feeling Small: Exploring the Tactile Perception Limits. Sci Rep 3, 2617 (2013).

(24) Viswanathan, K., Sundaram, N. K. & Chandrasekar, S. Stick-slip at soft adhesive interfaces mediated by slow frictional waves. Soft Matter 12, 5265–5275 (2016).

(25) Wiertlewski, M., Hudin, C. & Hayward, V. On the 1/f noise and non-integer harmonic decay of the interaction of a finger sliding on flat and sinusoidal surfaces. in 2011 IEEE World Haptics Conference 25–30 (2011). doi:10.1109/WHC.2011.5945456.

(26) Fehlberg, M., Monfort, E., Saikumar, S., Drewing, K. & Bennewitz, R. Perceptual Constancy in the Speed Dependence of Friction During Active Tactile Exploration. IEEE Transactions on Haptics 17, 957–963 (2024).

Author Response

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

eLife assessment

This study presents a valuable finding for the treatment of PCCs by sequencing 16 tumor specimens from five patients with pheochromocytomas by single-cell transcriptomics and proposing a new molecular classification criterion based on the sequencing results and characterization of tumor microenvironmental features. The evidence supporting the claims of the authors is solid, although the inclusion of more patient samples would strengthen the study's conclusions. The work will be of interest to clinicians or medical biologists working on rare pheochromocytomas (PCCs).

Firstly, we sincerely appreciate the positive feedback from the editor and extend our gratitude to the three reviewers for their meticulous review and valuable comments. Our detailed responses to each recommendation are outlined below.

Response to reviewers’ recommendations

Reviewer #1 (Recommendations for The Authors):

1) Transcriptomal clonal dynamics of different PCCs is well written. However for conclusion sample size needs to be more.

Acknowledging the rarity of PCCs with an incidence of approximately 0.2 to 0.6 cases per 100,000 person-years (Farrugia & Charalampopoulos, 2019; Neumann et al, 2019), our study recognizes the limitation in sample size, as discussed in the limitations section (Page 22). In response to this concern, we are committed to undertaking further research with an expanded sample size to bolster the robustness of our conclusions, seeking a more comprehensive understanding of tumor microenvironment characterization and molecular classification in PCCs. We appreciate the valuable guidance provided by the reviewer.

2) Clinical, biochemistry data of 5 cases can be analysed. Any findings in different categories as per postulated classification can be noted for further studies. Example: epinephrine levels

We have now included the clinical information of 5 PCC patients, encompassing signs and symptoms, the tumor size, and laboratory test results in the revised manuscript as Supplemental Table S3 (Page 11-12). Notably, our analysis revealed that the kinase-type PCC patient (P4) exhibited higher blood pressures and plasma levels of catecholamine metabolites (3-methoxytyramine and normetanephrine) compared to metabolism-type PCC patients (P1-P3, and P5). This observation aligns with the elevated expression of phenylethanolamine N-methyltransferase (PNMT), an enzyme involved in the biosynthesis of catecholamine and linked to hypertension, in P4, as identified in the scRNA-seq data (Figure 4B and 4D) (Kennedy et al, 1993; Konosu-Fukaya et al, 2018; Nguyen et al, 2015). As suggested, we plan to conduct further research to explore the correlation of our molecular classification with plasma levels of catecholamine metabolites, and the relevant points have been discussed in the revision (Page 20).

We would like to take this chance to again thank the reviewer for the careful review and very helpful guidance about how to improve our study.

References for Reviewer #1:

Farrugia FA, Charalampopoulos A (2019) Pheochromocytoma. Endocrine regulations 53: 191-212 Neumann HPH, Young WF, Jr., Eng C (2019) Pheochromocytoma and Paraganglioma. The New England journal of medicine 381: 552-565

Kennedy B, Elayan H, Ziegler MG (1993) Glucocorticoid hypertension and nonadrenal phenylethanolamine N-methyltransferase. Hypertension (Dallas, Tex : 1979) 21: 415419

Konosu-Fukaya S, Omata K, Tezuka Y, Ono Y, Aoyama Y, Satoh F, Fujishima F, Sasano H, Nakamura Y (2018) Catecholamine-Synthesizing Enzymes in Pheochromocytoma and Extraadrenal Paraganglioma. Endocrine pathology 29: 302309

Nguyen P, Khurana S, Peltsch H, Grandbois J, Eibl J, Crispo J, Ansell D, Tai TC (2015) Prenatal glucocorticoid exposure programs adrenal PNMT expression and adult hypertension. The Journal of endocrinology 227: 117-127

Reviewer #2 (Recommendations for The Authors):

1) Please revise all references to "malignant potential", "malignant behavior", etc. throughout the article, including the abstract and introduction, and replace them with the word "metastasis" as appropriate. Since all PCCs are malignant non-epithelial neuroendocrine neoplasms originating from the paraganglia, which are themselves malignant tumors, it is unacceptable to describe them as "malignant potential" or "malignant potential". Please review the 2022 WHO/IARC classification and description of pheochromocytoma/paraganglioma (reference: Mete O, Asa SL, Gill AJ, Kimura N, de Krijger RR, Tischler A. Overview of the 2022 WHO Classification of Paragangliomas and Pheochromocytomas. Endocr Pathol. 2022;33(1):90-114. doi:10.1007/s12022-022-09704-6).

As suggested, we have replaced all occurrences of “malignant potential” or “malignant behavior” with “metastasis” throughout the revised manuscript. We have also included a citation to the 2022 WHO/IARC classification for further clarity.

• Similarly, it is not advisable to use the PASS score to predict "malignant" PCC; this type of scoring system evaluates the "metastasis risk" or the "metastasis potential" of PCC.

We appreciate the reviewer for this insight and have revised our statements accordingly.

• Also, "MALIGNANT CHAFFIN CELLS" needs to be modified; in fact, it is the "tumor cell of PCC" that the authors are trying to express.

As suggested, we have amended the term “malignant chromaffin cells” to “PCC cells” in the revised manuscript (Page 9-10).

2) How does the PASS score specifically relate to intra-tumor heterogeneity as reflected by scRNA-seq? In fact, the PASS score evaluates the histological or pathological invasiveness of PCC, and different sections of the same tumor tissue may have different histological manifestations, which may affect the score; however, scRNA-seq analyzes the cellular composition of the tumor, which is not the same as the information reflected by the PASS score. Both represent different levels and dimensions of intra-tumor heterogeneity and should be analyzed together. Please specifically list, one by one, the proportion of each item score of the PASS system and cell type of scRNA-seq for each sample and the results of the comparisons with each other to better present the conclusions.

As suggested, we have included the proportion of each item score from the PASS system in the revised manuscript as Supplemental Table S2 (Page 8). Integrating this data with the cell type composition of each sample from Figure 2B, our analysis suggests that intra-tumor heterogeneity, as assessed by the PASS system, is more extensive compared to scRNA-seq. We concur with the reviewer’s judgement that scRNA-seq analysis and PASS score represent different levels and dimensions of intratumor heterogeneity, and we have adjusted our claim throughout the revised manuscript accordingly (Page 8, 9, and 19).

3) Where is the specific mutation site of the VHL gene in patient 5? Please advise.

The VHL gene mutation site, c.499C>T (missense mutation), in patient 5 was identified through whole exome sequencing (WES) analysis. We have now added the information to Supplemental Table S1 in the revised manuscript (Page 6).

4) Please revise Supplementary Figure 1, the scale should not appear in the picture of the staining result of P5.

As suggested, we have adjusted the position of the scale bar.

Author response image 1.

Hematoxylin-eosin staining and immunohistochemistry staining of CGA marker in formalin-fixed paraffin-embedded PCC tissue sections matched to scRNA-seq specimens. Scale bar, 100 μm.

5) What were the clinical presentation and biochemical findings in the five patients?

The information regarding tumor sizes, signs and symptoms, and plasma levels of catecholamine metabolites [3-methoxytyramine (3-MT), metanephrine (MN), and normetanephrine (NMN)] has been added to the revised manuscript as Supplemental Table S3 (Page 11-12).

• Were there any preoperative symptoms of hypertension?

With the exception of P2, preoperative symptoms of hypertension were observed in all PCC patients. The information has been added to the revised manuscript as Supplemental Table S3 (Page 11-12).

• What was the size and catecholamine secretion phenotype of each tumor? What was the relationship between these data and the scRNA-seq results?

The secretion phenotype showed that the kinase-type PCC patient (P4) exhibited higher plasma levels of catecholamine metabolites (3-methoxytyramine and normetanephrine) compared to metabolism-type PCC patients (P1-P3, and P5). This observation aligns with the elevated expression of phenylethanolamine Nmethyltransferase (PNMT), an enzyme involved in the biosynthesis of catecholamine and linked to hypertension, in P4, as identified in the scRNA-seq data (Figure 4B and 4D) (Kennedy et al, 1993; Konosu-Fukaya et al, 2018; Nguyen et al, 2015). Meanwhile, we have not observed the correlation between tumor sizes and molecular classification. We have now included tumor sizes and laboratory test results of 5 PCC patients in the revised manuscript as Supplemental Table S3 (Page 11-12), and the relevant points have been discussed in the revision (Page 20).

6) Please revise Figure 1A, the meaning shown in the figure appears to dissociate the tissues of the patient's normal adrenal glands, which can be misleading.

We appreciate the reviewer for raising this concern. The schematic in Figure 1A has been revised accordingly.

Author response image 2.

(1A) Schematic of the experimental pipeline. 11 tumor specimens and 5 adjacent normal adrenal medullary specimens were isolated from 5 PCC patients, dissociated into single-cell suspensions, and analyzed using 10x Genomics Chromium droplet scRNA-seq.

• Please revise the figure note for Figure 1B, where the symbol (B) appears twice.

As suggested, we have revised the figure legends for Figure 1B and 1C (Page 42).

7) Please indicate in the figure legends and text what exactly is meant by "adjacent specimens"? medulla? cortex? normal tissue? I believe the authors mean adjacent normal adrenal medullary tissue, please check the article.

As suggested, we have revised the term “adjacent specimens” to “adjacent normal adrenal medullary tissues” throughout the revised manuscript.

8) Please review the pathologic diagnostic criteria of this study in light of the 2022 WHO/IARC guidelines for pathologic diagnosis: "For the pathological diagnosis, the inclusion criteria were neuroendocrine neoplasm originating from the adrenal medulla and retroperitoneal origin, i.e. pheochromocytoma and paraganglioma, with consistent morphologic and immunohistochemical confirmation in relevant cases and positivity for chromogranin A and synaptophysin. The exclusion criteria were adrenocortical neoplasm and metastatic tumors." It is not rigorous enough to diagnose a tumor as PCC based on positive CgA immunohistochemical staining results alone.

We have revised the statements about pathologic diagnostic criteria in accordance with the suggestion and have cited the reference (Page 6).

We would like to express our gratitude to the reviewer for the thorough review and invaluable guidance provided to enhance the quality of our study.

References for Reviewer #2:

Kennedy B, Elayan H, Ziegler MG (1993) Glucocorticoid hypertension and nonadrenal phenylethanolamine N-methyltransferase. Hypertension (Dallas, Tex: 1979) 21: 415419

Konosu-Fukaya S, Omata K, Tezuka Y, Ono Y, Aoyama Y, Satoh F, Fujishima F, Sasano H, Nakamura Y (2018) Catecholamine-Synthesizing Enzymes in Pheochromocytoma and Extraadrenal Paraganglioma. Endocrine pathology 29: 302309

Nguyen P, Khurana S, Peltsch H, Grandbois J, Eibl J, Crispo J, Ansell D, Tai TC (2015) Prenatal glucocorticoid exposure programs adrenal PNMT expression and adult hypertension. The Journal of endocrinology 227: 117-127

Reviewer #3 (Recommendations For The Authors):

I have several concerns and suggestions, which if addressed would improve the manuscript.

1) The statements of “plasmas” in the manuscript and figures are confusing, which should be revised as “plasma cells”.

As suggested, we have revised the terminology from “plasmas” to “plasma cells” throughout the revised manuscript and figures.

2) The marker genes used for defining plasma cells (IGHG1 and IGLC2) showed low expressing percentage in Figure 1D. Please consider providing other genes as the marker of plasma cells.

As suggested, we performed additional analysis to pinpoint marker genes for accurate definition of plasma cells. Applying stricter statistical criteria (cut-off pvalue < 0.05, log2 fold change ≥ 1.5, and expressing percentage ≥ 0.6), we identified XBP1 (a transcription factor playing key roles in the final stages of plasma cell development) and IGKC (a type of light-chain immunoglobulins) (Todd et al, 2009; Poulsen et al, 2002) as top significant differentially expressed genes (DEGs) suitable for defining plasma cells. These data are now presented as Figure 1D in the revised manuscript (Page 7).

Author response image 3.

(1D) Dot plot of representative marker genes for each cell type. The color scale represents the average marker gene expression level; dot size represents the percentage of cells expressing a given marker gene.

3) The statement “Our clustering and cell type annotation analysis identified diverse adrenal cells, stromal cells, and immune cells within the PCC microenvironment” seems not be exhibited in Figure 1, so the clustering result of adrenal cells, stromal cells, and immune cells need to be added.

As suggested, we performed clustering analysis for adrenal cells, stromal cells, and immune cells (including lymphocytes and myeloid cells), and visualized by the Uniform Manifold Approximation and Projection (UMAP) plot. These data have been added to the revised manuscript as Supplemental Figure S3 (Page 8).

Author response image 4.

Integration Analysis across 5 PCC Patients Revealing the Cell Type Composition of the PCC Microenvironment. UMAP plot depicting the distribution of adrenal cells, stromal cells, and immune cells (including lymphocytes and myeloid cells) within the PCC microenvironment.

4) Given the classification of “metabolism-type PCCs” and “kinase-type PCCs” have not been presented in Figure 2D, the statement “Combined with our findings of a higher proportion of neutrophils and monocyts/macrophages in metabolism-type as compared with kinase-type” in Result 6 should be supported by using additional data.

As suggested, we performed additional analysis to evaluate the proportion of neutrophils and monocytes/macrophages in metabolism-type and kinasetype PCC patients. These data have been added to the revised manuscript as Supplemental Figure S4 (Page 14).

Author response image 5.

The frequency distribution of cell types within the microenvironment of metabolism-type and kinase-type PCC patients.

5) What makes the difference of scRNA-seq analysis and multispectral immunofluorescent staining in judging the immune escape of PCCs? Please provide an explanation.

We appreciate the reviewer's concern. scRNA-seq lacks spatial details, and multispectral immunofluorescent staining is constrained in the number of detected proteins. To address this, we employed both methods for analysis. scRNA-seq revealed limited communication between tumor and T cells, with lower HLA-I expression in kinase-type PCCs compared to metabolism-type PCCs. This was supported by multispectral staining using antibodies against CD4+ T cells, CD8+ T cells, M1 macrophages, or M2 macrophages markers, indicating sparse immune cell infiltration around tumor cells, mainly in the stroma (Figure 7A and 7B). This dual approach strengthens our understanding of immune escape in both PCC types. The explanation has been added to the revised manuscript (Page 21).

6) Figure 7G missed the scale bar for the staining results of marker proteins. Please add the scale bar into the figure.

As suggested, we have added to the scale bar accordingly.

7) In the method part of the manuscript, the authors should describe the minimum and maximum number used for quality control of the number of genes and the percentage of mitochondrial genes.

For quality control, we established a minimum threshold of no less than 200 genes and a maximum threshold of no more than 5000 genes. Additionally, the quality control process included a maximum threshold of 30% for mitochondrial genes. These specific criteria have been added to the methods section of the revised manuscript (Page 25-26).

We express our gratitude to the reviewer for their supportive recommendations and invaluable guidance on enhancing the rigor of our data.

References for Reviewer #3:

Todd DJ, McHeyzer-Williams LJ, Kowal C, Lee AH, Volpe BT, Diamond B, McHeyzer-Williams MG, Glimcher LH (2009) XBP1 governs late events in plasma cell differentiation and is not required for antigen-specific memory B cell development. The Journal of experimental medicine 206: 2151-2159

Poulsen TS, Silahtaroglu AN, Gisselø CG, Tommerup N, Johnsen HE (2002) Detection of illegitimate rearrangements within the immunoglobulin light chain loci in B cell malignancies using end sequenced probes. Leukemia 16: 2148-2155

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer 1 (Public review):

Summary:

This paper attempts to measure the complex changes of consciousness in the human brain as a whole. Inspired by the perturbational complexity index (PCI) from classic research, authors introduce simulation PCI (_s_PCI) of a time series of brain activity as a measure of consciousness. They first use large-scale brain network modeling to explore its relationship with the network coupling and input noise. Then the authors verify the measure with empirical data collected in previous research.

Strengths:

The conceptual idea of the work is novel. The authors measure the complexity of brain activity from the perspective of dynamical systems. They provide a comparison of the proposed measure with four other indexes. The text of this paper is very concise, supported by experimental data and theoretical model analysis.

We would like to thank the reviewer for evaluation of our work and the positive feedback. In what follows we would like to clarify the ambiguities in our initial submission, and the respective changes to the manuscript.

(1) Consciousness is a network phenomenon. The measure defined by the authors is to consider the maximal sPCI across the nodes stimulated. This measure is based on the time series of one node. The measure may be less effective in quantifying the ill relationship between nodes. This may contribute to the less predictive power of anesthesia (Figure 4b).

Thank you for this comment, consciousness is indeed a network phenomenon. sPCI is in fact measured across the whole network: to compute sPCI we apply PCI to simulated activity of the whole network. The perturbation is applied to individual nodes of network (different node for each trial) and each time, the response to the stimulus is measured through sPCI in the whole network. To make this explicit, the relevant section now reads:

“In line with the PCI experimental protocol, we sampled from multiple initial conditions and stimulated regions, presenting the maximum sPCI for each regime (i.e., each {G,σ}). For each simulation, we measured the complexity of the activity of the whole network over a 10-second period post-stimulus.”

(2) One of the focuses of the work is the use of a dynamic model of brain networks. The explanation of the model needs to be in more detail.

Thank you for your feedback. We expanded the method section.

(3) The equations should be checked. For example, there should be no max on the left side of the first equation on page 13.

We thank the reviewer for spotting this typo, and we removed the max on the left side of this equation, and also checked all the other equations for correctness. The equation now reads:

(4) The quality of the figures should be improved.

Thank you for your comment. We have made adjustments to several figures and we hope that they are clearer now...

(5) Figure 4 should be discussed and analyzed more in the text.

Thank you for pointing this out. We added the following paragraph discussing the figure (now number 5) in the results section:

“Classification results using a linear SVM are reported in Fig. 5. We report the crossplots of PCI and each of the resting-state metrics for all subjects and conditions in Fig. 5a. Each point corresponds to the calculation of the given metric over the whole recording normalized by its duration. We find that for fluidity (Fig. 5a, third panel), there is a perfect linear separation between Propofol and Xenon anesthesia on the left side and Wakefulness and Ketamine anesthesia on the right side. This corresponds to the classification accuracy result of 100% for the consciousness class in Fig. 5b, which is the same for PCI. As expected, PCI and fluidity behave poorly at classifying the presence of an anesthetic agent due to the confusion induced by Ketamine. However, the size of the functional repertoire performs an almost perfect classification for this grouping. Only one subject under Ketamine has a high functional repertoire (Fig. 5a, left panel), but all other subjects in the anesthesia condition have a size of functional repertoire roughly under 100. Classification accuracies for complexity and GAP at the group level are less performant but are shown for completeness.”

(6) The usage of the terms PCI and sPCI should be distinguished.

We would like to thank the reviewer for pointing out this ambiguity. The PCI metric had to be adapted for the synthetic data. We have now further emphasized this in the methods sections – “Perturbational Complexity”.

Reviewer 2 (Public review):

Summary:

Breyton and colleagues analysed the emergent dynamics from a neural mass model, characterised the resultant complexity of the dynamics, and then related these signatures of complexity to datasets in which individuals had been anaesthetised with different pharmacological agents. The results provide a coherent explanation for observations associated with different time series metrics, and further help to reinforce the importance of modelling when integrating across scientific studies.

Strengths:

(1) The modelling approach was clear, well-reasoned, and explicit, allowing for direct comparison to other work and potential elaboration in future studies through the augmentation with richer neurobiological detail.

(2) The results serve to provide a potential mechanistic basis for the observation that the Perturbational Complexity Index changes as a function of the consciousness state.

We would like to thank the reviewer for assessing our work, and the valuable feedback.

Weaknesses:

(3) Coactivation cascades were visually identified, rather than observed through an algorithmic lens. Given that there are numerous tools for quantifying the presence/absence of cascades from neuroimaging data, the authors may benefit from formalising this notion.

Thank you for bringing this to our attention. We added a quantification of the cascades in Fig 2 and 3. We computed the absolute value of the mean signal across sources (following z-scoring) to obtain a cascade profile and calculated the area under the curve as quantification of the overall presence of cascades. As it can be seen in the two figures, the presence of cascades is the highest around the working point. We have also added the precise definition to the methods section, which now reads:

“Coactivation Cascades

The profile of cascades over time was computed, first by z-scoring each source activity, and second by averaging the absolute value of the activity across all sources. The quantification of cascades was then obtained by calculating numerically the Area Under the Curve (AUC) of the profile of cascades.”

(4) It was difficult to tell, graphically, where the model’s operating regime lay. Visual clarity here will greatly benefit the reader.

Thank you for pointing out this ambiguity, we have marked the working point explicitly in the Figure 3.

Recommendations For The Authors

Reviewer 1 (Recommendations for the authors):

(1) In the method section, the technical details of the other four indexes should be elaborated.

Thank you for your recommendation, we agree that the description in the submitted manuscript was too brief. We expanded the method section about the functional repertoire and the bursting potential.

Reviewer 2 (Recommendations for the authors):

(1) The authors could more clearly label the ”working point” of their parameter space. Perhaps a label/arrow on Figure 2c that directs the readers’ eyes towards the location in state-space that you define as the working point?

Thank you for pointing out this ambiguity, we updated the figure 3 to mark the working point precisely.

(2) While ’fluidity’ is quite an evocative term and does a great job of suggesting to the uninitiated reader the character of the time series in question, I wonder whether a more descriptive term might be better suited for this variable, even if as an adjunct to the term, fluidity. In the past, we (and others) have used the term dynamic functional connectivity variability (Mu¨ller et al., 2020 NeuroImage) to refer to this feature, as it links the measure directly to the technique from which it was estimated.

Thank you for your feedback. You are correct, dynamic functional connectivity variability could have been a wording of choice for some of our results. However the term “fluidity” was chosen to convey a broader theoretical concept linked to dynamical systems but not exclusive to the brain. Here, dynamic functional connectivity variability is merely a measure of the fluidity of the system. We added the following in the method section describing the metrics:

“[...] Fluidity is related to previously defined metrics such as functional connectivity variability [10] that relied on a non-overlapping windowing procedure. We chose the term fluidity to convey a cocept linked to dynamical systems in general and states exploration. [...]”

(3) The term ”bursting potential” is also potentially problematic, as ”bursting” refers to a different concept at the cellular level (i.e., multiple action potentials in a short window of time) than it does in the context that the authors are presumably using it here (i.e., the capacity for the dynamics of the population to ”burst” into the fat-tail of their activity distribution). To avoid ambiguity, it could be worth considering altering this terminology, perhaps again by using a term that is descriptive of the technique used to estimate it, rather than the concept that it evokes.

Thank you for pointing out this ambiguity in the naming of the bursting potential. We have renamed it to “Global Activation Potential (GAP)” as we believe this term is a better description of the metric. We have switched to this term across the whole manuscript.

(4) There is a range of other modelling studies that have compared brain dynamics in the awake vs. anaesthetised patient. In my opinion, the reader would benefit from the ability to place this work into the broader context created by the literature, particularly as there are subtle (yet potentially important) differences in the models used in each case. Note - as this is a subjective opinion, I don’t view this as a crucial addition to the paper’s potential strength of evidence, though I do believe that it would have a positive effect on its potential impact.

We thank you for the suggestion. We have modified the before-to-last paragraph of the discussion to bring more context from the literature models of anethesia and wakefulness:

“Several studies have employed computational modeling approaches to investigate the differences in brain dynamics across states of consciousness. These studies present varying degrees of physiological detail and focus on complementary aspects of unconsciousness. They start from simple abstract models (Ising model) addressing for example the increased correlation between stuctural and functional connectivity in aneshesia [15], or oscillator-based models (Hopf model) capturing a brain state dependent response to simulated perturbation [4]. More neurobiologically realistic models (Dynamic Mean Field) have also been used to combine multimodal imaging data together with receptor density maps to address the macroscopic effects of general aneshesia and their relationship to spatially heterogeneous properties of the neuronal populations [8]. Similarly, using anatomically constrained parameters for brain regions has already been shown to increase the predictive value of brain network models [6, 18]. Furthermore, employing biophysically grounded mean-field and spiking neuron models (AdEx) allows addressing phenomena propagating in effect across multiple scales of description such as the molecular effects of anesthetics targeting specific receptor types [12]. Related work has shown that adaptation successfully reproduces dynamical regimes coherent with NREM and wakefulness [3] with corresponding realistic PCI values Goldman2021comprehensive. Here, we don’t address these biological questions but rather give a proof of concept that large-scale brain models can help understand the dynamics related to brain function. We used a model derived from QIF neurons Montbrio2015Macroscopic that lacks biological parameters such as ion concentration or synaptic adaptation. Nevertheless, we demonstrate that even the symmetry breaking caused by the connectome is sufficient for setting the global working point of the brain, which then links the brain’s capacity for generating complex behavior in the different paradigms, that is, rest and stimulation.”

(5) I saw the label ”digital brain twin” in the abstract but then did not find a location in the main text/methods wherein this aspect of the modelling was explained.

Thank you for pointing out this discrepancy, we have removed the term “digital brain twin” and replaced it by “whole-brain model” everywhere.

References

John M. Beggs and Dietmar Plenz. Neuronal Avalanches in Neocortical Circuits. The Journal of Neuroscience, 23(35):11167–11177, dec 3 2003.

A. G. Casali, O. Gosseries, M. Rosanova, M. Boly, S. Sarasso, K. R. Casali, S. Casarotto, M.-A. Bruno, S. Laureys, G. Tononi, and M. Massimini. A Theoretically Based Index of Consciousness Independent of Sensory Processing and Behavior. Science Translational Medicine, 5(198):198ra105–198ra105, aug 14 2013.

Anna Cattani, Andrea Galluzzi, Matteo Fecchio, Andrea Pigorini, Maurizio Mattia, and Marcello Massimini. Adaptation shapes local cortical reactivity: From bifurcation diagram and simulations to human physiological and pathological responses. eneuro, 10(7):ENEURO.0435– 22.2023, July 2023.

Gustavo Deco, Joana Cabral, Victor M. Saenger, Melanie Boly, Enzo Tagliazucchi, Helmut Laufs, Eus Van Someren, Beatrice Jobst, Angus Stevner, and Morten L. Kringelbach. Perturbation of whole-brain dynamics in silico reveals mechanistic differences between brain states. NeuroImage, 169:46–56, April 2018.

Rahul S. Desikan, Florent S´egonne, Bruce Fischl, Brian T. Quinn, Bradford C. Dickerson, Deborah Blacker, Randy L. Buckner, Anders M. Dale, R. Paul Maguire, Bradley T. Hyman, Marilyn S. Albert, and Ronald J. Killiany. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage, 31(3):968–980, 7 2006.

Xiaolu Kong, Ru Kong, Csaba Orban, Peng Wang, Shaoshi Zhang, Kevin Anderson, Avram Holmes, John D. Murray, Gustavo Deco, Martijn van den Heuvel, and B. T. Thomas Yeo. Sensory-motor cortices shape functional connectivity dynamics in the human brain. Nature Communications, 12(1), November 2021.

A. Lempel and J. Ziv. On the Complexity of Finite Sequences. IEEE Transactions on Information Theory, 22(1):75–81, 1 1976. event-title: IEEE Transactions on Information Theory.

Andrea I. Luppi, Pedro A. M. Mediano, Fernando E. Rosas, Judith Allanson, John D. Pickard, Guy B. Williams, Michael M. Craig, Paola Finoia, Alexander R. D. Peattie, Peter Coppola, Adrian M. Owen, Lorina Naci, David K. Menon, Daniel Bor, and Emmanuel A. Stamatakis. Whole-brain modelling identifies distinct but convergent paths to unconsciousness in anaesthesia and disorders of consciousness. Communications Biology, 5(1), April 2022.

Ernest Montbri´o, Diego Paz´o, and Alex Roxin. Macroscopic Description for Networks of Spiking Neurons. Physical Review X, 5(2):021028, jun 19 2015.

Eli J. Mu¨ller, Brandon Munn, Luke J. Hearne, Jared B. Smith, Ben Fulcher, Aurina Arnatkeviˇciu¯te˙, Daniel J. Lurie, Luca Cocchi, and James M. Shine. Core and matrix thalamic sub-populations relate to spatio-temporal cortical connectivity gradients. NeuroImage, 222:117224, November 2020.

J. Matias Palva, Alexander Zhigalov, Jonni Hirvonen, Onerva Korhonen, Klaus LinkenkaerHansen, and Satu Palva. Neuronal long-range temporal correlations and avalanche dynamics are correlated with behavioral scaling laws. Proceedings of the National Academy of Sciences, 110(9):3585–3590, feb 26 2013. publisher: Proceedings of the National Academy of Sciences.

Maria Sacha, Federico Tesler, Rodrigo Cofre, and Alain Destexhe. A computational approach to evaluate how molecular mechanisms impact large-scale brain activity. Nature Computational Science, 5(5):405–417, May 2025.

Simone Sarasso, Melanie Boly, Martino Napolitani, Olivia Gosseries, Vanessa Charland-Verville, Silvia Casarotto, Mario Rosanova, Adenauer Girardi Casali, Jean-Francois Brichant, Pierre Boveroux, Steffen Rex, Giulio Tononi, Steven Laureys, and Marcello Massimini. Consciousness and Complexity during Unresponsiveness Induced by Propofol, Xenon, and Ketamine. Current Biology, 25(23):3099–3105, 12 2015.

Pierpaolo Sorrentino, Rosaria Rucco, Fabio Baselice, Rosa De Micco, Alessandro Tessitore, Arjan Hillebrand, Laura Mandolesi, Michael Breakspear, Leonardo L. Gollo, and Giuseppe Sorrentino. Flexible brain dynamics underpins complex behaviours as observed in Parkinson’s disease. Scientific Reports, 11(1):4051, feb 18 2021. number: 1 publisher: Nature Publishing Group.

S. Stramaglia, M. Pellicoro, L. Angelini, E. Amico, H. Aerts, J. M. Cort´es, S. Laureys, and D. Marinazzo. Ising model with conserved magnetization on the human connectome: Implications on the relation structure-function in wakefulness and anesthesia. Chaos: An Interdisciplinary Journal of Nonlinear Science, 27(4), April 2017.

Enzo Tagliazucchi, Pablo Balenzuela, Daniel Fraiman, and Dante Chialvo. Criticality in LargeScale Brain fMRI Dynamics Unveiled by a Novel Point Process Analysis. Frontiers in Physiology, 3, 2012. [Online; accessed 2022-12-23].

David C. Van Essen, Stephen M. Smith, Deanna M. Barch, Timothy E.J. Behrens, Essa Yacoub, and Kamil Ugurbil. The WU-Minn Human Connectome Project: An Overview. NeuroImage, 80:62–79, oct 15 2013. PMID: 23684880 PMCID: PMC3724347.

Peng Wang, Ru Kong, Xiaolu Kong, Rapha¨el Li´egeois, Csaba Orban, Gustavo Deco, Martijn P. van den Heuvel, and B.T. Thomas Yeo. Inversion of a large-scale circuit model reveals a cortical hierarchy in the dynamic resting human brain. Science Advances, 5(1), January 2019.

Farnaz Zamani Esfahlani, Youngheun Jo, Joshua Faskowitz, Lisa Byrge, Daniel P. Kennedy, Olaf Sporns, and Richard F. Betzel. High-amplitude cofluctuations in cortical activity drive functional connectivity. Proceedings of the National Academy of Sciences of the United States of America, 117(45):28393–28401, November 2020.