eLife assessment

Supported by solid evidence, this work provides valuable insights into theanine metabolism and regulation at single-cell resolution. The study paves the way for addressing the multicellular compartmentation of secondary metabolites in various plant systems, making it a valuable resource for future research.

Reviewer #1 (Public Review):

Summary:

The study used root tips from semi-hydroponic tea seedlings. The strategy followed sequential steps to draw partial conclusions.

Initially, protoplasts obtained from root tips were processed for scRNA-seq using the 10x Genomics platform. The sequencing data underwent pre-filtering at both the cell and gene levels, leading to 10,435 cells. These cells were then classified into eight clusters using t-SNE algorithms. The present study scrutinised cell typification through protein sequence similarity analysis of homologs of cell type marker genes. The analysis was conducted to ensure accuracy using validated genes from previous scRNA-seq studies and the model plant Arabidopsis thaliana. The cluster cell annotation was confirmed using in situ RT-PCR analyses. This methodology provided a comprehensive insight into the cellular differentiation of the sample under study. The identified clusters, spanning 1 to 8, have been accurately classified as xylem, epidermal, stem cell niche, cortex/endodermal, root cap, cambium, phloem, and pericycle cells.

Then, the authors performed a pseudo-time analysis to validate the cell cluster annotation by examining the differentiation pathways of the root cells. Lastly, they created a differentiation heatmap from the xylem and epidermal cells and identified the biological functions associated with the highly expressed genes.

Upon thoroughly analysing the scRNA-seq data, the researchers delved into the cell heterogeneity of nitrate and ammonium uptake, transport, and nitrogen assimilation into amino acids. The scRNA-seq data was validated by in situ RT-PCR. It allows the localisation of glutamine and alanine biosynthetic enzymes along the cell clusters and confirms that both constituent the primary amino acid metabolism in the root. Such investigation was deemed necessary due to the paramount importance of these processes in theanine biosynthesis since this molecule is synthesised from glutamine and alanine-derived ethylamine.

Afterwards, the authors analysed the cell-specific expression patterns of the theanine biosynthesis genes, combining the same molecular tools. They concluded that theanine biosynthesis is more enriched in cluster 8 "pericycle cells" than glutamine biosynthesis (Lines 271-272). However, the statement made in Line 250 states that the highest expression levels of genes responsible for glutamine biosynthesis were observed in Clusters 1, 3, 4, 6, and 8, leading to an unclear conclusion.

The regulation of theanine biosynthesis by the MYB transcription factor family is well-established. In particular, CsMYB6, a transcription factor expressed specifically in roots, has been found to promote theanine biosynthesis by binding to the promoter of the TSI gene responsible for theanine synthesis. However, their findings indicate that CsMYB6 expression is present in Cluster 3 (SCN), Cluster 6 (cambium cells), and Cluster 1 (xylem cells) but not in Cluster 8 (pericycle cells), which is known for its high expression of CsTSI. Similarly, their scRNA-seq data indicated that CsMYB40 and CsHHO3, which activate and repress CsAlaDC expression, respectively, did not show high expression in Cluster 1 (the cell cluster with high CsAlaDC expression). Based on these findings, the authors hypothesised that transcription factors and target genes are not necessarily always highly expressed in the same cells. Nonetheless, additional evidence is essential to substantiate this presumption.

Lastly, the authors have discovered a novel transcription factor belonging to the Lateral Organ Boundaries Domain (LBD) family known as CsLBD37 that can co-regulate the synthesis of theanine and the development of lateral roots. The authors observed that CsLBD37 is located within the nucleus and can repress the CsAlaDC promoter's activity. To investigate this mechanism further, the authors conducted experiments to determine whether CsLBD37 can inhibit CsAlaDC expression in vivo. They achieved this by creating transiently CsLBD37-silenced or over-expression tea seedlings through antisense oligonucleotide interference and generation of transgenic hairy roots. Based on their findings, the authors hypothesise that CsLBD37 regulates CsAlaDC expression to modulate the synthesis of ethylamine and theanine.

Additionally, the available literature suggests that the transcription factors belonging to the Lateral Organ Boundaries Domain (LBD) family play a crucial role in regulating the development of lateral roots and secondary root growth. Considering this, they confirmed that pericycle cells exhibit a higher expression of CsLBD37. A recent experiment revealed that overexpression of CsLBD37 in transgenic Arabidopsis thaliana plants led to fewer lateral roots than the wild type. From this observation, the researchers concluded that CsLBD37 regulates lateral root development in tea plants. I respectfully submit that the current conclusion may require additional research before it can be considered definitive.

Further efforts should be made to investigate the signalling mechanisms that govern CsLBD37 expression to arrive at a more comprehensive understanding of this process. In the context of Arabidopsis lateral root founder cells, the establishment of asymmetry is regulated by LBD16/ASL18 and other related LBD/ASL proteins, as well as the AUXIN RESPONSE FACTORs (ARF7 and ARF19). This is achieved by activating plant-specific transcriptional regulators such as LBD16/ASL18 (Go et al., 2012, https://doi.org/10.1242/dev.071928). On the other hand, other downstream homologues of LBD genes regulated by cytokinin signalling play a role in secondary root growth (Ye et al., 2021, https://doi.org/10.1016/j.cub.2021.05.036). It is imperative to shed light on the hormonal regulation of CsLBD37 expression in order to gain a comprehensive understanding of its involvement in the morphogenic process.

Strength:

The manuscript showcases significant dedication and hard work, resulting in valuable insights that serve as a fundamental basis for generating knowledge. The authors skillfully integrated various tools available for this type of study and meticulously presented and illustrated every step involved in the survey. The overall quality of the work is exceptional, and it would be a valuable addition to any academic or professional setting.

Weaknesses:

In its current form, the article presents certain weaknesses that need to be addressed to improve its overall quality. Specifically, the authors' conclusions appear to have been drawn in haste without sufficient experimental data and a comprehensive discussion of the entire plant. It is strongly advised that the authors devote additional effort to resolving the abovementioned issues to bolster the article's credibility and dependability. This will ensure that the article is of the highest quality, providing readers with reliable and trustworthy information.

Reviewer #2 (Public Review):

Summary:

In their manuscript, Lin et al. present a comprehensive single-cell analysis of tea plant roots. They measured the transcriptomes of 10,435 cells from tea plant root tips, leading to the identification and annotation of 8 distinct cell clusters using marker genes. Through this dataset, they delved into the cell-type-specific expression profiles of genes crucial for the biosynthesis, transport, and storage of theanine, revealing potential multicellular compartmentalization in theanine biosynthesis pathways. Furthermore, their findings highlight CsLBD37 as a novel transcription factor with dual regulatory roles in both theanine biosynthesis and lateral root development.

Strengths:

This manuscript provides the first single-cell dataset analysis of roots of the tea plants. It also enables detailed analysis of the specific expression patterns of the gene involved in theanine biosynthesis. Some of these gene expression patterns in roots were further validated through in-situ RT-PCR. Additionally, a novel TF gene CsLBD37's role in regulating theanine biosynthesis was identified through their analysis.

Weaknesses:

Several issues need to be addressed:

(1) The annotation of single-cell clusters (1-8) in Figure 2 could benefit from further improvement. Currently, the authors utilize several key genes, such as CsAAP1, CsLHW, CsWAT1, CsIRX9, CsWOX5, CsGL3, and CsSCR, to annotate cell types. However, it is notable that some of these genes are expressed in only a limited number of cells within their respective clusters, such as CsAAP1, CsLHW, CsGL3, CsIRX9, and CsWOX5. It would be advisable to utilize other marker genes expressed in a higher percentage of cells or employ a combination of multiple marker genes for more accurate annotation.

(2) Figure 3 could enhance clarity by displaying the trajectory of cell differentiation atop the UMAP, similar to the examples demonstrated by Monocle 3.

(3) The identification of CsLBD37 primarily relies on bulk RNA-seq data. The manuscript could benefit from elaborating on the role of the single-cell dataset in this context.

(4) The manuscript's conclusions predominantly rely on the expression patterns of key genes. This reliance might stem from the inherent challenges of tea research, which often faces limitations in exploring molecular mechanisms due to the lack of suitable genetic and molecular methods. The authors may consider discussing this point further in the discussion section.

Reviewer #3 (Public Review):

Summary:

Lin et al., performed a scRNA-seq-based study of tea roots, as an example, to elucidate the biosynthesis and regulatory processes for theanine, a root-specific secondary metabolite, and established the first map of tea roots comprised of 8 cell clusters. Their findings contribute to deepening our understanding of the regulation of the synthesis of important flavor substances in tea plant roots. They have presented some innovative ideas.

It is notable that the authors - based on single-cell analysis results - proposed that TFs and target genes are not necessarily always highly expressed in the same cells. Many of the important TFs they previously identified, along with their target genes (CsTSI or CsAlaDC), were not found in the same cell cluster. Therefore, they proposed a model in which the theanine biosynthesis pathway occurs via multicellular compartmentation and does not require high co-expression levels of transcription factors and their target genes within the same cell cluster. Since it is not known whether the theanine content is absolutely high in the cell cluster 1 containing a high CsAlaDC expression level (due to the lack of cell cluster theanine content determination, which may be a current technical challenge), it is difficult to determine whether this non-coexpressing cell cluster 1 is a precise regulatory mechanism for inhibiting theanine content in plants. In fact, there are actually a small number of cells where TFs and CsAlaDC are simultaneously highly expressed, but the quantity is insufficient to form a separate cluster. However, these few cells may be sufficient to meet the current demands for theanine synthesis. This possibility may better align with some previous experiments and validation results in this study. Moreover, I feel that under normal conditions, plants may not mobilize a large number of cells to synthesize a particular substance. Perhaps, cell cluster 1 is actually a type of cell that inhibits the synthesis of theanine, aiming to prevent excessive theanine production? I do not oppose the model proposed by the author, but I feel there is a possibility as I mentioned. If it seems reasonable, the author may consider adding it to an appropriate position in the discussion.

eLife assessment

The authors present an important resource to quantify mitochondrial function across many organs in mice. The convincing conclusions are supported by the identification of processes that specifically differ between young and old, or between male and female mice. All reviewers point to the merit of this study in providing a comprehensive resource to contextualize mitochondrial functions across the body. Some further suggestions are made to clarify conclusions in terms of data normalization, interpretations of comparative analyses between organs.

Reviewer #1 (Public Review):

In this study, Sarver and colleagues carried out an exhaustive analysis of the functioning of various components (Complex I/II/IV) of the mitochondrial electron transport chain (ETC) using a real-time cell metabolic analysis technique (commonly referred as Seahorse oxygen consumption rate (OCR) assay). The authors aimed to generate an atlas of ETC function in about 3 dozen tissue types isolated from all major mammalian organ systems. They used a recently published improvised method by which ETC function can be quantified in freshly frozen tissues. This method enabled them to collect data from almost all organ systems from the same mouse and use many biological replicates (10 mice/experiment) required for an unbiased and statistically robust analysis. Moreover, they studied the influence of sex (male and female) and aging (young adult and old age) on ETC function in these organ systems. The main findings of this study are (1) cells in the heart and kidneys have very active ETC complexes compared to other organ systems, (2) the sex of the mice has little influence on the ETC function, and (3) aging undermined the mitochondrial function in most tissue, but surprisingly in some tissue aging promoted the activity of ETC complexes (e.g., Quadriceps, plantaris muscle, and Diaphragm). Although this study provides a comprehensive outlook on the ETC function in various tissues, the main caveat is that it's too technical and descriptive. The authors didn't invest much effort in putting their findings in the context of the biological function of the tissue analyzed, i.e., some tissues might be more glycolytic than others and have low ETC activity. Also, it is unclear what slight changes in the activity of one or the other ETC complex mean in terms of mitochondrial ATP production. Likely, these small changes reported do not affect the mitochondrial respiration. With such a detailed dataset, the study falls short of deriving more functionally relevant conclusions about the heterogeneity of mitochondrial function in various tissues. In the current format, the readers get lost in the large amount of data presented in a technical manner. Also, it is highly recommended that all the raw data and the values be made available as an Excel sheet (or other user-friendly formats) as a resource to the community.

Major concerns

(1) In this study, the authors used the method developed by Acin-Perez and colleagues (EMBO J, 2020) to analyze ETC complex activities in mitochondria derived from the snap-frozen tissue samples. However, the preservation of cellular/mitochondrial integrity in different types of tissues after being snap-frozen was not validated. Additionally, the conservation of mitochondrial respiration in snap-frozen tissues might differ, especially in those derived from old mice. For example, quadriceps (young male/female), plantaris (young male/female), intestinal segments (duodenum), and pancreas preparations show almost no activity (nearly flat OCR in Seahorse assays). For such a comprehensive study, the author must at least validate those tissues where the OCR plots looked suboptimal with the mitochondrial preparations derived from the fresh tissue. Since aging has been identified as the most important effector in this study, it is essential to validate how aging affects respiration in various fresh frozen tissues. Such analysis will ensure that the results presented are not due to the differential preservation of the mitochondrial respiration in the frozen tissue. In addition, such validations will further strengthen the conclusions and promote the broad usability of this "new" method.

(2) In this study, the authors sampled the maximal activity of ETC complex I, II, and IV, but throughout the manuscript, they discussed the data in the context of mitochondrial function. However, it is unclear how the changes in CI, CII, and CIV activity affect overall mitochondrial function (if at all) and how small changes seen in the maximal activity of one or more complexes affect the efficiency and efficacy of ATP production (OxPhos). The authors report huge variability between the activity of different complexes - in some tissues all three complexes (CI, CII, and CIV) and often in others, just one complex was affected. For example, as presented in Figure 4, there is no difference in CI activity in the hippocampus and cerebellum, but there is a slight change in CII and CIV activity. In contrast, in heart atria, there is a change in the activity of CI but not in CII and CIV. However, the authors still suggest that there is a significant difference in mitochondrial activity (e.g., "Old males showed a striking increase in mitochondrial activity via CI in the heart atria....reduced mitochondrial respiration in the brain cortex..." - Lines 5-7, Page 9). Until and unless a clear justification is provided, the authors should not make these broad claims on mitochondrial respiration based on small changes in the activity of one or more complexes (CI/CII/CIV). With such a data-heavy and descriptive study, it is confusing to track what is relevant and what is not for the functioning of mitochondria.

(3) What do differences in the ETC complex CI, CII, and CIV activity in the same tissue mean? What role does the differential activity of these complexes (CI, CII, and CIV) play in mitochondrial function? What do changes in Oxphos mean for different tissues? Does that mean the tissue (cells involved) shift more towards glycolysis to derive their energy? In the best world, a few experiments related to the glycolytic state of the cells would have been ideal to solidify their finding further. The authors could have easily used ECAR measurements for some tissues to support their key conclusions.

(4) The authors further analyzed parameters that significantly changed across their study (Figure 7, 98 data points analyzed). The main caveat of such analysis is that some tissue types would be represented three or even more times (due to changes in the activity of all three complexes - CI, CII, and CIV, and across different ages and sexes), and some just once. Such a method of analysis will skew the interpretation towards a few over-represented organ/tissue systems. Perhaps the authors should separately analyze tissue where all three complexes are affected from those with just one affected complex.

(5) The current protocol does not provide cell-type-specific resolution and will be unable to identify the cellular source of mitochondrial respiration. This becomes important, especially for those organ systems with tremendous cellular heterogeneity, such as the brain. The authors should discuss whether the observed changes result from an altered mitochondria respiratory capacity or if changes in proportions of cell types in the different conditions studied (young vs. aged) might also contribute to differential mitochondrial respiration.

(6) Another critical concern of this study is that the same datasets were repeatedly analyzed and reanalyzed throughout the study with almost the same conclusion - namely, aging affects mitochondrial function, and sex-specific differences are limited to very few organs. Although this study has considerable potential, the authors missed the chance to add new insights into the distinct characteristics of mitochondrial activity in various tissue and organ systems. The author should invest significant efforts in putting their data in the context of mitochondrial function.

Reviewer #2 (Public Review):

Summary:

The authors utilize a new technique to measure mitochondrial respiration from frozen tissue extracts, which goes around the historical problem of purifying mitochondria prior to analysis, a process that requires a fair amount of time and cannot be easily scaled up.

Strengths:

A comprehensive analysis of mitochondrial respiration across tissues, sexes, and two different ages provides foundational knowledge needed in the field.

Weaknesses:

While many of the findings are mostly descriptive, this paper provides a large amount of data for the community and can be used as a reference for further studies. As the authors suggest, this is a new atlas of mitochondrial function in mouse. The inclusion of a middle aged time point and a slightly older young point (3-6 months) would be beneficial to the study.

Reviewer #3 (Public Review):

The aim of the study was to map, a) whether different tissues exhibit different metabolic profiles (this is known already), what differences are found between female and male mice and how the profiles changes with age. In particular, the study recorded the activity of respirasomes, i.e. the concerted activity of mitochondrial respiratory complex chains consisting of CI+CIII2+CIV, CII+CIII2+CIV or CIV alone.

The strength is certainly the atlas of oxidative metabolism in the whole mouse body, the inclusion of the two different sexes and the comparison between young and old mice. The measurement was performed on frozen tissue, which is possible as already shown (Acin-Perez et al, EMBO J, 2020).

Weakness:

The assay reveals the maximum capacity of enzyme activity, which is an artificial situation and may differ from in vivo respiration, as the authors themselves discuss. The material used was a very crude preparation of cells containing mitochondria and other cytosolic compounds and organelles. Thus, the conditions are not well defined and the respiratory chain activity was certainly uncoupled from ATP synthesis. Preparation of more pure mitochondria and testing for coupling would allow evaluation of additional parameters: P/O ratios, feedback mechanism, basal respiration, and ATP-coupled respiration, which reflect in vivo conditions much better. The discussion is rather descriptive and cautious and could lead to some speculations about what could cause the differences in respiration and also what consequences these could have, or what certain changes imply.

Nevertheless, this study is an important step towards this kind of analysis.

eLife assessment

This valuable study partially succeeds in providing evidence to support the therapeutic potential of the plant-derived compound eugenol for ameliorating symptoms associated with Type 1 Diabetes, identifying Nuclear factor E2 - related factor (Nrf2) as a mediator of the effects induced by eugenol. Although the study provides some interesting data, the evidence for the proposed mechanism is currently incomplete.

Reviewer #1 (Public Review):

Summary

Type 1 diabetes mellitus (T1DM) progression is accelerated by oxidative stress and apoptosis. Eugenol (EUG) is a natural compound previously documented as anti-inflammatory, anti-oxidative, and anti-apoptotic. In this manuscript by Jiang et al., the authors study the effects of EUG on T1DM in MIN6 insulinoma cells and a mouse model of chemically induced T1DM. The authors show that EUG increases nuclear factor E2-related factor 2 (Nrf2) levels. This results in a reduction of pancreatic beta-cell damage, apoptosis, oxidative stress markers, and a recovery of insulin secretion. The authors highlight these effects as indicative of the therapeutic potential of EUG in managing T1DM.

Strengths

Relevant, timely, and addresses an interesting question in the field. The authors consistently observe enhanced beta cell functionality following EUG treatment, which makes the compound a promising candidate for T1DM therapy.

Weaknesses

The in vivo experiments have too few biological replicates. With an n=3 (as all figure legends indicate) in complex mouse studies such as these, drawing robust conclusions becomes challenging. It is important to reproduce these results in a larger cohort, to validate the conclusions of the authors. Another big concern is the lack of quantifications and statistical analysis throughout the manuscript. Although the authors claim statistical significance in various experiments, the limited information provided makes it difficult to verify. The authors use vague and minimal descriptions of their experiments, which further reduces the reader's comprehension and the reproducibility of the experiments. Finally, the use of Min6 cells as a model for pancreatic beta cells is a strong limitation of this study. Future studies should seek to reproduce these findings in a more translational model and use more relevant in vitro cell systems (eg. Islets).

Reviewer #2 (Public Review):

Summary:

In this manuscript, the authors consider the effects of eugenol (EUG), a plant-produced substance known to reduce oxidative stress in various cellular contexts via Nrf2, in alleviating the effects of streptozotocin (STZ), a known rodent beta cell toxin. They claim that EUG treatment would be useful for T1D therapy.

Strengths:

The experiments shown are sufficiently clear and rather convincing in documenting that eugenol can revert the effects of streptozotocin on animal physiology as well as beta cell oxidative stress and cell death via activation of Nrf2.

Weaknesses:

In my view, there are major concerns with the basic premises of the manuscript.

(1) While oxidative stress may be implicated in T1D they are neither the primary nor the main reason for autoimmune beta cell destruction. In T1DM, ER stress rather than oxidative stress is the main intracellular mediator of cell death. Thus, the abstract statement that 'oxidative stress plays a major role in T1D' is an exaggeration.

(2) Streptozotocin induces beta cell death through mechanisms that only partially overlap with autoimmune beta cell destruction. The main players ie beta cell / immune system crosstalk and T-cell mediated cell death are not present in the STZ model.

In short, because the interplay between the immune system and beta cell-intrinsic factors that trigger and accelerate the disease is completely missing, STZ treatment cannot be used as a T1DM model when beta cell demise mechanisms are concerned. The statement that STZ-treated mice are, in this context, a T1DM model, is misleading.

There are inconsistencies in the manuscript. Mechanistically, the manuscript remains at a rather superficial level demonstrating that the eugenol effects are mediated by Nrf2 upregulation and a downregulation of its partner inhibitor protein Keap1. How is eugenol penetrating the cell, is there a receptor that could be potentially targeted? Are there intermediary proteins that convey the effect to the Nrf2/Keap1 complex or is eugenol directly disrupting their interaction? What are direct downstream Nrf2 effectors? Besides, streptozotocin is also a powerful DNA alkylating agent. Are these effects mitigated by EUG?

Reviewer #3 (Public Review):

Summary:

This study by Jiang et al. aims to establish the streptozotocin (STZ)-induced type 1 diabetes mellitus (T1DM) mouse model in vivo and the STZ-induced pancreatic β cell MIN6 cell model in vitro to explore the protective effects of Eugenol (EUG) on T1DM. The authors tried to elucidate the potential mechanism by which EUG inhibits the NRF2-mediated anti-oxidative stress pathway. Overall, this study is well executed with solid data, offering an intriguing report from animal studies for a potential new treatment strategy for T1DM.

Strengths:

The in vivo efficacy study is comprehensive and solid. Given that STZ-induced T1DM is a devastating and harsh model, the in vivo efficacy of this compound is really impressive.

Weaknesses:

The Mechanism is linked with the anti-oxidant property of the compound, which is common for many natural compounds, such as flavonoids and polyphenol. However, rarely, this kind of compound has been successfully developed into therapeutics in clinical usage. Indeed, if that is the case, Vitamin C or Vitamin E could be used here as the positive control.

eLife assessment

This valuable simulation study proposes a new coarse-grained model to explain the effects of CpG methylation on nucleosome wrapping energy and nucleosome positioning. The evidence to support the claims in the paper looks solid, although the novelty of the findings should be discussed in connection with the previous works. This work will be of interest to the researchers working on gene regulation and mechanisms of DNA methylation.

Reviewer #1 (Public Review):

Summary:

In this manuscript, the authors used a coarse-grained DNA model (cgNA+) to explore how DNA sequences and CpG methylation/hydroxymethylation influence nucleosome wrapping energy and the probability density of optimal nucleosomal configuration. Their findings indicate that both methylated and hydroxymethylated cytosines lead to increased nucleosome wrapping energy. Additionally, the study demonstrates that methylation of CpG islands increases the probability of nucleosome formation.

Strengths:

The major strength of this method is that the model explicitly includes elastic constraints on the positions of phosphate groups facing a histone octamer, as DNA-histone binding site constraints. The authors claim that their model enhances the accuracy and computational efficiency and allows comprehensive calculations of DNA mechanical properties and deformation energies.

Weaknesses:

A significant limitation of this study is that the parameter sets for the methylated and hydroxymethylated CpG steps in the cgNA+ model are derived from all-atom molecular dynamics (MD) simulations that suggest that both methylated and hydroxymethylated cytosines increase DNA stiffness and nucleosome wrapping energy (Pérez A, et al. Biophys J. 2012; Battistini, et al. PLOS Comput Biol. 2021). It could predispose the coarse-grained model to replicate these findings. Notably, conflicting results from other all-atom MD simulations, such as those by Ngo T in Nat. Commun. 2016, shows that hydroxymethylated cytosines increase DNA flexibility, contrary to methylated cytosines. If the cgNA+ model was trained on these later parameters or other all-atom force fields, different conclusions might be obtained regarding the effects of methylated and hydroxymethylation on nucleosome formation.

Despite the training parameters of the cgNA+ model, the results presented in the manuscript indicate that methylated cytosines increase both DNA stiffness and nucleosome wrapping energy. However, when comparing nucleosome occupancy scores with predicted nucleosome wrapping energies and optimal configurations, the authors find that methylated CGIs exhibit higher nucleosome occupancies than unmethylated ones, which seems to contradict their findings from the same paper which showed that increased stiffness should reduce nucleosome formation affinity. In the manuscript, the authors also admit that these conclusions "apparently runs counter to the (perhaps naive) intuition that high nucleosome forming affinity should arise for fragments with low wrapping energy". Previous all-atom MD simulations (Pérez A, et al. Biophys J. 2012; Battistini, et al. PLOS Comput Biol. 202; Ngo T, et al. Nat. Commun. 20161) show that the stiffer DNA upon CpG methylation reduces the affinity of DNA to assemble into nucleosomes or destabilizes nucleosomes. Given these findings, the authors need to address and reconcile these seemingly contradictory results, as the influence of epigenetic modifications on DNA mechanical properties and nucleosome formation are critical aspects of their study.<br /> Understanding the influence of sequence-dependent and epigenetic modifications of DNA on mechanical properties and nucleosome formation is crucial for comprehending various cellular processes. The authors' study, focusing on these aspects, will definitely garner interest from the DNA methylation research community.

Reviewer #2 (Public Review):

Summary:

This study uses a coarse-grained model for double-stranded DNA, cgNA+, to assess nucleosome sequence affinity. cgNA+ coarse-grains DNA on the level of bases and accounts also explicitly for the positions of the backbone phosphates. It has been proven to reproduce all-atom MD data very accurately. It is also ideally suited to be incorporated into a nucleosome model because it is known that DNA is bound to the protein core of the nucleosome via the phosphates.

It is still unclear whether this harmonic model parametrized for unbound DNA is accurate in describing DNA inside the nucleosome. Previous models by other authors, using more coarse-grained models of DNA, have been rather successful in predicting base pair sequence-dependent nucleosome behavior. This is at least the case as far as DNA shape is concerned whereas assessing the role of DNA bendability (something this paper focuses on) has been consistently challenging in all nucleosome models, to my knowledge.

It is thus of major interest whether this more sophisticated model is also more successful in handling this issue. As far as I can tell the work is technically sound and properly accounts for not only the energy required in wrapping DNA but also entropic effects, namely the change in entropy that DNA experiences when going from the free state to the bound state. The authors make an approximation here which seems to me to be a reasonable first step.

Of interest is also that the authors have the parameters at hand to study the effect of methylation of CpG-steps. This is especially interesting as it allows us to study a scenario where changes in the physical properties of base pair steps via methylation might influence nucleosome positioning and stability in a cell-type-specific way.

Overall, this is an important contribution to the question of how the sequence affects nucleosome positioning and affinity. The findings suggest that cgNA+ has something new to offer. But the problem is complex, also on the experimental side, so many questions remain open.

Strengths:

The authors use their state-of-the-art coarse-grained DNA model which seems ideally suited to be applied to nucleosomes as it accounts explicitly for the backbone phosphates.

Weaknesses:

(1) According to the abstract the authors consider two "scalar measures of the sequence-dependent propensity of DNA to wrap into nucleosomes". One is the bending energy and the other, is the free energy. Specifically in the latter, the authors take the difference between the free energies of the wrapped and the free DNA. Whereas the entropy of the latter can be calculated exactly, they assume that the bound DNA always has the same entropy (independent of sequence) in its more confined state. The problem is the way in which this is written (e.g. below Eq. 6) which is hard to understand. The authors should mention that the negative of Eq. 6 is what physicists call free energy, namely especially the free energy difference between bound and free DNA.

(2) In Eq. 5 the authors introduce penalty coefficients c_i. They write that values are "set by numerical experiment to keep distances ... within the ranges observed in the PDB structure, while avoiding sterical clashes in DNA." This is rather vague, especially since it is unclear to me what type of sterical clashes might occur. Figure 1 shows then a comparison between crystal structures and simulated structures. They are reasonably similar but standard deviations in the fluctuations of the simulation are smaller than in the experiments. Why did the authors not choose smaller c_i-values to have a better fit? Do smaller values lead to unwanted large fluctuations that would lead to steric clashes between the two DNA turns? I also wonder what side views of the nucleosomes look like (experiments and simulations) and whether in this side view larger fluctuations of the phosphates can be observed in the simulation that would eventually lead to turn-turn clashes for smaller c_i-values.

Reviewer #3 (Public Review):

Summary:

In this study, the authors utilize biophysical modeling to investigate differences in free energies and nucleosomal configuration probability density of CpG islands and nonmethylated regions in the genome. Toward this goal, they develop and apply the cgNA+ coarse-grained model, an extension of their prior molecular modeling framework.

Strengths:

The study utilizes biophysical modeling to gain mechanistic insight into nucleosomal occupancy differences in CpG and nonmethylated regions in the genome.

Weaknesses:

Although the overall study is interesting, the manuscripts need more clarity in places. Moreover, the rationale and conclusion for some of the analyses are not well described.

eLife assessment

This study presents valuable findings linking circHMGCS1 and miR-4521 in diabetes-induced vascular endothelial dysfunction. The evidence supporting the claims of the authors is solid, but addressing concerns around how certain experiments were performed and controlled could enhance clarity and further strengthen the study. The work will be of interest to biomedical scientists working with cardiovascular and/or RNA biology, particularly those studying diabetes.

Reviewer #1 (Public Review):

Summary:

HMGCS1, 3-hydroxy-3-methylglutaryl-CoA synthase1 is predicted to be involved in Acetyl-CoA metabolic process and mevalonate-cholesterol pathway. To induce diet-induced diabetes, they fed wild-type littermates either a standard chow (Control) or a high fat-high sucrose (HFHG) diet, where the diet composition consisted of 60% fat, 20% protein, and 20% carbohydrate (H10060, Hfkbio, China). The dietary regimen was maintained for 14 weeks. Throughout this period, body weight and fasting blood glucose (FBG) levels were measured on a weekly basis. Although the authors induced diabetes with a diet also rich in fat, the cholesterol concentration or metabolism was not investigated. After the treatment, were the animals with endothelial dysfunction? How was the blood pressure of the animals?

Strengths:

To explore the potential role of circHMGCS1 in regulating endothelial cell function, the authors cloned exons 2-7 of HMGCS1 into lentiviral vectors for ectopic overexpression of circHMGCS1 (Figure S2). The authors could use this experiment as a concept proof and investigate the glucose concentration in the cell culture medium. Is the pLV-circ HMGCS1 transduction in HUVEC increasing the glucose release? (Line 163)

Weaknesses:

(1) Pg 20. The cells were transfected with miR-4521 mimics, miR-inhibitor, or miR-NC and incubated for 24 hours. Subsequently, the cells were treated with PAHG for another 24 hours.

Were the cells transfected with lipofectanine? The protocol or the lipofectamine kit used should be described. The lipofectamine protocol suggests using an incubation time of 72 hours. Why did the authors incubate for only 24 hours?

If the authors did the mimic and inhibitor curves, these should be added to the supplementary figures. Please, describe the miRNA mimic and antagomir concentration used in cell culture.

(2) Pg 20, line 507. What was the miR-4521 agomiR used to treatment of the animals?

(3) Figure 1B. The results are showing the RT-qPCR for only 5 circRNA, however, the results show 48 circRNAs were upregulated, and 18 were downregulated (Figure S1D). Why were the other cicRNAs not confirmed? The circRNAs upregulated with high expression are not necessarily with the best differential expression comparing control vs. PAHG groups. Furthermore, Figure 1A and S1D show circRNAs downregulated also with high expression. Why were these circRNAs not confirmed?

(4) Figure 1B shows the relative circRNAs expression. Were host genes expressed in the same direction?

(5) Line 128. The circRNA RT-qPCR methodology was not described. The methodology should be described in detail in the Methods Session.

(6) Line 699. The relative gene expression was calculated using the 2-ΔΔCt method. This is not correct, the expression for miRNA and gene expression are represented in percentage of control.

(7) Line 630. Detection of ROS for tissue and cells. The methodology for tissue was described, but not for cells.

(8) Line 796. RNA Fluorescent In Situ Hybridization (RNA-FISH). Figure 1F shows that the RNA-Fluorescence in situ hybridization (RNA-FISH) confirmed the robust expression of cytoplasmic circHMGCS1 in HUVECs (Figure 1F). However, in the methods, lines 804 and 805 described the probes targeting circMAP3K5 and miR-4521 were applied to the sections. Hybridization was performed in a humid chamber at 37{degree sign}C overnight. Is it correct?

(9) Line 14. Fig 1-H. The authors discuss qRT-PCR demonstrated that circHMGCS1 displayed a stable half-life exceeding 24 h, whereas the linear transcript HMGCS1 mRNA had a half-life less than 8 h (Figure 1H).<br /> Several of the antibodies may contain trace amounts of RNases that could degrade target RNA and could result in loss of RNA hybridization signal or gene expression. Thus, all of the solutions should contain RNase inhibitors. The HMGCS1 mRNA expression could be degraded over the incubation time (0-24hs) leading to incorrect results. Moreover, in the methods is not mentioned if the RNAse inhibitor was used. Please, could the authors discuss and provide information?

(10) Further experiments demonstrated that the overexpression of circHMGCS1 stimulated the expression of adhesion molecules (VCAM1, ICAM1, and ET-1) (Figures 2B and 2C), suggesting that circHMGCS1 is involved in VED. How were these genes expressed in the RNA-seq?

(11) Line 256. By contrast, the combined treatment of circHMGCS1 and miR-4521 agomir did not significantly affect the body weight and blood glucose levels. OGTT and ITT experiments demonstrated that miR-4521 agomir considerably enhanced glucose tolerance and insulin resistance in diabetic mice (Figures 5C, 5D, and Figures S5B and S5C). Why didi the miR-4521 agomir treatment considerably enhance glucose tolerance and insulin resistance in diabetic mice, but not the blood glucose levels?

(12) In the experiments related to pull-down, the authors performed Biotin-coupled miR-4521 or its mutant probe, which was employed for circHMGCS1 pull-down. This result only confirms the Luciferase experiments shown in Figure 4A. The experiment that the authors need to perform is pull-down using a biotin-labeled antisense oligo (ASO) targeting the circHMGCS1 backsplice junction sequence followed by pulldown with streptavidin-conjugated magnetic beads to capture the associated miRNAs and RNA binding proteins (RBPs). Also, the ASO pulldown assay can be coupled to miRNA RT-qPCR and western blotting analysis to confirm the association of miRNAs and RBPs predicted to interact with the target circRNA.

(13) In Figure 5, the authors showed that the results suggest that miR-4521 can inhibit the occurrence of diabetes, whereas circHMGCS1 specifically dampens the function of miR-4521, weakening its protective effect against diabetes. In this context, what are the endogenous target genes for the miR-4521 that could be regulating diabetes?

(14) In the western blot of Figure 5, the β-actin band appears to be different from the genes analyzed. Was the same membrane used for the four proteins? The Ponceau S membrane should be provided.

(15) Why did the authors use AAV9, since the AAV9 has a tropism for the liver, heart, skeletal muscle, and not to endothelial vessels?

Reviewer #2 (Public Review):

Summary:

The authors observed an aggravated vascular endothelial dysfunction upon overexpressing circHMGCS1 and inhibiting miR-4521. This study discovered that circHMGCS1 promotes arginase 1 expression by sponging miR-4521, which accelerated the impairment of vascular endothelial function.

Strengths:

The study is systematic and establishes the regulatory role of the circHMGCS1-miR-4521 axis in diabetes-induced cardiovascular diseases.

Weaknesses:

(1) The authors selected the miR-4521 as the target based on their reduced expression upon circHMGCS1 overexpression. Since the miRNA level is downregulated, the downstream target gene is expected to be upregulated even in the absence of circRNA. The changes in miRNA expression opposite to the levels of target circRNA could be through Target RNA-Directed MicroRNA Degradation. In addition, miRNA can also be stabilized by circRNAs. Hence, selecting miRNA targets based on opposite expression patterns and concluding miRNA sponging by circRNA needs further evidence of direct interactions.

(2) The majority of the experiments were performed with an overexpression vector which can generate a lot of linear RNAs along with circRNAs. The linear RNAs produced by the overexpression vectors can have a similar effect to the circRNA due to sequence identity.

(3) There is a lack of data of circHMGCS1 silencing and its effect on target miRNA & mRNAs.

eLife assessment

This useful manuscript presents an interesting multi-modal omics analysis of lung adenocarcinoma patients with distinct clinical clusters, mutation hotspots, and potential risk factors identified in cases linked to air pollution. The findings show potential for high clinical and therapeutic impact. However, some of the conclusions are incomplete as they are based on correlative or suggestive findings, and would benefit from further functional investigation and validating approaches.

Reviewer #1 (Public Review):

Summary:

This is a well-written and detailed manuscript showing important results on the molecular profile of 4 different cohorts of female patients with lung cancer.

The authors conducted comprehensive multi-omic profiling of air-pollution-associated LUAD to study the roles of the air pollutant BaP. Utilizing multi-omic clustering and mutation-informed interface analysis, potential novel therapeutic strategies were identified.

Strengths:

The authors used several different methods to identify potential novel targets for therapeutic interventions.

Weaknesses:

Statistical test results need to be provided in comparisons between cohorts.

Reviewer #2 (Public Review):

Summary:

Zhang et al. performed a proteogenomic analysis of lung adenocarcinoma (LUAD) in 169 female never-smokers from the Xuanwei area (XWLC) in China. These analyses reveal that XWLC is a distinct subtype of LUAD and that BaP is a major risk factor associated with EGFR G719X mutations found in the XWLC cohort. Four subtypes of XWLC were classified with unique features based on multi-omics data clustering.

Strengths:

The authors made great efforts in performing several large-scale proteogenomic analyses and characterizing molecular features of XWLCs. Datasets from this study will be a valuable resource to further explore the etiology and therapeutic strategies of air-pollution-associated lung cancers, particularly for XWLC.

Weaknesses:

(1) While analyzing and interpreting the datasets, however, this reviewer thinks that authors should provide more detailed procedures of (i) data processing, (ii) justification for choosing methods of various analyses, and (iii) justification of focusing on a few target gene/proteins in the datasets for further validation in the main text.

(2) Importantly, while providing the large datasets, validating key findings is minimally performed, and surprisingly there is no interrogation of XWLC drug response/efficacy based on their findings, which makes this manuscript descriptive and incomplete rather than conclusive. For example, testing the efficacy of XWLC response to afatinib combined with other drugs targeting activated kinases in EGFR G719X mutated XWLC tumors would be one way to validate their datasets and new therapeutic options.

(3) The authors found MAD1 and TPRN are novel therapeutic targets in XWLC. Are these two genes more frequently mutated in one subtype than the other 3 XWLC subtypes? How these mutations could be targeted in patients?

(4) In Figures 2a and b: while Figure 2a shows distinct genomic mutations among each LC cohort, Figure 2b shows similarity in affected oncogenic pathways (cell cycle, Hippo, NOTCH, PI3K, RTK-RAS, and WNT) between XWLC and TNLC/CNLC. Considering that different genomic mutations could converge into common pathways and biological processes, wouldn't these results indicate commonalities among XWLC, TNLC, and CNLC? How about other oncogenic pathways not shown in Figure 2b?

(5) In Figure 2c, how and why were the four genes (EGFR, TP53, RBM10, KRAS) selected? What about other genes? In this regard, given tumor genome sequencing was done, it would be more informative to provide the oncoprints of XWLC, TSLC, TNLC, and CNLC for complete genomic alteration comparison.

(6) Supplementary Table 11 shows a number of mutations at the interface and length of interface between a given protein-protein interaction pair. Such that, it does not provide what mutation(s) in a given PPI interface is found in each LC cohort. For example, it fails to provide whether MAD1 R558H and TPRN H550Q mutations are found significantly in each LC cohort.

(7) Figure 7c and d are simulation data not from an actual binding assay. The authors should perform a biochemical binding assay with proteins or show that the mutation significantly alters the interaction to support the conclusion.

Reviewer #3 (Public Review):

Summary:

The manuscript from Zhang et al. utilizes a multi-omics approach to analyze lung adenocarcinoma cases in female never smokers from the Xuanwei area (XWLC cohort) compared with cases associated with smoking or other endogenous factors to identify mutational signatures and proteome changes in lung cancers associated with air pollution. Mutational signature analysis revealed a mutation hotspot, EGFR-G719X, potentially associated with BaP exposure, in 20% of the XWLC cohort. This correlated with predicted MAPK pathway activations and worse outcomes relative to other EGFR mutations. Multi-omics clustering, including RNA-seq, proteomics, and phosphoproteomics identified 4 clusters with the XWLC cohort, with additional feature analysis pathway activation, genetic differences, and radiomic features to investigate clinical diagnostic and therapeutic strategy potential for each subgroup. The study, which nicely combines multi-modal omics, presents potentially important findings, that could inform clinicians with enhanced diagnosis and therapeutic strategies for more personalized or targeted treatments in lung adenocarcinoma associated with air pollution. The authors successfully identify four distinct clusters with the XWLC cohort, with distinct diagnostic characteristics and potential targets. However, many validating experiments must be performed, and data supporting BaP exposure linkage to XWLC subtypes is suggestive but incomplete to conclusively support this claim. Thus, while the manuscript presents important findings with the potential for significant clinical impact, the data presented are incomplete in supporting some of the claims and would benefit from validation experiments.

Strengths:

Integration of omics data from multimodalities is a tremendous strength of the manuscript, allowing for cross-modal comparison/validation of results, functional pathway analysis, and a wealth of data to identify clinically relevant case clusters at the transcriptomic, translational, and post-translational levels. The inclusion of phosphoproteomics is an additional strength, as many pathways are functional and therefore biologically relevant actions center around activation of proteins and effectors via kinase and phosphatase activity without necessarily altering the expression of the genes or proteins.

Clustering analysis provides clinically relevant information with strong therapeutic potential both from a diagnostic and treatment perspective. This is bolstered by the individual microbiota, radiographic, wound healing, outcomes, and other functional analyses to further characterize these distinct subtypes.

Visually the figures are well-designed and presented and for the most part easy to follow. Summary figures/histograms of proteogenomic data, and specifically highlighted genes/proteins are well presented.

Molecular dynamics simulations and 3D binding analysis are nice additions.

While I don't necessarily agree with the authors' interpretation of the microbiota data, the experiment and results are very interesting, and clustering information can be gleaned from this data.

Weaknesses:

Statistical methods for assessing significance may not always be appropriate.

Necessary validating experiments are lacking for some of the major conclusions of the paper.

Many of the conclusions are based on correlative or suggestive results, and the data is not always substantive to support them.

Experimental design is not always appropriate, sometimes lacking necessary controls or large disparity in sample sizes.

Conclusions are sometimes overstated without validating measures, such as in BaP exposure association with the identified hotspot, kinase activation analysis, or the EMT function.

eLife assessment

This valuable work provides novel insights into the substrate binding mechanism of a tripartite ATP-independent periplasmic (TRAP) transporter. The structural analysis is convincing, but evidence to support some of the conclusions regarding the mechanism is incomplete. This study will be of interest to the membrane transport and bacterial biochemistry communities.

Reviewer #1 (Public Review):

Summary:

This manuscript reports the substrate-bound structure of SiaQM from F. nucleatum, which is the membrane component of a Neu5Ac-specific Tripartite ATP-dependent Periplasmic (TRAP) transporter. Until recently, there was no experimentally derived structural information regarding the membrane components of the TRAP transporter, limiting our understanding of the transport mechanism. Since 2022, there have been 3 different studies reporting the structures of the membrane components of Neu5Ac-specific TRAP transporters. While it was possible to narrow down the binding site location by comparing the structures to proteins of the same fold, a structure with substrate bound has been missing. In this work, the authors report the Na+-bound state and the Na+ plus Neu5Ac state of FnSiaQM, revealing information regarding substrate coordination. In previous studies, 2 Na+ ion sites were identified. Here, the authors also tentatively assign a 3rd Na+ site. The authors reconstitute the transporter to assess the effects of mutating the binding site residues they identified in their structures. Of the 2 positions tested, only one of them appears to be critical to substrate binding.

Strengths:

The main strength of this work is the capture of the substrate-bound state of SiaQM, which provides insight into an important part of the transport cycle.

Weaknesses:

The main weakness is the lack of experimental validation of the structural findings. The authors identified the Neu5Ac binding site, but only tested 2 residues for their involvement in substrate interactions, which was very limited. The authors tentatively identified a 3rd Na+ binding site, which if true would be an impactful finding, but this site was not tested for its contribution to Na+ dependent transport, and the authors themselves report that the structural evidence is not wholly convincing. This lack of experimental validation undermines the confidence of the findings. However, the reporting of these new data is important as it will facilitate follow-up studies by the authors or other researchers.

Reviewer #2 (Public Review):

In this exciting new paper from the Ramaswamy group at Purdue, the authors provide a new structure of the membrane domains of a tripartite ATP-independent periplasmic (TRAP) transporter for the important sugar acid, N-acetylneuraminic acid or sialic acid (Neu5Ac). While there have been a number of other structures in the last couple of years (the first for any TRAP-T) this is the first to trap the structure with Neu5Ac bound to the membrane domains. This is an important breakthrough as in this system the ligand is delivered by a substrate-binding protein (SBP), in this case, called SiaP, where Neu5Ac binding is well studied but the 'hand over' to the membrane component is not clear. The structure of the membrane domains, SiaQM, revealed strong similarities to other SBP-independent Na+-dependent carriers that use an elevator mechanism and have defined Na+ and ligand binding sites. Here they solve the cryo-EM structure of the protein from the bacterial oral pathogen Fusobacterium nucleatum and identify a potential third (and theoretically predicted) Na+ binding site but also locate for the first time the Neu5Ac binding site. While this sits in a region of the protein that one might expect it to sit, based on comparison to other transporters like VcINDY, it provides the first molecular details of the binding site architecture and identifies a key role for Ser300 in the transport process, which their structure suggests coordinates the carboxylate group of Neu5Ac. The work also uses biochemical methods to confirm the transporter from F. nucleatum is active and similar to those used by selected other human and animal pathogens and now provides a framework for the design of inhibitors of these systems.

The strengths of the paper lie in the locating of Neu5Ac bound to SiaQM, providing important new information on how TRAP transporters function. The complementary biochemical analysis also confirms that this is not an atypical system and that the results are likely true for all sialic acid-specific TRAP systems.

The main weakness is the lack of follow-up on the identified binding site in terms of structure-function analysis. While Ser300 is shown to be important, only one other residue is mutated and a much more extensive analysis of the newly identified binding site would have been useful.

Reviewer #3 (Public Review):

The manuscript by Goyal et al reports substrate-bound and substrate-free structures of a tripartite ATP-independent periplasmic (TRAP) transporter from a previously uncharacterized homolog, F. nucleatum. This is one of the most mechanistically fascinating transporter families, by means of its QM domain (the domain reported in his manuscript) operating as a monomeric 'elevator', and its P domain functioning as a substrate-binding 'operator' that is required to deliver the substrate to the QM domain; together, this is termed an 'elevator with an operator' mechanism. Remarkably, previous structures had not demonstrated the substrate Neu5Ac bound. In addition, they confirm the previously reported Na+ binding sites and report a new metal binding site in the transporter, which seems to be mechanistically relevant. Finally, they mutate the substrate binding site and use proteoliposomal uptake assays to show the mechanistic relevance of the proposed substrate binding residues.

The structures are of good quality, the functional data is robust, the text is well-written, and the authors are appropriately careful with their interpretations. Determination of a substrate-bound structure is an important achievement and fills an important gap in the 'elevator with an operator' mechanism. Nevertheless, I have concerns with the data presentation, which in its current state does not intuitively demonstrate the discussed findings. Furthermore, the structural analysis appears limited, and even slight improvements in data processing and resulting resolution would greatly improve the authors' claims. I have several suggestions to hopefully improve the clarity and quality of the manuscript.

Author response:

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

eLife assessment

This structural and biochemical study of the mouse homolog of acidic mammalian chitinase (AMCase) enhances our understanding of the pH-dependent activity and catalytic properties of mouse AMCase and sheds light on its adaptation to different physiological pH environments. The methods and analysis of data are solid, providing several lines of evidence to support a development of mechanistic hypotheses. While the findings and interpretation will be valuable to those studying AMCase in mice, the broader significance, including extension of the results to other species including human, remain unclear.

Public Reviews:

Reviewer #1 (Public Review):

General comments:

This paper investigates the pH-specific enzymatic activity of mouse acidic mammalian chitinase (AMCase) and aims to elucidate its function's underlying mechanisms. The authors employ a comprehensive approach, including hydrolysis assays, X-ray crystallography, theoretical calculations of pKa values, and molecular dynamics simulations to observe the behavior of mouse AMCase and explore the structural features influencing its pH-dependent activity.

The study's key findings include determining kinetic parameters (Kcat and Km) under a broad range of pH conditions, spanning from strong acid to neutral. The results reveal pH-dependent changes in enzymatic activity, suggesting that mouse AMCase employs different mechanisms for protonation of the catalytic glutamic acid residue and the neighboring two aspartic acids at the catalytic motif under distinct pH conditions.

The novelty of this research lies in the observation of structural rearrangements and the identification of pH-dependent mechanisms in mouse AMCase, offering a unique perspective on its enzymatic activity compared to other enzymes. By investigating the distinct protonation mechanisms and their relationship to pH, the authors reveal the adaptive nature of mouse AMCase, highlighting its ability to adjust its catalytic behavior in response to varying pH conditions. These insights contribute to our understanding of the pH-specific enzymatic activity of mouse AMCase and provide valuable information about its adaptation to different physiological conditions.

Overall, the study enhances our understanding of the pH-dependent activity and catalytic properties of mouse AMCase and sheds light on its adaptation to different physiological pH environments.

Reviewer #2 (Public Review):

Summary:

In this study of the mouse homolog of acidic mammalian chitinase, the overall goal is to provide a mechanistic explanation for the unusual observation of two pH optima for the enzyme. The study includes biochemical assays to establish kinetic parameters at different solution pH, structural studies of enzyme/substrate complexes, and theoretical analysis of amino acid side chain pKas and molecular dynamics.

Strengths:

The biochemical assays are rigorous and nicely complemented by the structural and computational analysis. The mechanistic proposal that results from the study is well rationalized by the observations in the study.

Weaknesses:

The overall significance of the work could be made more clear. Additional details could be provided about the limitations of prior biochemical studies of mAMC that warranted the kinetic analysis. The mouse enzyme seems unique in terms of its behavior at high and low pH, so it remains unclear how the work will enhance broader understanding of this enzyme class. It was also not clear can the findings be used for therapeutic purposes, as detailed in the abstract, if the human enzyme works differently.

We have edited the paper to address these concerns

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

Major comments:

(1) Regarding the pH profiles of mouse AMCase, previous studies have reported its activity at pH 2.0 and within the pH range of 3-7. In this paper, the authors conducted kinetic measurements and showed that pH 6.5 is optimal for kcat/Km. The authors emphasize the significance of mouse AMCase's activity in the neutral region, particularly at pH 6.5, for understanding its physiological relevance in humans. To provide a comprehensive overview, it would be valuable for the authors to summarize the findings from previous and current studies, discuss their implications for future pulmonary therapy in humans, and cite relevant literature. Additionally, the authors should highlight their research's specific contributions and novel findings, such as the determination of kinetic parameters (Kcat and Km) under different pH conditions. Emphasizing why previous studies may have required these observations and underscoring the importance of the present findings in addressing those knowledge gaps will help readers understand the significance of the study and its impact on the field of enzymology.

We thank the reviewer for this comment. In keeping with the knowledge gaps addressed directly by this paper, we have not augmented the discussion of future pulmonary therapy in humans. We have summarized the present findings at the end of the introduction as follows:

“We measured the mAMCase hydrolysis of chitin, which revealed significant activity increase under more acidic conditions compared to neutral or basic conditions. To understand the relationship between catalytic residue protonation state and pH-dependent enzyme activity, we calculated the theoretical pKa of the active site residues and performed molecular dynamics (MD) simulations of mAMCase at various pHs. We also directly observed conformational and chemical features of mAMCase between pH 4.74 to 5.60 by solving X-ray crystal structures of mAMCase in complex with oligomeric GlcNAcn across this range.”

(2) Regarding the implications of the pKa values and Asp138 orientation for the pH optima, it would be valuable for the authors to discuss the variations in optimal activity by pH among GH-18 chitinases and investigate the underlying factors contributing to these differences. In particular, exploring the role of Asp138 orientation in chitotriosidase, another mammalian chitinase, would provide important insights. Chitotriosidase is known to be inactive at pH 2.0, and it would be interesting to investigate whether the observed orientation of Asp138 towards Glu140 in mouse AMCase for pH 2.0 activity is lacking in chitotriosidase.

There are similar rotations of the two acidic residues in the literature on Chit1. The variety of crystal pH conditions and the lack of a straightforward mechanism for pKa shifts in AMCase make it difficult to draw a comparison to why Chit1 is inactive at low pH, but this is an interesting area for future study. See a more full discussion in: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2760363/

Furthermore, considering the lower activity of human AMCase at pH 2.0, it would be worthwhile to examine whether the Asp138 orientation towards Glu140, as observed in mouse AMCase, is also absent in human AMCase. Exploring this aspect will help determine if the orientation of Asp138 plays a critical role in pH-dependent activity in human AMCase.

The situation for hAMCase is similar to Chit1 as the rotations observed here for mAMCase are also present. It is not the whether Asp138 can rotate, but rather the relevant energetic penalties as we discuss in the manuscript.

(3) In a previous study by Okawa et al.(Loss and gain of human acidic mammalian chitinase activity by nonsynonymous SNPs. Mol Biol Evol 33, 3183-3193, 2016), it was reported that specific amino acid substitutions (N45D, D47N, and R61M) encoded by nonsynonymous single nucleotide polymorphisms (nsSNPs) in the N-terminal region of human AMCase had distinct effects on its chitinolytic activity. Introducing these three residues (N45D, D47N, and R61M) could activate human AMCase. This activation significantly shifted the optimal pH from 4-5 to 2.0.

Considering the significant impact of these amino acid substitutions on the pH-dependent activity of human AMCase, the authors should discuss this point in the manuscript's discussion section. Incorporating the findings and relating them to the current study's observations on pH optima and Asp138 orientation can provide a comprehensive understanding of the factors influencing pH-dependent activity in AMCase.

We added a citation and dicuss how the mutations identified by this study could potentially shift the pKa of key catalytic residues:

“Okawa et al identified how primate AMCase lost activity by integration of specific, potentially pKa-shifting, mutations relative to the mouse counterpart42b.”

(4) To further strengthen the discussion, the authors could explore the ancestral insectivorous nature of placental mammals and the differences in chitinase activity between herbivorous and omnivorous species. Incorporating these aspects would add depth and relevance to the overall discussion of AMCase. AMCase is an enzyme known for its role in digesting insect chitin in the stomachs of various insectivorous and omnivorous animals, including bats, mice, chickens, pigs, pangolins, common marmosets, and crab-eating monkeys 1-7. However, in certain animals, such as dogs (carnivores) and cattle (herbivores), AMCase expression and activity are significantly low, leading to impaired chitin digestion 8. These observations suggest a connection between dietary habits and the expression and activity of the AMCase gene, ultimately influencing chitin digestibility across different animal species 8.

(1) Strobelet al. (2013). Insectivorous bats digest chitin in the stomach using acidic mammalian chitinase. PloS one 8, e72770.

(2) Ohno et al. (2016). Acidic mammalian chitinase is a proteases-resistant glycosidase in mouse digestive system. Sci Rep 6, 37756.

(3) Tabata et al. (2017). Gastric and intestinal proteases resistance of chicken acidic chitinase nominates chitin-containing organisms for alternative whole edible diets for poultry. Sci Rep 7, 6662.

(4) Tabata et al. (2017). Protease resistance of porcine acidic mammalian chitinase under gastrointestinal conditions implies that chitin-containing organisms can be sustainable dietary resources. Sci Rep 7, 12963.

(5) Ma et al. (2018). Acidic mammalian chitinase gene is highly expressed in the special oxyntic glands of Manis javanica. FEBS Open Bio 8, 1247-1255.

(6) Tabata et al. (2019). High expression of acidic chitinase and chitin digestibility in the stomach of common marmoset (Callithrix jacchus), an insectivorous nonhuman primate. Sci. Rep. 9. 159.

(7) Uehara et al. (2021). Robust chitinolytic activity of crab-eating monkey (Macaca fascicularis) acidic chitinase under a broad pH and temperature range. Sci. Rep. 11, 15470.

(8) Tabata et al. (2018). Chitin digestibility is dependent on feeding behaviors, which determine acidic chitinase mRNA levels in mammalian and poultry stomachs. Sci Rep 8, 1461.

This overall point is covered by our brief discussion on diet differences:

“However, hAMCase is likely too destabilized at low pH to observe an increase in _k_cat. hAMCase may be under less pressure to maintain high activity at low pH due to humans’ noninsect-based diet, which contains less chitin compared to other mammals with primarily insect-based diets42. “

(5) It is important for the authors to clearly state the limitations of their simulations and emphasize the need for experimental validation or additional supporting evidence. This will provide transparency and enable readers to understand the boundaries of the study's findings. A comprehensive discussion of limitations would contribute to a more robust interpretation of the results.

We added a sentence to the discussion:

“Our simulations have important limitations that could be overcome by quantum mechanical simulations that allow for changes in protonation state and improved consideration of polarizability.”

Minor comments:

(1) Regarding the naming of AMCase, it is important to accurately describe it based on its acidic isoelectric point rather than its enzymatic activity under acidic conditions based on the original paper (Reference #14 (Boot, R. G. et al. Identification of a novel acidic mammalian chitinase distinct from chitotriosidase. J. Biol. Chem. 276, 6770-6778 (2001)).

We have made this modification

(2) In the introduction, providing more context regarding the terminology of acidic mammalian chitinase (AMCase) would be beneficial. While AMCase was initially discovered in mice and humans, subsequent research has revealed its presence in various vertebrates, including birds, fish, and other species. Therefore, it would be appropriate to include the alternative enzyme name, Chia (chitinase, acidic), in the introduction to reflect its broader distribution across different organisms. This clarification would enhance the readers' understanding of the enzyme's taxonomy and facilitate further exploration of its functional significance in diverse biological systems.

We have made this modification

(3) The authors mention that AMCase is active in tissues with neutral pHs, such as the lung. However, it is important to consider that the pH in the lung is lower, around 5, due to the presence of dissolved CO2 that forms carbonic acid. The lung microenvironment is known to vary, and specific regions or conditions within the lung may have slightly different pH levels. By addressing the pH conditions in the lungs and their relationship to AMCase's activity, the authors can enhance our understanding of the enzyme's function within its physiological context. A thorough discussion of the specific pH conditions in the lung and their implications for AMCase's activity would provide valuable insights into the enzyme's role in lung pathophysiology.

To keep the focus on the insights we have made, we have elected not to expand this discussion.

(4) It would be helpful for the authors to provide more information about the substrate or products of AMCase. The basic X-ray crystal structures used in this study are GlcNAc2 or GlcNAc3, known products of AMCase. Including details about the specific ligands involved in the enzymatic reactions would enhance the understanding of the study's focus.

We are unclear about what this means - and since it is a minor comment, we have elected not to change the discussion of substrates here.

(5) The authors should critically evaluate the inclusion of the term "chitin-binding" in the Abstract and Introduction. Suppose substantial evidence or discussion regarding the specific chitin-binding properties of the enzyme or its relevance to the immune response needs to be included. In that case, removing or modifying that statement might be appropriate.

We are unclear about what this means - and since it is a minor comment, we have elected not to change the discussion of “chitin-binding” here.

(6) The authors developed an endpoint assay to measure the activity of mouse AMCase across a broad pH range, allowing for direct measurement of kinetic parameters. The authors should provide a more detailed description of the methods used, including any specific modifications made to the previous assay, to ensure reproducibility and facilitate further research in the field. It is important to clearly show the novelty of their endpoint assay compared to previous methods employed in other reports. The authors should also explain how their modified endpoint assay differs from existing assays and highlight its advancements or improvements. This will help readers understand the unique features and contributions of the assay in the context of previous methods.

We have included a detailed method description and figures already. See also our previous paper by Barad which includes other, related, assays.

(7) The authors suggest that mouse AMCase may be subject to product inhibition, potentially due to its transglycosylation activity, which can affect the Michaelis-Menten model predictions at high substrate concentrations. However, the reviewer needed help understanding the specific impact of transglycosylation on the kinetic parameters. It would be helpful for the authors to provide a more appropriate and detailed explanation, clarifying how transglycosylation activity influences the kinetic behavior of AMCase and its implications for the observed results.

The experiments to conclusively demonstrate this are beyond our current capabilities.

(8) In the Abstract, the authors state, "We also solved high resolution crystal structures of mAMCase in complex with chitin, where we identified extensive conformational ligand heterogeneity." This reviewer suggests replacing "chitin" with "oligomeric GlcNAcn" throughout the text, specifically about biochemical experiments. It is important to accurately describe the experimental conditions and ligands used in the study.

We have made these changes throughout the manuscript

(9) In the introduction, the authors mention "a polymer of β(1-4)-linked N-acetyl-D-glucosamine (GlcNAc)". In this case, the letter "N" should be italicized to conform to the proper notation for the monosaccharide abbreviation.

corrected (and hopefully would have been done so by the copy editor!)

(10) In the introduction, the authors state, "In the absence of AMCase, chitin accumulates in the airways, leading to epithelial stress, chronic activation of type 2 immunity, and age-related pulmonary fibrosis5,6". It is recommended to clarify that "AMCase" refers to "acidic mammalian chitinase (AMCase)" in this context, as it is the first mention of the enzyme in the introduction.

We moved that section so that it flows better and is introduced with the full name.

(11) In the introduction, the authors state, "Mitigating the negative effects of high chitin levels is particularly important for mammalian lung and gastrointestinal health." This reviewer requests further clarification on the connection between chitin and gastrointestinal health. Please provide an explanation or reference to support this statement.

We have modified this sentence to:

“Chitin levels can be potentially important for mammalian lung and gastrointestinal health.”

(12) In the introduction, the authors mention that "Acidic Mammalian Chitinase (AMCase) was originally discovered in the stomach and named for its high enzymatic activity under acidic conditions." It is recommended to include Reference #14 (Boot et al. J. Biol. Chem. 276, 6770-6778, 2001) as it provides the first report on mouse and human AMCase, contributing to the understanding of the enzyme.

However, it is worth noting that while this paragraph primarily focuses on human tissues, Reference #14 primarily discusses mouse AMCase but also reports on human AMCase. Additionally, References #8 and #9 mainly discuss mouse AMCase. This creates confusion in the description of human and mouse AMCase within the paragraph.

Considering that this paper aims to focus on the unique features of mouse AMCase, it is suggested that the authors provide a more specific and balanced description of both human and mouse AMCase throughout the main text..

We have clarified the origin of the name AMCase and the results distinguish the two orthologs in the text with h or mAMCase.

(13) Figure 1A in the Introduction section has been previously presented in several papers. The authors should consider moving this figure to the Results section and present an alternative figure based on their experimental results to enhance the novelty and impact of the study.

We have considered this option, but prefer the original placement.

(14) In the Results section, the authors mentioned, "Prior studies have focused on relative mAMCase activity at different pH18,20, limiting the ability to define its enzymological properties precisely and quantitatively across conditions of interest." It would be beneficial for the authors to include reference #14, the first report showing the pH profile of mouse AMCase, to support their statement.

We have added this reference

(15) Regarding the statement, "To overcome the pH-dependent fluorescent properties of 4MU-chitobioside, we reverted the assay into an endpoint assay, which allowed us to measure substrate breakdown across different pH (Supplemental Figure 1A)", the authors should provide a more detailed description of the improvements made to measure AMCase activity. Additionally, it would be helpful to include a thorough explanation of the figure legend for Supplementary Figure 1A to provide clarity to readers.

We have included a detailed method description and figures already. See also our previous paper by Barad which includes other, related, assays.

(16) Figure 1B shows that the authors used the AMCase catalytic domain. It would benefit the authors to explain the rationale behind this choice in the figure legend or the main text.

This point is addressed in the text:

“Previous structural studies on AMCase have focused on interactions between inhibitors like methylallosamidin and the catalytic domain of the protein.”

(17) For Figures 1C-E, it is recommended that the authors include error bars in their results to represent the variability or uncertainty of the data. In Figure 1E, the authors should clarify the units of the Y-axis (e.g., sec-1 µM-1). Additionally, in Figure 1F, the authors should explain how the catalytic acidity is shown.

We have added error bars and axis labels. Figure 1F is conceptual, so we are leaving it as is.

(18) The authors stated, "These observations raise the possibility that mAMCase, unlike other AMCase homologs, may have evolved an unusual mechanism to accommodate multiple physiological conditions." It would be helpful for the authors to compare and discuss the pH-dependent AMCase activity of mouse AMCase with other AMCase homologs to support this statement.

That is an excellent idea for future comparative studies, but beyond the scope of what we are examining in this paper.

(19) The authors should explain Supplemental Figures 1B and C in the Results or Methods sections to provide context for these figures.

We are unclear about what this means - and since it is a minor comment, we have elected not to change these sections.

(20) Supplemental Figure 3 is missing any description. It would be important for the authors to include a mention of this figure in the main text before Supplemental Figure 4 to guide the readers.

The full legend is in there now and the reference to Supplemental 4 was mislabeled.

(21) For Supplemental Figure 4, the authors should explain the shape of the symbol used in the figure. Additionally, they should explain "apo" and "holoenzyme" in the context of this figure.

Unclear what a shape means in this context - perhaps the confusion arises because these are violin plots showing distributions.

(22) Table 1 requires a more detailed explanation of its contents. Additionally, Tables 2 and 3 need to be included. The authors should include these missing tables in the revised version and explain their contents appropriately.

Table 1 is the standard crystallographic table - there isn’t much more detailed explanation that can be offered. Tables 2 and 3 were not transferred properly by BioRxiv but were included in the review packet as requested a day after submission.

(23) In Figure 4, it would be beneficial to enlarge Panels A-C to improve the ease of comprehension for readers. Additionally, it is recommended to use D136, D138, and E140 instead of D1, D2, and E to label the respective parts. The authors should also explain the meaning of the symbol used in the figure.

Since it is a minor comment, we have elected not to change these figures.

(24) In Figure 5, it would be beneficial to enlarge Panels A-C to improve the ease of comprehension for readers.

Since it is a minor comment, we have elected not to change these figures.

(25) Similarly, in Figure 6, all panels should be enlarged to enhance the ease of comprehension for readers.

Since it is a minor comment, we have elected not to change these figures.

Reviewer #2 (Recommendations For The Authors):

In general, I did not identify many detailed or technical concerns with the work. A few items for the authors to consider are listed below.

(1) The interpretation of the crystallographic datasets seems complicated by the heterogeneity in the substrate component. It might be nice to see more critical analysis of the approach here. Are there other explanations or possible models that were considered? Do other structures of chitinases or other polysaccharide hydrolases exhibit the same phenomenon?

We have tried in writing it to provide a very critical approach to this and it is quite likely that other structures contain unmodeled density containing similar heterogeneity (but it is just unmodeled).

(2) It would be ideal to include more experimental validation of the proposed mechanism. Much of the manuscript includes theoretical validations (pKa estimation, dynamics, etc) - but it would be optimal to make an enzyme variant or do an experiment with a substrate analog.

Yes - we agree that follow on experiments are needed to fully test the mechanism and that those will be the subject of future work.

(3) For an uninitiated reviewer, I think the major issue with this study is that the broader significance of the work and how it fits into the context of other work on these enzymes is not clear. It would be helpful to be more specific about what we know of mechanism from work on other enzymes to help the reader understand the motivation for this study.

We have added w few additional references, guided by reviewer 1 comments, that should help in this respect.

eLife assessment

This structural and biochemical study of the mouse homolog of acidic mammalian chitinase (AMCase) enhances our understanding of the pH-dependent activity and catalytic properties of mouse AMCase, and it sheds light on its adaptation to different physiological pH environments. The methods and analysis of data are solid, providing several lines of evidence to support the development of mechanistic hypotheses. While the findings and interpretation will be valuable to those studying AMCase in mice, the broader significance, including extension of the results to other species including human, remain less clear.

Reviewer #1 (Public Review):

General comments:

This paper investigates the pH-specific enzymatic activity of mouse acidic mammalian chitinase (AMCase) and aims to elucidate its function's underlying mechanisms. The authors employ a comprehensive approach, including hydrolysis assays, X-ray crystallography, theoretical calculations of pKa values, and molecular dynamics simulations to observe the behavior of mouse AMCase and explore the structural features influencing its pH-dependent activity.

The study's key findings include determining kinetic parameters (Kcat and Km) under a broad range of pH conditions, spanning from strong acid to neutral. The results reveal pH-dependent changes in enzymatic activity, suggesting that mouse AMCase employs different mechanisms for protonation of the catalytic glutamic acid residue and the neighboring two aspartic acids at the catalytic motif under distinct pH conditions.<br /> The novelty of this research lies in the observation of structural rearrangements and the identification of pH-dependent mechanisms in mouse AMCase, offering a unique perspective on its enzymatic activity compared to other enzymes. By investigating the distinct protonation mechanisms and their relationship to pH, the authors reveal the adaptive nature of mouse AMCase, highlighting its ability to adjust its catalytic behavior in response to varying pH conditions. These insights contribute to our understanding of the pH-specific enzymatic activity of mouse AMCase and provide valuable information about its adaptation to different physiological conditions.<br /> Overall, the study enhances our understanding of the pH-dependent activity and catalytic properties of mouse AMCase and sheds light on its adaptation to different physiological pH environments.

Comments on revised version:

In their revised manuscript, the authors have made significant efforts to address the reviewers' comments.

eLife assessment

This important manuscript presents several structures of the Kv1.2 voltage-gated potassium channel, based on state-of-the-art cryoEM techniques and algorithms. The authors present solid evidence for structures of DTX-bound Kv1.2 and of Kv1.2 in potassium-free solution (with presumably sodium ions bound within the selectivity filter). These structures advance our knowledge of the molecular basis of the channel inactivation process.

Reviewer #1 (Public Review):

In this manuscript by Wu et al., the authors present the high resolution cryoEM structures of the WT Kv1.2 voltage-gated potassium channel. Along with this structure the authors have solved several structures of mutants or experimental conditions relevant to the slow inactivation process that these channels undergo and which is not yet completely understood.

One of the main findings is the determination of the structure of a mutant (W366F) that is thought to correspond to the slow inactivated state. These experiments confirm results in similar mutants in different channels from Kv1.2 that indicate that inactivation is associated with an enlarged selectivity filter.

Another interesting structure is the complex of Kv1.2 with the pore blocking toxin Dendrotoxin 1. The results shown in the revised version indicate that the mechanism of block is similar to that of related blocking-toxins, in which a lysine residue penetrates in the pore. Surprisingly, in these new structures, the bound toxin results in a pore with empty external potassium binding sites.

The quality of the structural data presented in this revised manuscript is very high and allows for unambiguous assignment of side chains. The conclusions are supported by the data. This is an important contribution that should further our understanding of voltage-dependent potassium channel gating. In the revised version, the authors have addressed my previous specific comments, which are appended below.

(1) In the main text's reference to Figure 2d residues W18' and S22' are mentioned but are not labeled in the insets.

(2) On page 8 there is a discussion of how the two remaining K+ ions in binding sites S3 and S4 prevent permeation K+ in molecular dynamics. However, in Shaker, inactivated W434F channels can sporadically allow K+ permeation with normal single-channel conductance but very reduced open times and open probability at not very high voltages.

(3) The structures of WT in the absence of K+ shows a narrower selectivity filter, however Figure 4 does not convey this finding. In fact, the structure in Figure 4B is constructed in such an angle that it looks as if the carbonyl distances are increased, perhaps this should be fixed. Also, it is not clear how the distances between carbonyls given in the text on page 12 are measured. Is it between adjacent or kitty-corner subunits?

(4) It would be really interesting to know the authors opinion on the driving forces behind slow inactivation. For example, potassium flux seems to be necessary for channels to inactivate, which might indicate a local conformational change is the trigger for the main twisting events proposed here.

Reviewer #2 (Public Review):

Cryo_EM structures of the Kv1.2 channel in the open, inactivated, toxin complex and in Na+ are reported. The structures of the open and inactivated channels are merely confirmatory of previous reports. The structures of the dendrotoxin bound Kv1.2 and the channel in Na+ are new findings that will of interest to the general channel community.

Review of the resubmission:

I thank the authors for making the changes in their manuscript as suggested in the previous review. The changes in the figures and the additions to the text do improve the manuscript. The new findings from a further analysis of the toxin channel complex are welcome information on the mode of the binding of dendrotoxin.

A few minor concerns:<br /> (1) Line 93-96, 352: I am not sure as to what is it the authors are referring to when they say NaK2P. It is either NaK or NaK2K. I don't think that it has been shown in the reference suggested that either of these channels change conformation based on the K+ concentration. Please check if there is a mistake and that the Nichols et. al. reference is what is being referred to.

(2) Line 365: In the study by Cabral et. al., Rb+ ions were observed by crystallography in the S1, S3 and S4 site, not the S2 site. Please correct.

Reviewer #3 (Public Review):

Wu et al. present cryo-EM structures of the potassium channel Kv1.2 in open, C-type inactivated, toxin-blocked and presumably sodium-bound states at 3.2 Å, 2.5 Å, 2.8 Å, and 2.9 Å. The work builds on a large body of structural work on Kv1.2 and related voltage-gated potassium channels. The manuscript presents a plethora of structural work, and the authors are commended on the breadth of the studies. The structural studies are well-executed. Although the findings are mostly confirmatory, they do add to the body of work on this and related channels. Notably, the authors present structures of DTx-bound Kv1.2 and of Kv1.2 in a low concentration of potassium (which may contain sodium ions bound within the selectivity filter). These two structures add considerable new information. The DTx structure has been markedly improved in the revised version and the authors arrive at well-founded conclusions regarding its mechanism of block. Regarding the Na+ structure, the authors claim that the structure with sodium has "zero" potassium - I caution them to make this claim. It is likely that some K+ persists in their sample and that some of the density in the "zero potassium" structure may be due to K+ rather than Na+. This can be clarified by revisions to the text and discussion. I do not think that any additional experiments are needed. Overall, the manuscript is well-written, a nice addition to the field, and a crowning achievement for the Sigworth lab.

Most of this reviewer's initial comments have been addressed in the revised manuscript. Some comments remain that could be addressed by revisions of the text.

Specific comments on the revised version:<br /> Quotations indicate text in the manuscript.<br /> (1) "While the VSD helices in Kv1.2s and the inactivated Kv1.2s-W17'F superimpose very well at the top (including the S4-S5 interface described above), there is a general twist of the helix bundle that yields an overall rotation of about 3o at the bottom of the VSD."

Comment: This seemed a bit confusing. I assume the authors aligned the complete structures - the differences they indicate seem to be slight VSD repositioning relative to the pore rather than differences between the VSD conformations themselves. The authors may wish to clarify. As they point out in the subsequent paragraph, the VSDs are known to be loosely associated with the pore.

(2) Comment: The modeling of DTx into the density is a major improvement in the revision. Figure 3 displays some interactions between the toxin and Kv1.2 - additional side views of the toxin and the channel might allow the reader to appreciate the interactions more fully. The overall fit of the toxin structure into the density is somewhat difficult to assess from the figure. (The authors might consider using ChimeraX to display density and model in this figure.)

(3) "We obtained the structure of Kv1.2s in a zero K+ solution, with all potassium replaced with sodium, and were surprised to find that it is little changed from the K+ bound structure, with an essentially identical selectivity filter conformation (Figure 4B and Figure 4-figure supplement 1)."

Comment: It should be noted in the manuscript that K+ and Na+ ions cannot be distinguished by the cryo-EM studies - the densities are indistinguishable. The authors are inferring that the observed density corresponds to Na+ because the protein was exchanged from K+ into Na+ on a gel filtration (SEC) column. It is likely that a small amount of K+ remains in the protein sample following SEC. I caution the authors to claim that there is zero K+ in solution without measuring the K+ content of the protein sample. Additionally, it should be considered that K+ may be present in the blotting paper used for cryo-EM grid preparation (our laboratory has noted, for example, a substantial amount of Ca2+ in blotting paper). The affinity of Kv1.2 for K+ has not been determined, to my knowledge - the authors note in the Discussion that the Shaker channel has "tight" binding for K+. It seems possible that some portion of the density in the selectivity filter could be due to residual K+. This caveat should be clearly stated in the main text and discussion. More extensive exchange into Na+, such as performing the entire protein purification in NaCl, or by dialysis (as performed for obtaining the structure of KcsA in low K+ by Y. Zhou et al. & Mackinnon 2001), would provide more convincing removal of K+, but I suspect that the Kv1.2 protein would not have sufficient biochemical stability without K+ to endure this treatment. One might argue that reduced biochemical stability in NaCl could be an indication that there was a meaningful amount of K+ in the final sample used for cryo-EM (or in the particles that were selected to yield the final high-resolution structure).

(4) Referring to the structure obtained in NaCl: "The ion occupancy is also similar, and we presume that Kv1.2 is a conducting channel in sodium solution."

Comment: Stating that "Kv1.2 is a conducting channel in sodium solution" and implying that conduction of Na+ is achieved by an analogous distribution of ion binding sites as observed for K+ are strong statements to make - and not justified by the experiments provided. Electrophysiology would be required to demonstrate that the channel conducts sodium in the absence of K+. More complete ionic exchange, better control of the ionic conditions (Na+ vs K+), and affinity measurements for K+ would be needed to determine the distribution of Na+ in the filter (as mentioned above). At minimum, the authors should revise and clarify what the intended meaning of the statement "we presume that Kv1.2 is a conducting channel in sodium solution". As mentioned above, it seems possible/likely that a portion of the density in the filter may be due to K+.

Author response:

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

Public Reviews: 

Reviewer #1 (Public Review): 

In this manuscript by Wu et al., the authors present the high-resolution cryoEM structures of the WT Kv1.2 voltage-gated potassium channel. Along with this structure, the authors have solved several structures of mutants or experimental conditions relevant to the slow inactivation process that these channels undergo and which is not yet completely understood. 

One of the main findings is the determination of the structure of a mutant (W366F) that is thought to correspond to the slow inactivated state. These experiments confirm results in similar mutants in different channels from Kv1.2 that indicate that inactivation is associated with an enlarged selectivity filter. 

Another interesting structure is the complex of Kv1.2 with the pore-blocking toxin Dendrotoxin 1. The results show that the mechanism of the block is different from similar toxins, in which a lysine residue penetrates the pore deep enough to empty most external potassium binding sites. 

The quality of the structural data presented in this manuscript is very high and allows for the unambiguous assignment of side chains. The conclusions are supported by the data. This is an important contribution that should further our understanding of voltagedependent potassium channel gating. Specific comments are appended below. 

(1) In the mains text's reference to Figure 2d residues W18' and S22' are mentioned but are not labeled in the insets. 

Now labeled in Fig. 2D

(2) On page 8 there is a discussion of how the two remaining K+ ions in binding sites S3 and S4 prevent permeation K+ in molecular dynamics. However, in Shaker, inactivated W434F channels can sporadically allow K+ permeation with normal single-channel conductance but very reduced open times and open probability at not very high voltages. 

Addressed in the Discussion, lines 480-490.

(3) The structures of WT in the absence of K+ show a narrower selectivity filter, however, Figure 4 does not convey this finding. In fact, the structure in Figure 4B is constructed at such an angle that it looks as if the carbonyl distances are increased, perhaps this should be fixed. Also, it is not clear how the distances between carbonyls given in the text on page 12 are measured. Is it between adjacent or kitty-corner subunits? 

We decided to remove mention of carbonyl distances, because at our resolutions the atoms are not resolved.

(4) It would be really interesting to know the authors' opinions on the driving forces behind slow inactivation. For example, potassium flux seems to be necessary for channels to inactivate, which might indicate a local conformational change is the trigger for the main twisting events proposed here. 

We cite Sauer et al. (2011) for the idea that the intact selectivity filter is a strained conformation, and its relaxation yields the wide vestibule seen in NaK2K and Kv channels.  Lines 434-439.

Reviewer #2 (Public Review): 

There are four Kv1.2 channel structures reported: the open state, the C-type inactivated state, a dendrotoxin-bound state, and a structure in Na+. 

A high-resolution crystal structure of the open state for a chimeric Kv1.2 channel was reported in 2007 and there is no new information provided by the cryoEM structure reported in this study. 

The cryo-EM structure of the C-type inactivated state of the Kv1.2 channel was determined for a channel with the W to F substitution in the pore helix. A cryo-EM structure of the Shaker channel and a crystal structure of a chimeric Kv1.2 channel with an equivalent W to F mutation were reported in 2022. Cryo-EM structures of the C-type inactivated Kv1.3 channel are also available. All these previous structures have provided a relatively consistent structural view of the C-type inactivated state and there is no significant new information that is provided by the structure reported in this study. 

A structure of the Kv1.2 channel blocked by dendrotoxin is reported. A crystal structure of charybdotoxin and the chimeric Kv1.2 channel was reported in 2013. Density for dendrotoxin could not be clearly resolved due to symmetry issues and so the definitive information from the structure is that dendrotoxin binds, similarly to charybdotoxin, at the mouth of the pore. A potential new finding is that there is a deeper penetration of the blocking Lys residue in dendrotoxin compared to charybdotoxin. It will however be necessary to use approaches to break the symmetry and resolve the electron density for the dendrotoxin molecule to support this claim and to make this structure significant.  

We have now succeeded in breaking the symmetry and present in Fig. 3 a C1 structure of the toxin-channel complex. In the improved map we now see that our previous conclusion was wrong: the penetration of Lys5 cannot be much deeper than that seen in CTx and ShK structures. However for some reason the pattern of ion-site occupancies in the blocked state is different in this structure than in the others. Fig. 3, Fig. 4E; text lines 559-568.

The final structure reported is the structure of the Kv1.2 channel in K+ free conditions and with Na+ present. The structure of the KcsA channel by the MacKinnon group in 2001 showed a constricted filter and since then it has been falsely assumed by the K channel community that the lowering of K concentration leads to a construction of the selectivity filter. There have been structural studies on the MthK and the NaK2K channels showing a lack of constriction in the selectivity filter in the absence of K+. These results have been generally ignored and the misconception of filter constriction/collapse in the absence of K+ still persists. The structure of the Kv1.2 channel in Na+ provided a clear example that loss of K+ does not necessarily lead to filter constriction. 

We are grateful to the reviewer for pointing out this serious omission. We now cite other work including from the Y. Jiang and C. Nichols labs showing examples of outer pore expansion and destabilization. Page p. 4, lines 90-104; lines 421-439.

The structure in Na+ is significant while the other structures are either merely reproductions of previous reports or are not resolved well enough to make any substantial claims. 

We now state more clearly the confirmatory nature of our Kv1.2 open structure (lines 71-74) and the similarities of the inactivated-channel structures (lines 193196).

Reviewer #3 (Public Review): 

Wu et al. present cryo-EM structures of the potassium channel Kv1.2 in open, C-type inactivated, toxin-blocked and presumably sodium-bound states at 3.2 Å, 2.5 Å, 2.8 Å, and 2.9 Å. The work builds on a large body of structural work on Kv1.2 and related voltage-gated potassium channels. The manuscript presents a large quantity of structural work on the Kv1.2 channel, and the authors should be commended on the breadth of the studies. The structural studies seem well-executed (this is hard to fully evaluate because the current manuscript is missing a data collection and refinement statistics table). The findings are mostly confirmatory, but they do add to the body of work on this and related channels. Notably, the authors present structures of DTXbound Kv1.2 and of Kv1.2 in a low concentration of potassium (with presumably sodium ions bound within the selectivity filter). These two structures add new information, but the studies seem somewhat underdeveloped - they would be strengthened by accompanying functional studies and further structural analyses. Overall, the manuscript is well-written and a nice addition to the field. 

The data collection and refinement table has been added (Fig. 4 supplement 3.)

We agree and regret the lack of functional studies. We have not been able to carry them out because work in our laboratory is winding down and the lab soon will be closing.

Recommendations for the authors: 

Reviewer #2 (Recommendations For The Authors): 

(1) It is not obvious from the data shown how well the side chain positions in the inactivated state are defined by the electron density. These figures should be redone. Maybe the use of stereo would be useful. This will be particularly useful for the reader to decide if the small changes in, for example, the positioning of the carbonyl oxygens are believable. 

Figure 2 – figure supplement 4 shows the stereo views.

(2) The authors note the changes observed (though small) in the VSD which were not observed in other structures. The relevance of this observation is not described. Do these changes arise due to the different environments of detergents versus nanodisc etc. in the different structures?

We’ve now inserted a note about variety of environments and how this might be a cause of the difference: lines 280-285.  

Are there changes in the pore-VSD interface in the inactivated and the open channel structures and if yes, then do mutations at these residues affect inactivation?

There is surprisingly little movement at the S4-S5 interface residues identified by Bassetto et al. (2022) as having effects on inactivation. Lines 262-267.

(3) For the structures in Na+, it is important to provide analytical data showing the biochemical behavior of the channel. This is also true for the wild type and the W to F mutant channel. Size exclusion profiles should be included. 

The SEC profile (noisy, but showing a clear peak) of the channel in Na+ is now shown in Fig. 4 supplement 1. Low expression of the W366F mutant produced even worse SEC results, but we include a representative micrograph of W366F in Na+ to show the monodispersed protein prep. In Figure 5 – figure supplement 1.

Reviewer #3 (Recommendations For The Authors): 

Portions of text from the manuscript are indicated by quotations. 

Introduction: "One goal of the current study was to examine the structure of the native Kv1.2 channel." 

Comment, minor points: The authors refer to the Kv1.2 construct used for the structural studies as "native Kv1.2". I found this somewhat confusing because the word "native" suggests derived from a native source. The phrasing above also gives the impression that the structure by Wu et al is the first structure of Kv1.2. The Kv1.2 construct is essentially identical to the one used by Long et al in 2005 to determine the initial structure of Kv1.2 (PDB 2A79). The authors discuss a subsequent paddle-chimera Kv1.2-2.1 structure from 2007 (PDB 2R9R) in the introduction, but it would be prudent to mention the 2005 one of Kv1.2 as well. The open structure determined by Wu et al. is an improvement on the 2A79 structure in that the 2A79 structure was modeled as a poly-alanine model within the voltage sensor domain. Nevertheless, the Kv1.2-2.1 structure (2R9R) is highly similar to the 2A79 structure of Kv1.2. The 2007 structure indicated that Kv1.2-2.1 recapitulates structural features of Kv1.2. It is therefore not surprising that the open structure presented here is highly similar to that of both PDB 2A79 (Kv1.2) and PDB 2R9R (Kv1.2-2.1).  

We failed to point out the high quality of the original Long et al. 2005 structure and its comparisons with the chimeric structure in Long et al. 2007. We now have tried to correct this: lines 70-74.

Comment: The cryo-EM analyses suggest that a large percentage (most?) of the particles are missing the beta subunit. This should be commented on somewhere.      

Now noted on lines 120-132, we pooled particles with and without beta subunits. 

Regarding ions in the selectivity filter, one-dimensional plots of the density would strengthen the analysis.

Now included in Fig. 4.

Also, one should mention caveats associated with identifying ions in cryo-EM maps and the added difficulty/uncertainty when the density is located along a symmetry axis (C4 axis, due to the possible build-up of noise). C1 reconstructions, showing density within the filter, if possible, would strengthen the analyses.

You are correct. However local resolution is highest in the selectivity filter region. So I think that since the CTF-based filtering is constant over all the structure I think the SNR will be good on axis. 

Comment: The section on channel inactivation could be simplified by stating that the structure is highly similar to W17'F structures of other Kv channels. (And then discussing possible differences).  

We now note, “overall conformational difference is identical…” p. 7, lines 193-196.

"Salt bridges involving the S4 Arg and Lys residues are shifted slightly (Figure 2-figure supplement 3A-D). Arg300 (R3) is in close proximity to Glu226 on the S2 helix for the open channel, while R3 is closer to Glu183 in the S2 helix. The Glu226 side chain adopts a visible interaction with R4 in the inactivated state." 

Comment: The density for these acidic amino acids seems weak, especially in the inactivated state. It seems like a stretch to make much of their possible conformational changes. 

We’ve included stereo pairs in Fig. 2 – figure supplement 4.

"By adding 100 nM α-DTx to detergent solubilized Kv1.2 protein we obtained a cryo-EM structure at 2.8 Å resolution of the complex." 

Comment: 100 nm. might be lower than the Kv concentration. The current methods are ambiguous on the concentration of Kv channel used for the DTx sample. From the methods, it seems possible that 100 nM DTX is a sub-stoichiometric amount relative to the channel. Regardless, the cryo-EM data seems to suggest that a large percentage of particles do not have DTx bound. This surely complicates the interpretation of density within the filter (which has partly been ascribed to a lysine side chain from DTx).

The reviewer correctly points a potentially serious problem. It turns out that the 100nM figure we quoted was incorrect, and the actual concentration of toxin, >400 nM, was substantially greater than the protein concentration. This is confirmed by the small fraction (<1%) of 3D class particles that do not show the toxin density (lines 303-306).

Comment: The methods on atomic structure building/refinement (Protein model building, refinement, and structural analysis) are sparse. A table is needed showing data collection and refinement statistics for each of the structures. This data should also provide average B factors for the ions in the filter. An example can be found in PMID 36224384. 

Data collection and statistics are now in Fig. 4 – figure supplement 3.

"In the selectivity filter of the toxin-bound channel (Figure 3E) a continuous density is seen to extend downward from the external site IS0 through to the boundary between IS1 and IS2. This density is well modeled by an extended Lys side chain from the bound toxin, with the terminal amine coordinated by the carbonyls of G27”.

Comment: While there seems to be extra density in site IS0 from the figures, the density ascribed to lysine in the filter doesn't seem that distinct from those of ions in the open structure. 1-dimensional density plots and some degree of caution may be prudent. Could there, for example, be a mixture of toxin-bound and free channels in the dataset?

Could the lysine penetrate to different depths? If the toxin binds with nM affinity, why are any channels missing the toxin? Have the authors modeled an atomic structure of the entire toxin bound to the channel to evaluate how plausible the proposed binding of the lysine is? Can the toxin be docked onto Kv1.2 with the deep positioning of the lysine and not clash with the extracellular surface of Kv1.2? 

We also were concerned about these issues. We have been able to obtain a C1 reconstruction of the toxin-channel complex. In building the atomic model we found that indeed the Lys5 side chain could not penetrate as far as we had thought, and appears to be coordinated by the first carbonyl pair. Fig. 3; text lines 331-332. 

"Toxin binding shrinks the distances between opposing carbonyl oxygens in the selectivity filter, forming a narrower tunnel into which the Lys side chain fits (Figure 3F). The second and fourth carbonyl oxygen distances are substantially reduced from 4.7 Å and 4.6 Å in an open state to 3.7 Å and 3.9 Å, respectively (Figure 4E). In a superposition of Kv1.2 open-state and α-DTX-bound P-loop structures, there is also an upward shift of the first three carbonyl groups by 0.7~1.0 Å (Figure 4F). " 

Comment: I suspect the authors intend to refer to Figure 3F rather than 4. I would be cautious here. The refined positions of the carbonyl oxygens are almost certainly affected by the presence or absence of ions in the atomic model during refinement. The density and the resolution of the map may not be able to distinguish small changes to the positions of the carbonyl oxygens (and these differences/uncertainties are compounded by the C4 symmetry). 

"On the other hand, the terminal amine of lysine in α-DTX is deeply wedged at the second set of carbonyls, narrowing both IS1 and IS2 while displacing ions from the sites (Figure 3-figure supplement 2A). CTX does not cause narrowing of the selectivity filter or displacements of the carbonyls (Figure 3-figure supplement 2B). "

Comment: Again, caution would be prudent here.  

We are very grateful to the reviewer for pointing out these problems. We have removed these statements that are weakly supported at our resolution level.

"Shaker channels are able to conduct Na+ in the absence of K+ (Melishchuk et al., 1998)." 

Comment: How about the Kv1.2 channel? Is Kv1.2 able to conduct Na+ in the absence of K+ ? This would certainly be relevant for interpreting the conformation of the filter and the density ascribed to Na+ for the structure in sodium.  

We agree wholeheartedly, but unfortunately we are no longer capable of doing the measurements as our lab will soon close.

"Ion densities are seen in the IS1, IS3, and IS4 ion binding sites, but the selectivity filter shows a general narrowing as would be expected for binding of sodium ions. The second, third, and fourth carbonyl oxygen distances are reduced from 4.7 Å, 4.7 Å, and 4.6 Å in the open state to 4.4 Å, 3.9 Å, and 4.5 Å, respectively. The rest of the channel structure is very little perturbed. " 

Comment: The density for IS4 seems weak. To me, it looks like IS1 and IS3 are occupied, whereas IS2 and IS4 are much weaker. 1-dimensional density plots would be helpful. I would suggest caution in commenting too strongly on the "general narrowing" since the resolution of the maps, the local density, and the atomic structure refinement would be consistent with coordinate errors of 0.5 Å or more - and would be compounded (~ doubled) by measuring between symmetry-related atoms.  

We present 1D plots in Fig. 4E. We no longer comment on “narrowing”

"Finally, the snake toxin a-Dendrotoxin (DTx) studied here is seen to block Kv1.2 by insertion of a lysine residue into the pore." 

Comment: Discussion (and references) should be given regarding what was known prior to this study on the mode of inhibition by DTx. 

Discussion and references now added, lines 287-301.

"On the other hand, a lengthy molecular-dynamics simulation of deactivation in the Kv1.2-2.1..." 

Comment: I don't think mentioning this personal communication adds to the manuscript. 

Actually the original “personal communication” reference was there because the situation is complicated. The movie S3 accompanying the Jensen et al. paper shows deactivation and dewetting of the channel during a 250 us simulation. In the movie there are ions visible in the selectivity filter for the first 50 us, but after that the SF appears empty. Puzzled by this we contacted Dr. Jensen who explained that the movie was in error, ions remain in the SF throughout the entire 250 us. We now cite Jensen (2012) along with the personal communication.

"The difference between the open and inactivated Kv1.2 structures, like the difference in Kv1.2-2.1 (Reddi et al., 2022) and Shaker (Tan et al., 2022) can be imagined as resulting from a two-step process." 

Comment: Confusing phrasing because the authors mean to compare their structure to inactivated structures of Kv1.2-2.1 and shaker. 

Fixed, lines 220-222.

"Molecular dynamics simulations by Tan et al. based on the Shaker-W17'F structure show that IS3 and IS4 are simultaneously occupied by K+ ions in the inactivated state." 

Comment: I think that the word "show" is too strong. Perhaps "suggest" 

The MD result seems to us to be unequivocal, that most of the time the two sites are occupied by ions.

References are needed for the following statements:  

-  "as well as the charge-transfer center phenylalanine"

Now citing Tao et al. 2010, line 156.

- "total gating charge movement in Shaker channels is larger, about 13 elementary charges per channel" 

Now citing the review by Islas, 2015 (line 166-169).

"The selectivity filter of potassium channels consists of an array of four copies of the extended loop (the P-loop) formed by a highly conserved sequence, in this case, TTVGYGD. Two residues anchor the outer half of the selectivity filter and are particularly important in inactivation mechanisms (Figure 2B, right panels). Normally, the tyrosine Y28' (Y377 in Kv1.2) is constrained by hydrogen bonds to residues in the pore helix and helix S6 and is key to the conformation of the selectivity filter. The final aspartate of the P-loop, D30' (D379 in Kv1.2) is normally located near the extracellular surface and has a side chain that also participates in H-bonds with W17' (W366 in Kv1.2) on the pore helix." 

Citations added (Pless 2013, Sauer 2011) lines 211-214.

- "During normal conduction, ion binding sites in the selectivity filter are usually occupied by K+ and water molecules in alternation." 

Added Morais-Cabral et al. 2001, p. 17, lines 463-465.

eLife assessment:

This paper characterises a novel gene (Spar), and presenting valuable findings in the field of insect biology and behaviour. The experiments are well designed, with attention to detail, showcasing the potential of the Drosophila melanogaster model and the use of online resources. The mixed approach presents a convincing argument for a genetic interaction between Alk and Spar.

eLife assessment

Receptor tyrosine kinases such as ALK play critical roles during appropriate development and behaviour and are nodal in many disease conditions, through molecular mechanisms that weren't completely understood. This manuscript identifies a previously unknown neuropeptide precursor as a downstream transcriptional target of Alk signalling in Clock neurons in the Drosophila brain. The experiments are well designed with attention to detail, the data are convincing, and the findings will be valuable to those interested in events downstream of signalling by receptor tyrosine kinases.

Reviewer #3 (Public Review):

In this manuscript, the authors explored the interaction between the pattern recognition receptor MDA5 and 5'ppp-RNA in the Miiuy croaker. They found that MDA5 can serve as a substitute for RIG-I in detecting 5'ppp-RNA of Siniperca cheilinus rhabdovirus (SCRV) when RIG-I is absent in Miiuy croaker. Furthermore, they observed MDA5's recognition of 5'ppp-RNA in chickens (Gallus gallus), a species lacking RIG-I. Additionally, the authors documented that MDA5's functionality can be compromised by m6A-mediated methylation and degradation of MDA5 mRNA, orchestrated by the METTL3/14-YTHDF2/3 regulatory network in Miiuy croaker during SCRV infection. This impairment compromises the innate antiviral immunity of fish, facilitating SCRV's immune evasion. These findings offer valuable insights into the adaptation and functional diversity of innate antiviral mechanisms in vertebrates.

eLife assessment

The authors present evidence suggesting that MDA5 can substitute as a sensor for triphosphate RNA in a species that naturally lacks RIG-I. The key findings are potentially important for our understanding of the evolution of innate immune responses. Compared to an earlier version of the paper, the strength of evidence has improved but it is still partially incomplete due to a few key missing experiments and controls.

Reviewer #1 (Public Review):

This study offers valuable insights into host-virus interactions, emphasizing the adaptability of the immune system. Readers should recognize the significance of MDA5 in potentially replacing RIG-I and the adversarial strategy employed by 5'ppp-RNA SCRV in degrading MDA5 mediated by m6A modification in different species, further indicating that m6A is a conservational process in the antiviral immune response.

However, caution is warranted in extrapolating these findings universally, given the dynamic nature of host-virus dynamics. The study provides a snapshot into the complexity of these interactions, but further research is needed to validate and extend these insights, considering potential variations across viral species and environmental contexts. Additionally, it is noted that the main claims put forth in the manuscript are only partially supported by the data presented.

Reviewer #2 (Public Review):

This manuscript by Geng et al. aims to demonstrate that MDA5 compensates for the loss of RIG-I in certain species, such as teleofish miiuy croacker. The authors use siniperca cheats rhabdovirus (SCRV) and poly(I:C) to demonstrate that these RNA ligands induce an IFN response in an MDA5-dependent manner in m.miiuy derived cells. Furthermore, they show that MDA5 requires its RD domain to directly bind to SCRV RNA and to induce an IFN response. They use in vitro synthesized RNA with a 5'triphosphate (or lacking a 5'triphosphate as a control) to demonstrate that MDA5 can directly bind to 5'-triphosphorylated RNA. The second part of the paper is devoted to m6A modification of MDA5 transcripts by SCRV as an immune evasion strategy. The authors demonstrate that the modification of MDA5 with m6A is increased upon infection and that this causes increased decay of MDA5 and consequently a decreased IFN response.

- One critical caveat in this study is that it does not address whether ppp-SCRV RNA induces IRF3-dimerization and type I IFN induction in an MDA5 dependent manner. The data demonstrate that mmiMDA5 can bind to triphosphorylated RNA (Fig. 4D). In addition, triphosphorylated RNA can dimerize IRF3 (4C). However, a key experiment that ties these two observations together is missing.<br /> - Specifically, although Fig. 4C demonstrates that 5'ppp-SCRV RNA induces dimerization (unlike its dephosphorylated or capped derivatives), this does not proof that this happens in an MDA5-dependent manner. This experiment should have been done in WT and siMDA5 MKC cells side-by-side to demonstrate that the IRF3 dimerization that is observed here is mediated by MDA5 and not by another (unknown) protein. The same holds true for Fig. 4J.<br /> - Fig 1C-D: these experiments are not sufficiently convincing, i.e. the difference in IRF3 dimerization between VSV-RNA and VSV-RNA+CIAP transfection is minimal.<br /> - Fig. 2N and 2O: why did the authors decide to use overexpression of MDA5 to assess the impact of STING on MDA5-mediated IFN induction? This should have been done in cells transfected with SCRV or polyIC (as in 2D-G) or in infected cells (as in 2H-K). In addition, it is a pity that the authors did not include an siMAVS condition alongside siSTING, to investigate the relative contribution of MAVS versus STING to the MDA5-mediated IFN response. Panel O suggests that the IFN response is completely dependent on STING, which is hard to envision.<br /> - Fig. 3F and 3G: where are the mock-transfected/infected conditions? Given that ectopic expression of hMDA5 is known to cause autoactivation of the IFN pathway, the baseline ISG levels should be shown (ie. In absence of a stimulus or infection). Normalization of the data does not reveal whether this is the case and is therefore misleading.<br /> - Fig. 4F and 4G: can the authors please indicate in the figure which area of the gel is relevant here? The band that runs halfway the gel? If so, the effects described in the text are not supported by the data (i.e. the 5'OH-SCRV and 5'pppGG-SCRV appear to compete with Bio-5'ppp-SCRV as well as 5'ppp-SCRV).<br /> - My concerns about Fig. 5 remain unaltered. The fact that MDA5 is an ISG explains its increased expression and increased methylation pattern. The authors should at the very least mention in their text that MDA5 is an ISG and that their observations may be partially explained by this fact.

Author response:

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

eLife assessment

The authors present evidence suggesting that MDA5 can substitute as a sensor for triphosphate RNA in a species that naturally lacks RIG-I. The key findings are potentially important for our understanding of the evolution of innate immune responses, but the evidence is incomplete, as additional biochemical and functional experiments are needed to unambiguously assign MDA5 as a bona fide sensor of triphosphate RNA in this model. This also leaves the title as overstating its case.

We would like to thank the editorial team for these positive comments on our manuscript and the constructive suggestions to improve our manuscript. According to the suggestions and valuable comments of the referees, we have added substantial amounts of new data and analysis to substantiate our claims, and the manuscript, including the title, has been carefully revised to better reflect our conclusions. We are now happy to send you our revised manuscript, we hope the modified manuscript addresses your and the reviewers’ concerns satisfactorily and is suitable for publication in eLife now.

Reviewer #1 (Public Review):

This study offers valuable insights into host-virus interactions, emphasizing the adaptability of the immune system. Readers should recognize the significance of MDA5 in potentially replacing RIG-I and the adversarial strategy employed by 5'ppp-RNA SCRV in degrading MDA5 mediated by m6A modification in different species, further indicating that m6A is a conservational process in the antiviral immune response.

However, caution is warranted in extrapolating these findings universally, given the dynamic nature of host-virus dynamics. The study provides a snapshot into the complexity of these interactions, but further research is needed to validate and extend these insights, considering potential variations across viral species and environmental contexts.

We concur with the viewpoint that virus-host coevolution complicates the derivation of universal conclusions. To address this challenge, incorporated additional experiments and data based on the suggestions of the reviewers. These experiments were carried out across diverse models, including two distinct vertebrate species (M. miiuy and G. gallus), two different viruses (SCRV and VSV), and the synthesis of corresponding 5’ppp-RNA probes. We believe that these supplementary data bolster the evidence supporting the immune replacement role of MDA5 in the recognition of 5'ppp-RNA in RIG-I deficient species (Figure 1C-1E, Figure 2O and 2P, Figure 4). Moreover, we have duly incorporated references in both the introduction and discussion sections to further support our conclusion that MDA5 in T. belangeri, a mammal lacking RIG-I, possesses the ability to detect RNA viruses posed as RIG-I agonists (doi: 10.1073/pnas.1604939113). Lastly, meticulous revisions have been undertaken in the manuscript, including adjustments to the title, to ensure harmonization with our research outcomes.

Reviewer#2 (Public Review):

This manuscript by Geng et al. aims to demonstrate that MDA5 compensates for the loss of RIG-I in certain species, such as teleost fish miiuy croaker. The authors use siniperca cheats rhabdovirus (SCRV) and poly(I:C) to demonstrate that these RNA ligands induce an IFN response in an MDA5-dependent manner in M. miiuy derived cells. Furthermore, they show that MDA5 requires its RD domain to directly bind to SCRV RNA and to induce an IFN response. They use in vitro synthesized RNA with a 5'triphosphate (or lacking a 5'triphosphate as a control) to demonstrate that MDA5 can directly bind to 5'-triphosphorylated RNA. The second part of the paper is devoted to m6A modification of MDA5 transcripts by SCRV as an immune evasion strategy. The authors demonstrate that the modification of MDA5 with m6A is increased upon infection and that this causes increased decay of MDA5 and consequently a decreased IFN response.

The key message of this paper, i.e. MDA5 can sense 5'-triphosphorylated RNA and thereby compensate for the loss of RIG-I, is novel and interesting, yet there is insufficient evidence provided to prove this hypothesis. Most importantly, it is crucial to test the capacity of in vitro synthesized 5'-triphosphorylated RNA to induce an IFN response in MDA5-sufficient and -deficient cells. In addition, a number of important controls are missing, as detailed below.

To further support the notion that MDA5 is capable of detecting 5'ppp-RNA in species lacking RIG-I, we conducted additional experiments. Initially, we isolated the RNA from SCRV and VSV viruses. Subsequently, we synthesized 5'ppp-RNA probes that corresponded to the genome termini of SCRV and VSV in vitro. Then, these RNAs were treated with Calf intestinal phosphatase (CIAP) to generate dephosphorylated derivatives. Next, we separately tested the activation ability of various RNAs on IRF3 dimer and IFN response in MKC (M. miiuy kidney cell line) and DF-1 (G. gallus fibroblast cell line) cells, and determined that the immune activation ability of SCRV/VSV viruses depends on their triphosphate structure (Figure 1C-1E, Figure 4C and 4J). In addition, the knockdown of MDA5 inhibited the immune response mediated by SCRV RNA (Figure 2P and 2Q). Finally, we incorporated essential experimental controls (Figure 4B and 4I). We think that the inclusion of these supplementary experimental data significantly enhances the credibility and further substantiates our hypothesis.

The authors describe an interaction between MDA5 and STING which, if true, is very interesting. However, the functional implications of this interaction are not further investigated in the manuscript. Is STING required to relay signaling downstream of MDA5?

To better explore the role of STING in MDA5 signal transduction, we constructed a STING expression plasmid and synthesized specific siRNA targeting STING. Next, we found that co-expression of STING and MDA5 significantly enhance MDA5-mediated IFN-1 response during SCRV virus infection (Figure 2N). Conversely, silencing of STING expression restored the MDA5-mediated IFN-1 response (Figure 2O). These findings provide important evidence for the critical involvement of STING in the immune signaling cascade mediated by MDA5 in response to 5'ppp-RNA viruses.

The second part of the paper is quite distinct from the first part. The fact that MDA5 is an interferon-stimulated gene is not mentioned and complicates the analyses (i.e. is there truly more m6A modification of MDA5 on a per molecule basis, or is there simply more total MDA5 and therefore more total m6A modification of MDA5).

For the experimental data analysis in Figure 5E and 5F, we first compared the m6A-IP group to the input group, and then normalized the control group (IgG group of 5E and Mock group of 5F) to a value of “1”. Given the observed variability in MDA5 expression levels within the input group of Mock and SCRV virus-infected cells, our analysis represents the actual m6A content of each MDA5 molecule. To enhance clarity, we have updated the label on the Y-axis in Figure 5E and 5F.

Finally, it should be pointed out that several figures require additional labels, markings, or information in the figure itself or in the accompanying legend to increase the overall clarity of the manuscript. There are frequently details missing from figures that make them difficult to interpret and not self-explanatory. These details are sometimes not even found in the legend, only in the materials and methods section. The manuscript also requires extensive language editing by the editorial team or the authors.

We acknowledge the valuable feedback from the reviewer and have made significant improvements to our manuscript based on the recommendations provided in the "Recommendation for the authors" section. Furthermore, we have conducted a thorough review of the entire article, resulting in substantial enhancements to the format, clarity, and overall readability of our manuscript.

Reviewer#3 (Public Review):

Summary: In this manuscript, the authors investigated the interaction between the pattern recognition receptor MDA5 and 5'ppp-RNA in a teleost fish called Miiuy croaker. They claimed that MDA5 can replace RIG-I in sensing 5'ppp-RNA of Siniperca cheats rhabdovirus (SCRV) in the absence of RIG-I in Miiuy croaker. The recognition of MDA5 to 5'ppp-RNA was also observed in the chicken (Gallus gallus), a bird species that lacks RIG-I. Additionally, they reported that the function of MDA5 can be impaired through m6A-mediated methylation and degradation of MDA5 mRNA by the METTL3/14-YTHDF2/3 regulatory network in Miiuy croaker under SCRV infection. This impairment weakens the innate antiviral immunity of fish and promotes the immune evasion of SCRV.

Strengths:<br /> These findings provide insights into the adaptation and functional diversity of innate antiviral activity in vertebrates.

Weaknesses:<br /> However, there are some major and minor concerns that need to be further addressed. Addressing these concerns will help the authors improve the quality of their manuscript.One significant issue with the manuscript is that the authors claim to be investigating the role of MDA5 as a substitute for RIG-I in recognizing 5'ppp-RNA, but their study extends beyond this specific scenario. Based on my understanding, it appears that sections 2.2, 2.3, 2.5, 2.6, and 2.7 do not strictly adhere to this particular scenario. Instead, these sections tend to investigate the functional involvement of Miiuy croaker MDA5 in the innate immune response to viral infection. Furthermore, the majority of the data is focused on Miiuy croaker MDA5, with only a limited and insufficient study on chicken MDA5. Consequently, the authors cannot make broad claims that their research represents events in all RIG-I deficient species, considering the limited scope of the species studied.

We agree with the reviewer's perspective that functional analysis of MDA5 in M. miiuy may not adequately represent all species lacking RIG-I. To address this concern, we have incorporated additional experimental data utilizing different model systems, including two different vertebrate species (M. miiuy and G. gallus), two distinct viruses (SCRV and VSV), and the synthesis of two corresponding 5’ppp-RNA probes. While the functional characterization of G. gallus MDA5 remains relatively limited compared to M. miiuy, our current experimental findings provide support for two key observations. Firstly, the triphosphate structure of the VSV virus is pivotal in activating the innate immune response in G. gallus against the virus (Figure 1D and 4J). Secondly, G. gallus MDA5 can recognize 5’ppp-RNA (Figure 4I, 4K and 4L). Consequently, although we cannot definitively establish the immune surrogate function of MDA5 in all RIG-I-deficient species, our research data further substantiates this hypothesis. Moreover, we have adopted a more cautious attitude in summarizing our experimental conclusions, thereby enhancing the rigor of our manuscript language.

The current title of the article does not align well with its actual content. It is recommended that the focus of the research be redirected to the recognition function and molecular mechanism of MDA5 in the absence of RIG-I concerning 5'ppp-RNA. This can be achieved through bolstering experimental analysis in the fields of biochemistry and molecular biology, as well as enhancing theoretical research on the molecular evolution of MDA5. It is advisable to decrease or eliminate content related to m6A modification.

Following the reviewer's recommendations, we have revised the title to emphasize that our main research focus is a teleost fish devoid of RIG-I. Furthermore, we have conducted additional molecular experiments to further elucidate the 5'ppp-RNA recognition function of MDA5 in RIG-I-deficient species. In an attempt to analyze the potential molecular evolution of MDA5 resulting from RIG-I deficiency, we collected MDA5 coding sequences from diverse vertebrates. However, due to multiple independent loss events of RIG-I in fish, fish with or without RIG-I genes in the phylogenetic tree cannot be effectively clustered separately, making it extremely difficult to perform this aspect of analysis. Consequently, we have regrettably opted to forgo the molecular evolution analysis of MDA5.

Our article topic is to reveal an antagonistic phenomenon between fish receptor and RNA viruses. The MDA5 of RIG-I-lost fish has evolved the ability to recognize 5’ppp-RNA virus and mediate IFN response to resist SCRV infection. Conversely, the m6A methylation mechanism endows the SCRV virus with a means to weaken the immune capacity of MDA5. Therefore, we believe that the latter part is an important part of the arms race between the virus and its host, and should be retained.

Additionally, the main body of the writing contains several aspects that lack rigor and tend to exaggerate, necessitating significant improvement.

We appreciate the reviewer’s comment and have improved the manuscript addressing the points raised in the “Recommendation for the authors”. We have added corresponding experiments to strengthen the verification of the conclusions, and in addition, we are more cautious in summarizing the language of the conclusions.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

(1) The evidential foundation within the Result 1 section appears somewhat tenuous.

Firstly, the author derives conclusions regarding the phenomenon of RIG-I loss in lower vertebrates by referencing external literature and conducting bioinformatics analyses. It is pertinent to inquire whether the author considered fortifying these findings through additional WB/PCR experiments, particularly for evaluating RIG-I expression levels across diverse vertebrates, encompassing both lower and higher orders.

Firstly, the species we analyzed are mostly model species with excellent genomic sequence information in the database. Secondly, the RIG-I protein sequences (at least some domain sequences) are relatively conserved in vertebrates. Therefore, the credibility of evaluating the existence of RIG-I in these species through homology comparison is high. Therefore, we do not intend to conduct additional PCR/WB experiments to confirm this.

Additionally, following the identification of RIG-I loss, the author postulates MDA5 as a substitute of RIG-I, grounding this speculation in the analysis of MDA5 and LGP2 protein structures. It is imperative to address whether the author could enhance the manuscript by supplying expression data for MDA5 and LGP2 across different vertebrates and elucidating further why MDA5 is posited as the compensatory mechanism for RIGI loss.

Like MDA5, LGP2 is also an interferon-stimulating gene, so they both likely exhibit high sensitivity to viral infections. Therefore, we think that comparing the expression data of these two genes is difficult to evaluate their function. In mammals, the regulatory mechanisms of LGP2 to RIG-I and MDA5 were complicated and ambiguous. To evaluate the potential function of LGP2 in M. miiuy, we further constructed LGP2 plasmid and synthesized siRNA targeting LGP2. Then, our results indicate that mmiLGP2 can enhance the antiviral immune response mediated by mmiMDA5 (Figure 1H and 1I), further indicating the regulatory role of mmiLGP2 in RLR signaling, rather than acting as a compensatory receptor for RIG-I.

Also, is it conceivable that other receptors contribute to this compensatory effect in lower vertebrates?

5’ triphosphate short blunt-end double-strand RNA is the ligand of RIG-I as contained in the panhandle of negative-strand viral genomes. We mainly focus on the immune recognition and compensatory effects of other receptors on RIG-I loss, and MDA5, as the protein with the most similar structure, first attracted our attention. In addition, IFIT proteins have been reported to recognize triphosphate single-stranded RNA (doi: 10.1038/nature11783). However, we used SCRV and VSV RNA as viral models, both of which have negative stranded genomes and meet the ligand standards of RIG-I, rather than IFIT. Therefore, we excluded the IFIT protein from our research scope.

(2) The article exclusively employs a singular type of 5'PPP-RNA virus and one specific lower vertebrate species, thereby potentially compromising the robustness of the assertion that this phenomenon is prevalent in lower vertebrates. To bolster this claim, could the author consider incorporating data from an alternative 5'PPP-RNA virus and a different lower vertebrate species?

To address this concern, we have incorporated additional experimental data utilizing different model systems, including two different vertebrate species (M. miiuy and G. gallus) and two distinct viruses (SCRV and VSV). While the functional characterization of G. gallus MDA5 remains relatively limited compared to M. miiuy, our current experimental findings provide support for two key observations. Firstly, the triphosphate structure of the VSV virus is pivotal in activating the innate immune response in G. gallus against the virus (Figure 1D and 4J). Secondly, G. gallus MDA5 can recognize 5’ppp-RNA (Figure 4I, 4K and 4L). Consequently, these experimental results further confirmed the conservatism of this immune compensation mechanism.

(3) A nuanced consideration of the statement in Result 5 is warranted. Examination of the results under SCRV infection conditions suggests dynamic fluctuations in MDA5 expression levels, challenging the veracity of the statement implying "increased expression", which contradicts the proposed working model of this article.

Because MDA5 acts as a receptor and plays a recognition immune role in the early stages of virus infection, the expression of MDA5 in the early stage of SCRV infection rapidly increases. In the later stage of infection, the expression of MDA5 may gradually decrease again due to the negative feedback mechanism in the host body to prevent excessive inflammation. However, compared to the uninfected group, the expression of MDA5 was significantly increased in the SCRV-infected group, so we believe that the term "increased expression" is not a problem. In addition, the m6A mechanism can weaken the function of MDA5, but it still cannot prevent the overall increase of MDA5 expression, which is not contradictory to the working model in this article.

Additionally, the alterations in m6A levels in miiuy croaker under SCRV infection conditions warrant clarification. Could the author employ m6A dot blotting to supplement the findings related to total m6A levels?

Our previous studies (doi: 10.4049/jimmunol.2200618) have suggested that the total m6A level is increased after SCRV infection in miiuy croaker. We cited this conclusion in the discussion of our manuscript.

(4) It would be beneficial if the editors could assist the author in enhancing the language of the manuscript.

We have carefully checked the full article and modified it with Grammarly tools, and we believe that the grammar, format, and readability of our articles have been greatly improved.

Reviewer #2 (Recommendations For The Authors):

Figure 1

(1) Figure 1B - some clarification needs to be added about this figure in the text. It is unclear what the main point is that the authors would like to convey.

What we want to emphasize is that some species with RIG-I, such as zebrafish, have also experienced RIG-I loss events, but have undergone whole genome replication events before the loss, thus preserving a copy of RIG-I. This indicates that loss events of RIG-I are very common in vertebrates and do not occur randomly. We have elaborated on this point in the results and discussion.

(2) Figure 1C - is not very informative other than showing Mm MDA5 and LGP2 side-by-side. It would be more useful to show a comparison of human RIG-I/MDA5 alongside Mm and Gg MDA5. Are there any conserved/shared key residues between hRIG-I/hMDA5 versus mmMDA5?

Homologous proteins are often known to adopt the same or similar structure and function. We have added human RIG-I domain information to this figure (Figure 1F). By comparing the domain information of human RIG-I with M. miiuy MDA5 and LGP2, M. miiuy MDA5 has a similar structure to human RIG-I, making it most likely to compensate for the missing RIG-I. While M. miiuy LGP2 lacks the CARD domain, which is crucial for signal transduction, so we will shift our focus to M. miiuy MDA5. In addition, we collected protein sequences of MDA5 and RIG-I from various vertebrates to identify key residues evolved in recognizing 5'ppp-RNA by M. miiuy MDA5. However, unfortunately, no potential residues were found during the comparison process.

Figure 2

(1) Figure 2B - It would be important to demonstrate MDA5-Flag expression by immunoblot and compare MDA5-Flag overexpression to endogenous MDA5 expression using the anti-MDA5 antibody from panel 2A. If IF is used, more cells need to be visible in the field.

After transfecting the MDA5 plasmid into MKC, endogenous MDA5 expression was detected using MDA5 antibodies. The results showed a significant increase in MDA5 protein levels, indicating that MDA5 antibodies can specifically recognize MDA5 protein. In addition, we retained the original immunofluorescence images to better demonstrate the subcellular localization of MDA5.

(2) Figure 2C - The 1:1 stoichiometry of MDA5:MAVS (in the absence of any stimulus) is quite surprising. How does the interaction between MDA5 and MAVS change upon stimulation with an RNA ligand (SCRV, poly(I:C))?

We do not believe that the actual stoichiometry between MDA5 and MAVS is what you described as 1:1. In fact, the proportion of proteins in the complex depends on many factors in the experimental results with Co-IP. Firstly, the MDA5 plasmid in this study has a 3 × Flag tag, while the MAVS only has a 1x Myc tag, which makes the antibody more sensitive for detecting MDA5-Flag. In addition, the Co-IP results are also affected by multiple factors such as the type of antibody and the number of recoveries, making it difficult to estimate the actual ratio of MDA5 to MAVS. Based on the above reasons and the fact that the detection of the interaction strength between MDA5 and MAVS after infection seems to be off-topic, we did not continue to explore this point.

(3) Figure 2D - The interaction between MDA5 and STING is a very interesting finding but is not elaborated on in the paper (even though the interaction between MDA5 and STING is mentioned in the abstract). The manuscript would be strengthened if the interaction between MDA5 and STING is further investigated. For example, does the IFN response that is reported in panels 2E to 2H require the presence of STING? Does mmMDA5 signal via STING in response to a DNA ligand?

We appreciate the referee's suggestion to study the mutual influence between MDA5 and STING. We found that co-expression of STING and MDA5 can enhance MDA5-mediated IFN-1 response during SCRV virus infection, while knocking down STING can restore MDA5-mediated IFN-1 (Figure 2N and 2O). This indicates that STING plays an important signaling role in the immune response of MDA5 to RNA viruses. We understand the importance of cGAS/STING pathways in identifying exogenous DNA, so exploring the MDA5 pathway for DNA ligand recognition is an interesting and meaningful perspective. But this seems to be detached from the theme of our article, so we didn't continue to explore this point.

(4) Figures 2F and 2H - the authors demonstrate that SCRV induces a type I IFN response in an MDA5-dependent manner. While SCRV is a single-stranded negative-sense RNA virus that contains 5'ppp-RNA, it cannot be excluded that MDA5 is activated here in response to a double-stranded RNA intermediate of viral origin or even a host-derived RNA whose expression or modification is altered during infection. To demonstrate in an unambiguous manner that MDA5 senses 5'ppp-RNA, it is crucial to use the in vitro synthesized 5'ppp-RNA (and its dephosphorylated derivative as a control) from Fig. 4 in these experiments.

We transfected 5 'ppp SCRV and 5' ppp VSV (and their dephosphorylated derivatives) synthesized in vitro into MKC cells and DF-1 cells, respectively. The results showed that 5’ppp-RNAs significantly promoted the formation of IRF3 dimers, while their dephosphorylated derivatives did not (Figure 4C and 4J). In addition, we extracted virus RNA from the SCRV and VSV viruses and dephosphorylated them with Calf intestinal phosphatase (CIAP). These RNAs were transfected into MKC and DF-1 cells and found that the immune response mediated by virus RNAs was much higher than the dephosphorylated form (Figure 1C-1E). The above results indicate that the immune response activated by SCRV and VSV is indeed dependent on their triphosphate structure. Finally, the IRF3 dimer and IFN induction activated by SCRV RNA can be inhibited by si-MDA5 (Figure 2P and 2Q), further demonstrating the involvement of MDA5 in the immune response mediated by 5’ppp-RNA ligands.

(5) In mice and humans, MDA5 is known to collaborate with LGP2 to jointly induce an IFN response. Does M.miiuy express LGP2? If so, it would be informative to include a siRNA targeting LGP2 in the experiments in panel F. In mammals, LGP2 potentiates the response via MDA5 while it may inhibit RIG-I activation.

M.miiuy express LGP2. We constructed an LGP2 plasmid and synthesized si-LGP2 to investigate the impact of LGP2 on MDA5-mediated immune processes (Figure 1G-1I). The results showed that LGP2 can enhance the IFN response mediated by MDA5 during SCRV virus infection, similar to that in mammals.

(6) Minor comment - Is the poly(I:C) used in this figure high or low molecular weight poly(I:C)? HMW poly(I:C) preferentially stimulates MDA5, while LMW poly(I:C) preferentially stimulates RIG-I.

We used poly(I:C)-HMW as a positive control for activating MDA5. We have modified the relevant information in Figure 2 and its legend.

Figure 3

(1) Figure 3F/G - The normalization in this Figure is difficult to interpret. It would be better to split Figure 3G into 4 separate graphs and include the mock-infected cells alongside the infected samples (as done in Figure 2).

To better demonstrate the function of the RD domain of MDA5 in M. miiuy, we have changed the experimental plan, as shown in figure 3F. We detected the induction of antiviral factors by overexpression of MDA5 and MDA5-△RD under poly (I:C)-HMW stimulation. This can indicate that the RD domain of MDA5 has a conserved function in the recognition of poly(I:C)-HMW in M. miiuy, and can serve as a positive control for the recognition of SCRV virus by the RD domain.

Figure 4

(1) Figure 4B - A number of important controls are missing. Was the immunoprecipitation of RNA successful? This could be shown by running a fraction of the immunoprecipitated material on an RNA gel and/or by showing that the input RNA was depleted after IP. In addition, a control IP (Streptavidin beads without biotinylated RNA) is missing to ensure that MDA5 does not stick non-specifically to the Streptavidin resin.

We appreciate the referee's suggestions. We rerun this experiment and added a non-biomarker RNA IP control group, and the results showed that MDA5 did not adsorb non-specific onto the beads (Figure 4B). In addition, based on the referee's suggestion, we tested the consumption of RNA before and after immunoprecipitation, and the results showed that biotin-labeled RNA, rather than non-biotin-labeled RNA, could be adsorbed by beads, indicating the success of RNA precipitation. However, we think that this is not necessary for the final presentation of the experimental results, so we did not show this in the figure.

(2) Figure 4B - It is unclear why there is such a large molecular weight difference between endogenous MDA5 and MDA5-Flag (110 kDa versus 130/140 kDa). Why is there less MDA5-Flag retrieved than endogenous MDA5?

After careful analysis, we believe that the significant difference in molecular weight between endogenous MDA5 and MDA5 Flag may be due to three reasons. Firstly, MDA5 flag has a 3× Flag tag. Secondly, as shown in the primer table, we constructed MDA5 between the NotI and XbaI cleavage sites in the pcDNA3.1 vector, which are located at the posterior position in the vector. This means that the Flag tag has a certain distance from the starting codon of MDA5, and these sequences on the vector can also be translated and increase the molecular weight of the exogenous MDA5 protein. Finally, in order to facilitate the amplification of the primers, the F-terminal primers of MDA5 contain a small portion of the 3'UTR sequence (excluding the stop codon). These above reasons may have led to significant differences in molecular weight. In addition, in order to supplement important experimental controls, we have conducted a new RNA pull-down experiment as shown in Figure 4B.

(3) Minor point: Figure 4B - please clarify in the figure whether RNA or protein is immunoprecipitated and via which tags.

We have conducted a new RNA pull-down experiment as shown in Fig 4B, and we have clearly labeled the relevant information in the figure.

(4) Figure 4E - the fraction of MDA5 that binds 5'ppp-RNA seems incredibly minor. And why is this experiment done using 5'OH-RNA as a competitor, rather than simply incubating MDA5 and 5'OH-RNA together and demonstrating that these do not form a complex?

The proportion of MDA5 combined with 5’ppp-RNA is influenced by many conditions, including the concentration and purity of the probe and purified protein. In addition, the dosage ratio between the RNA probe and MDA5 protein in the EMSA experiment can also have a significant impact on the results. Therefore, it is not possible to accurately determine the actual binding force between MDA5 and RNA. In the EMSA experimental program, both cold probes (5’ppp-RNA) and mutated cold probes (5’OH-RNA and 5’pppGG-RNA) are crucial for demonstrating the specific binding between MDA5 and 5’ppp-RNA, as they can exclude false positive errors caused by factors such as the presence of biotin in the purified MDA5 protein itself.

(5) Figure 4B/4C/4F - These experiments would be strengthened by including an MDA5 mutant that cannot bind to RNA. These mutants are well-described in mammals. If these residues are conserved, it is straightforward to generate this mutant.

As shown in Figure 3, the MDA5 of M. miiuy has an RD domain that can recognize the SCRV virus. We constructed MDA5-△RD mutant plasmids with 6x His-tags and purified them for EMSA experiments (Figure 4E). The experimental results further indicate that MDA5, rather than MDA5-△RD, can bind to 5’ppp-SCRV (Figure 4G). This further confirms the crucial role of the RD domain in recognizing the 5'ppp-RNA virus.

(6) Minor point: Figure 4E: please clarify in which lanes MDA5 has been added.

Thank you for the referee's suggestion. We have synthesized new 5'ppp-RNA probes (5’ppp-SCRV and their dephosphate derivatives) and rerun this experiment, and relevant information has been added in the Figure (Figure 4F).

Figure 5

(1) Figure 5C - As MDA5 is an interferon-stimulated gene (as shown in panel G/H/I)) the increased MDA5 expression could simply explain the increase in the amount of m6A-MDA5 that is immunoprecipitated after infection. Could this figure be improved by doing a fold change between input vs m6A-IP OR uninfected vs SCRV-infected conditions? This would reveal whether the modification of MDA5 with m6A is really increased after infection.

As shown in Figure 5F below, our data indicates that the proportion of m6A-modified MDA5 does indeed increase after SCRV infection, rather than solely due to the increased expression of MDA5 itself.

(2) Figure. 5E/F - The y-axis is unclear: relative MDA5 m6A levels. Relative to what? Input? Mock infected?

For experiments in Figure 5E/F, we first compared the m6A-IP group with the input group, and then normalized the control group (IgG group of 5E and Mock group of 5F) to “1”. We have replaced the Y-axis name with a clearer one (Figure 5E and 5F).

(3) General comment - It is not mentioned in the text that MDA5 is an interferon-stimulated gene. This would account for the increase in expression (qPCR) after viral infection or poly(I:C) transfection, hence there is no novelty in this finding. In addition, the authors suggest that MDA5 increases at the protein level (by immunoblot) but the increase on these blots is not convincing (figure 5H/5I).

We understand that the increase in expression of MDA5 as an interferon-stimulated gene after viral infection is a common phenomenon. We present this to further validate the m6A sequencing transcriptome data, and to demonstrate that although m6A modification interferes with MDA5 expression during viral infection, it cannot prevent the increase of mRNA level of MDA5. In addition, we rerun the experiment and the results showed that the expression of MDA5 protein can indeed be specifically activated by the SCRV virus and poly(I:C)-HMW.

Figure 6

(1) Figure 6E - What was the MOI of the virus used in this experiment? It is not mentioned in the figure legend.

MOI=5, we have added this point in the figure legend.

Figure 7

(1) Figure 7J - This graphic is somewhat misleading and should be altered to better reflect the conclusions that are drawn in the manuscript. The graphic suggests that MAVS and STING interact, but this is not demonstrated in the paper. In addition, the paper does not demonstrate whether MAVS or STING (or both) are needed downstream of MDA5 to relay signalling. Finally, please draw an arrow from type I IFNs to increased expression of MDA5 to illustrate that MDA5 is an ISG.

Thank you for the referee's suggestion. We have revised the images to more accurately match the conclusions of the manuscript (Figure 7J). Firstly, we have separated the STING protein from the MAVS protein. Secondly, arrows have been used to indicate that MDA5 is an IFN-stimulated gene. Finally, as we have added relevant experiments to demonstrate the importance of MITA protein in the signaling process of MDA5-activated IFN response. In addition, the function of MAVS binding to MDA5 protein and promoting its signal transduction is very conserved, and there is a good research background even in fish with RIG-I deficiency (10.1016/j.dci.2021.104235). Therefore, in Figure 7J, we still chose to bind MAVS to MDA5 protein and use it as a downstream signal transducer of MDA5.

Discussion<br /> (1) There is very little discussion about METTL and YTHDF proteins in the discussion despite the fact that the last 2 figures are entirely devoted to these proteins.

Based on the referee's suggestion, we have added relevant content about METTL and YTHDF proteins in the discussion. In addition, the basic mechanism and function of METTL and YTHDF proteins were briefly described in the introduction.

Reviewer #3 (Recommendations For The Authors):

Please refer to the specific suggestions and recommendations. They include proposals for experimental additions, improved methodologies, and suggestions to resolve writing-related concerns.

Major concerns

(1) I suggest changing the article title to "Functional Replacement of RIG-I with MDA5 in Fish Miiuy Croaker", or a similar title, to make it more focused and closely aligned with the content of the article.

Following the reviewer's recommendations, we have revised the title to emphasize our primary research subject is a teleost fish that lacks RIG-I. In addition, we have changed “5’ppp-RNA” to “5’ppp-RNA virus” to emphasize the interaction between the virus and the receptor. We believe that the revised title is more in line with the content of the article.

(2) Due to the inherent limitations in genome sequencing, assembly, and annotation for the Miiuy croaker, comprehensive annotation of immune-related genes remains incomplete. To address this critical gap, it is recommended that authors establish experimental protocols, such as Fluorescence In Situ Hybridization (FISH), to confirm the absence of RIG-I in the Miiuy croaker. They should simultaneously employ MDA5 probes as a positive control for validation purposes.

The miiuy croaker has good genomic information at the chromosomal level (doi: 10.1016/j.aaf.2021.06.001). In addition, studies have shown that RIG-I is absent in the orders of Perciformes (doi: 10.1016/j.fsirep.2021.100012), while miiuy croaker belongs to the order Perciformes, so it does indeed lose the RIG-I gene. Therefore, we do not intend to use FISH technology to prove this.

(3) Similarly, it is recommended that the authors first provide evidence of the presence of 5'ppp at the 5' terminus of the genome RNA of SCRV, as demonstrated in the study by Goubau et al. (doi: 10.1038/nature13590, Supplementary figure 1). This evidence is crucial before drawing conclusions about the compensatory role of MDA5 in recognizing 5'ppp RNA viruses, using SCRV as the viral model.

As suggested by the referee, we extracted SCRV RNA from SCRV virus particles and assessed the 5’-phosphate-dependence of stimulation by SCRV RNA. Calf intestinal phosphatase (CIAP) treatment substantially reduced the stimulatory activity of SCRV RNA in MKC cells of M. miiuy (Figure 1C and 1E). In addition, similar results were obtained by transfecting VSV-RNA isolated from VSV virus into DF-1 cells of G. gallus (Figure 1D). The above evidences confirm the presence of triphosphate molecular features between SCRV and VSV viruses, and indicating that birds and fish lacking RIG-I have other receptors that can recognize 5’ppp-RNA.

(4) The 62-nucleotide (nt) 5'ppp-RNA utilized in this study was obtained from Vesicular Stomatitis Virus (VSV). In order to provide direct evidence, it is necessary to include a 62-nt 5'ppp-RNA that is directly derived from SCRV itself.

We adopted this suggestion and synthesized a 67-nucleotide 5’ppp-SCRV RNA probe. We found that 5’ppp-SCRV activates dimerization of IRF3 and binds to MDA5 of M. miiuy in a 5’-triphosphate-dependent manner (Figure 4A-4F).

(5) Given that RNAs with uncapped diphosphate (PP) groups at the 5′ end also activate RIG-I, similar to RNAs with 5′-PPP moieties, and the 5′-terminal nucleotide must remain unmethylated at its 2′-O position to allow RNA recognition by RIG-I, it is necessary for the authors to conduct additional experiments to supplement and validate these two distinguishing features of RIG-I in RNA recognition. This will provide more reliable evidence for the replacement of RIG-I by MDA5 in RNA recognition.

Thank you for the reviewer's professional suggestions. We understand that exploring the combination of 5’pp-RNA and 2′-O-methylated RNA with MDA5 can further demonstrate the alternative function of MDA5. But we think that the use of 5’ppp-RNA and their dephosphorylation derivatives can fully demonstrate that the MDA5 of M. miiuy and G. gallus have evolved to recognize 5’triphosphate structure like human RIG-I. Therefore, we do not intend to conduct any additional experiments

(6) In section 2.3, the authors assert that Miiuy croaker recognizes SCRV through its RD domain. This claim is supported by their data showing that cells overexpressed with the MDA5 ΔRD mutant lost the ability to inhibit SCRV replication. As a result, the authors draw the conclusion that "these findings provide evidence that MDA5 may recognize 5'-triphosphate-dependent RNA (5'ppp-RNA) through its RD domain." However, to strengthen their argument, the authors should first demonstrate that during SCRV infection, MDA5-mediated antiviral immune response is indeed initiated by recognizing the 5'ppp part of the SCRV RNA, rather than the double-strand part (which can exist in ssRNA virus) of the viral RNA, as this is naturally a ligand for MDA5. Additionally, the authors should treat the isolated SCRV RNA with CIP to remove the phosphate group and examine the binding of MDA5 with SCRV RNA before and after treatment. They should also transfect CIP-treated or untreated SCRV RNA into MDA5 knockdown and wild-type MKC cells to investigate the induction of antiviral signaling and levels of viral replication. Finally, the authors should verify the binding ability of the mutants with isolated SCRV RNA, with or without CIP treatment, to determine which domain of MDA5 is responsible for SCRV 5'ppp-RNA recognition.

We understand the reviewer's concern that MDA5 may be identified by binding to dsRNA in the SCRV virus. Based on the reviewer's suggestion, we extracted SCRV RNA and obtained its dephosphorylated RNA using Calf intestinal phosphatase (CIAP). Next, we transfected them into MDA5-knockdown and wild-type MKC cells, and detected the dimerization of IRF3 and IFN reaction. The results indicate that SCRV RNA does indeed activate immunity in a triphosphate-dependent manner, and knockdown of MDA5 prevents immune activation of SCRV RNA (Figure 1C and 1E, Figure 2P and 2Q). Finally, we synthesized a 5'ppp-SCRV RNA probe and demonstrated that MDA5 binds to 5'ppp-SCRV through the RD domain (Figure 4E-4G). We believe that these results can better demonstrate that MDA5 recognizes 5’ppp-RNA through its RD domain and addresses the concerns of the reviewers.

(7) Similarly, merely presenting Co-IP data demonstrating the interaction between Miiuy croaker MDA5 and STING in overexpressed EPC cells does not justify the claim that "in vertebrates lacking RIG-I, MDA5 can utilize STING to facilitate signal transduction in the antiviral response". This is because interactions observed through overexpression may not accurately reflect the events occurring during viral infection or their actual antiviral functions. To provide more robust evidence, it is essential to conduct functional experiments after STING knockout (or at least knockdown). Furthermore, it is important to note that Miiuy Croaker alone cannot adequately represent all "vertebrates lacking RIG-I".

We found that co-expression of STING and MDA5 can enhance MDA5-mediated IFN-1 response during SCRV virus infection, while knocking down STING can restore MDA5-mediated IFN-1 response (Figure 2N and 2O). This indicates that STING plays an important signaling role in the immune response of MDA5 to RNA viruses. In addition, loss of RIG-I is a common phenomenon in vertebrates, and STING of birds such as chickens (doi: 10.4049/jimmunol.1500638) and mammalian tree shrews (doi: 10.1073/pnas.1604939113) can also bind to MDA5, indicating that STING can indeed play a crucial role in MDA5 signaling in species with RIG-I deficiency. We have added this section to our discussion and elaborated on our observations in more cautious language.

(8) In the manuscript, a series of experiments were conducted using an antibody (Beyotime Cat# AF7164) against endogenous MDA5. The corresponding immunogen for this MDA5 antibody is a recombinant fusion protein containing amino acids 1-205 of human IFIH1/MDA5 (NP_071451.2). However, the amino acid sequences of IFIH1/MDA5 differ substantially between humans and Miiuy croaker, which could introduce errors in the results. Therefore, it is essential to employ antibodies specifically designed for targeting Miiuy croaker's own MDA5 in the experiments.

As shown in Figure 2B, endogenous MDA5 antibodies can detect the MDA5 portion that is forcibly overexpressed by plasmids, suggesting that the MDA5 antibody can indeed specifically recognize the MDA5 protein of M. miiuy.

(9) It is recommended to investigate the phosphorylation of IRF3 in order to confirm the downstream signaling pathway during viral infection when MDA5 is knocked down or overexpressed.

Due to the lack of available phosphorylation antibodies for fish IRF3, we used IRF3 dimer experiments to detect downstream signaling (Figure 1C and 1D, Figure 2P, Figure 4C and 4J).

(10) The use of poly I:C as a mimic for dsRNA to investigate MDA5's recognition of 5'ppp-RNA in hosts lacking RIG-I, as well as the examination of the regulatory role of MDA5 m6A methylation upon activation by 5'ppp-RNA, may be inappropriate. Poly I:C does not possess 5'ppp, and while it has been identified as a ligand for MDA5 in various studies, MDA5 cannot serve as a substitute for RIG-I in recognizing poly (I:C). Therefore, the authors should utilize 5'ppp-dsRNA as the mimic and include the corresponding 5'ppp-dsRNA control without a 5'triphosphate as the negative control (both available from InvivoGen). This approach will specifically elucidate the mechanisms involved when MDA5 functions similarly to RIG-I in the recognition of 5'ppp-RNA.

In our study, we used poly(I:C)-HMW, a known dsRNA mimetic that can be preferentially recognized by MDA5 rather than RIG-I, as a positive control for activating MDA5. What we want to demonstrate is that, like poly(I:C)-HMW (positive control), SCRV can also promote MDA5-mediated IFN immunity, further indicating the important role of MDA5 in 5’ppp-RNA virus invasion. We have clearly labeled the type of poly(I:C) in the figures and legends to avoid misunderstandings for readers.

(11) In Figure 2, Figure 3, and Figure 6, the appearance of virus plaques is not readily apparent, and it is necessary to replace these images with clearer photographs. It appears that MKC or MPC cells are not appropriate for conducting plaque assays. To accurately assess viral proliferation, the authors should measure key indicators throughout the process, such as the production of positive-strand RNAs (+RNAs), replication intermediates (RF), and transcription of subgenomic RNAs. This approach is preferable to solely measuring the M and G protein genes from the virus genome as positive results can still be observed in contaminated cells.

As pointed out by the reviewer, we also think that the virus plaque images in Figure 2K and Figure 3D are not clear enough, so we have replaced them with new clear images (Figure 2J and Figure 3D). But we think that other images can clearly display the proliferation of the SCRV virus, so we did not replace them. In addition, the primers we currently use do measure +RNA, so the replication level of the SCRV virus can be accurately evaluated without being affected by virus contamination. Because the regions where the two pairs of primers are located belong to the SCRV-M and SCRV-G protein genes, we label them as SCRV-M and SCRV-G to distinguish between the two pairs of genes. To avoid reader misunderstanding, we have modified the Y-axis label in the figures (Figure 2I and 2K, Figure 3E, Figure 6E and 6O).

(12) There is a substantial disparity in the molecular size of M. miiuy MDA5 between endogenous and exogenously expressed proteins, as shown in Figure 2A and 2C-D. Please provide clarification.

Please refer to the response to Reviewer 2's question regarding Figure 4B above.

(13) The manuscript incorporates the evolutionary perspective, but lacks specific evolutionary analysis. Thus, it is essential to include relevant analysis to comprehend the evolutionary dynamics and positive selection on MDA5 and LGP2 in the absence of RIG-I in Miiuy croaker. This can be achieved through theoretical calculations using appropriate algorithms, such as the branch models and branch-site models based on the maximum-likelihood method implemented in the phylogenetic analysis by maximum likelihood (PAML) package.

In fact, we have analyzed the molecular evolution of MDA5 and LGP2. Unfortunately, even when analyzing only the MDA5/LGP2 CDS sequences in fish, we found that the topologies of gene trees of MDA5/LGP2 were largely consistent with the species tree. Thus, species with or without RIG-I in the gene trees cannot effectively separate clusters, making it extremely difficult to analyze the molecular evolution of MDA5/LGP2 caused by RIG-I deficiency. Consequently, we gave up this aspect of analysis.

(14) If the narrative regarding m6A methylation goes beyond the activation of MDA5 through recognition of 5'ppp-RNA and represents a regulatory mechanism for all MDA5 activation events, it is not relevant to the theme of "An arms race under RIG-I loss: 5'ppp-RNA and its alternative recognition receptor MDA5." Therefore, all investigations in this paper should focus solely on events when MDA5 recognizes 5'ppp-RNA. Any data associated with the broader regulatory mechanisms and m6A methylation of MDA5 should be excluded from this manuscript and instead be included in a separate study dedicated to exploring this specific topic.

Our theme aims to showcase RNA viruses, rather than an interaction between 5'ppp-RNA and host virus receptors, which our current topic cannot accurately express. Therefore, we made two main changes: firstly, we limited the study species to M. miiuy, although some studies on the functional substitution of MDA5 for RIG-I involved birds. Secondly, change “5’ppp-RNA” to “5’ppp-RNA virus”. We believe that the revised title is more in line with our current research contents.

(15) The running title appears to be hastily done.

We modified it to “MDA5 recognizes 5’ppp-RNA virus in species lacking RIG-I”.

(16) There are many descriptions that are not strongly related to the main theme of the article in the introduction section, making it lengthy and fragmented. Please focus on the research background of RIG-I and MDA5, including their structures, functions, and regulatory mechanisms, as well as the research progress on the compensatory effect of MDA5 in the absence of RIG-I and its evolutionary adaptation mechanism in other species.

Based on the suggestions of the reviewers, we have removed some of the less relevant content in the introduction and added research progress on the compensatory effect of MDA5 in the evolutionary adaptation mechanism of tree shrews in the absence of RIG-I.

(17) Lines 149-156 in the "Results" section include content that resembles an "Introduction" It is important to avoid duplicating information in the results section. Therefore, the authors are encouraged to revise this paragraph to ensure conciseness in the article.

We have streamlined this section to enhance the article's conciseness and clarity.

(18) In the "Results" section, at line 177, the authors assert, "As depicted in Figure 1F-1H," which should be corrected to Figure 2F-2H. Furthermore, the y-axis of the two figures on the right-hand side of Figure 2H represents the ISG15 genes. At line 182, "as demonstrated in Figure 1I-1L," should be revised as "as illustrated in Figure 2I-2L". The authors demonstrated a lack of attention to detail.

Thank you to the reviewer for pointing out our errors, and we have made the necessary corrections.

(19) In lines 197-198, the authors stated that "MDA5-ΔRD showed an inability to interact with SCRV." However, Figure 3D did not reveal any significant difference, thus it is advisable to repeat this experiment at least once.

We have replaced this virus spot image with a new one (Figure 3D).

(20) In lines 200-201 of the "2.3 RD domain is required for MDA5 to recognize SCRV" section, the authors report that the expression of antiviral genes was induced by the overexpression of both MDA5 and MDA5-ΔRD, even in the absence of infection (Figure 3F). Why does the expression of antiviral genes increase in the absence of viral RNA stimulation? Please provide a reasonable explanation.

In the absence of viral infection, overexpression of viral receptor proteins may still transmit erroneous signaling, affecting the body's immunity. We speculate that due to the preservation of the CARD domain by MDA5 and MDA5-ΔRD, they can still induce the expression of antiviral factors without ligands, although this induction effect is much smaller than that of viral infection. However, in order to better demonstrate the function of the RD domain of MDA5 in M. miiuy, we have changed the experimental plan, as shown in the figure 3F. We detected the induction of antiviral factors by overexpression of MDA5 and MDA5-△RD under poly (I:C)-HMW stimulation. This can indicate that the RD domain has a conserved function in the recognition of poly(I:C)-HMW in M. miiuy, and can serve as a positive control for the recognition of SCRV virus invasion by the RD domain of MDA5.

(21) Please provide the GeneBank accession number of M. miiuy MDA5.

The GeneBank accession number of M. miiuy MDA5 was added in the section 4.5 plasmids construction.

(22) The content of lines 228-233 in the "Results" section bears resemblance to that of the "Introduction." To ensure the avoidance of information duplication, it is recommended to remove this paragraph from the results section.

This section has been streamlined.

(23) The bands of mmiMDA5 in the 5'ppp-RNA and dsRNA lanes in Figure 4B are weak and almost unobservable. Please replace them with clear images.

We have rerun this experiment and replaced the images (Figure 4B).

(24) In Figure 5G and at line 253, there are only results presented for the SCRV infection group, while no results are shown for the control group. This raises the question of why the control group results are missing. It is necessary to provide a reasonable explanation or correction for this issue.

The "0 h" infection time point of the SCRV virus is the control group, and we have replaced it with a more intuitive image (Figure 5G).

(25) In Figure 7C, it would be necessary to include the western blot result of YTHDF protein expression in order to verify the efficiency of YTHDF siRNA.

In fact, we have attempted to detect the endogenous expression of YTHDF protein using available commercial antibodies. Unfortunately, only the YTHDF2 antibody can specifically recognize the endogenous protein expression of YTHDF2 in M. miiuy. In addition, the knockdown effect of si-YTHDF2 has been validated by YTHDF2 antibody (doi: 10.4049/jimmunol.2200618).

(26) In line 422 of the "4.3 Cell culture and treatment" section, the paragraph raises a question regarding the nature of Miiuy croaker kidney cells (MKCs) and spleen cells (MPCs) - whether they are cell lines or freshly isolated cells (or primary cultures) derived from kidney and spleen tissues. If these cells are indeed cell lines, it is requested to provide detailed information about the sources and properties of the cells (such as whether they are epithelial cells or other mixed cell types) and the generations of propagation. Alternatively, if the cells were freshly isolated or primary cultures obtained from fish, the method for cell isolation should be provided. The source and stability of cells are extremely important for ensuring the repeatability and reliability of experimental outcomes.

M. miiuy kidney cells (MKCs) and spleen cells (MPCs) are cell lines derived from the kidney and spleen tissues of M. miiuy, with passages ranging from 20 to 40 times. These details have been incorporated into section 4.3.

(27) There are many inaccurate descriptions in the text, which employ concepts that are too broad. These descriptions need to be narrowed down to specific species or objects. Here are a few examples, along with the necessary revisions. Other similar instances should also be revised accordingly. For instance, in line 119, "fish MDA5" should be changed to "Miiuy croaker MDA5." Similarly, in line 166, "fish MDA5-mediated signaling pathway" should be changed to "Miiuy croaker MDA5-mediated signaling pathway." In line 174, "fish MDA5" should be revised to "Miiuy croaker MDA5." Additionally, in line 185, "antiviral responses of teleost" should be changed to "antiviral responses of Miiuy croaker." In line 197, "interact with SCRV" should be revised to "interact with 5'ppp-RNA of SCRV." In line 337, "loss of RIG-I in the vertebrate" should be modified to "loss of RIG-I in Miiuy croaker and chicken." Similarly, in line 338, "MDA5 of fish" should be changed to "MDA5 of Miiuy croaker." Lastly, in line 348, "RIG-I deficient vertebrates" should be revised to "RIG-I deficient Miichthys miiuy and Gallus gallus."

Thank you for the reviewer's suggestions. We have made revisions to these inaccurate descriptions and reviewed the entire manuscript to address similar statements with broad concepts.

(28) Finally, it should be noted that a similar discovery has already been reported in tree shrews (Ling Xu, et al., Proc Natl Acad Sci., 2016, 113(39):10950-10955). This article shares similarities with that research report, therefore it is necessary to discuss in detail the relationship between the two in the discussion and compare and analyze the evolutionary patterns of MDA5 from it.

Based on the reviewer's suggestions, we have compared the similarities and differences between these two reports during the discussion and analyzed the evolutionary dynamics of MDA5 in these vertebrates lacking RIG-I.

Minor concerns:

Thank you to the reviewer for their meticulous examination to our manuscript, we have made revisions to the following suggestions.

(1) At line 120, the sentence "SCRV(one 5'ppp-RNA virus)" should have a space between "SCRV" and "(one 5'ppp-RNA virus)". Please make this correction.

Corrected.

(2) At lines 147-148, the sentence "However, the downstream gene of TOPORSa is missing a RIG-I" is not accurate and needs modification.

We have modified this sentence.

(3) At line 184, "findings indicate" should be corrected to "findings indicated".

Corrected.

(4) At line 189, "a 5'ppp-RNA virus" should be deleted and the text seems redundant.

Deleted.

(5) At line 198, "replication. (Figure 3C-3E)", please remove the punctuation between "replication" and "(Figure 3C-3E)".

Corrected.

(6) At line 416 in "Materials and methods" section, "4.2 Sample and challenge" should be corrected to "4.2 Fish and challenge".

Corrected.

(7) At line 419, the authors state that "The experimental procedure for SCRV infection was performed as described", please briefly describe the SCRV infection method and the infectious dose.

Based on the reviewer's suggestions, we have added relevant descriptions of SCRV infection in section 4.2.

(8) There are several formatting issues in the "Materials and Methods" section. For instance, in line 424, there is no space between the number and letter in "100 μg/ml" and "26 ℃" should be corrected to "26℃". Additionally, in line 430, "Cells" should be corrected to "cells".

Corrected.

(9) At line 446, "50 ng/ul" and "100 mU/ul" should be corrected to "50 ng/μl" and "100 mU/μl".

Corrected.

(10) At line 459, "primers 1)" should be corrected to "primers".

Corrected.

(11) At lines 461-464, the description "For protein purification, MDA5 plasmids with 6× His tag was constructed based on pcDNA3" seems to be no direct logical connection between protein purification and the plasmid construction. Please make the necessary corrections.

Corrected.

(12) At line 548, "cytoplasmic" should be corrected to "Cytoplasmic".

Corrected.

(13) At line 549, "5× 107" should be corrected to "5 × 107".

Corrected.

(14) At line 557, "MgCl2" should be corrected to "MgCl2".

Corrected.

(15) At line 558, "6 %" should be corrected to "6%".

Corrected.

(16) At line 565, "50μg" should be corrected to "50 μg".

Corrected.

(17) At line 571, "300{plus minus}50 bp." should be corrected to "300 {plus minus} 50 bp."

Corrected.

(18) At lines 592-593, the sentence "After several incubations, the m6A level was quantified colorimetrically at a wavelength of 450 nm" does not read smoothly, please improve it.

Revised.

(19) At line 786, "MDA5 recognize" should be corrected to "MDA5 recognized".

Corrected.

(20) At lines 788 and 798, "Pulldown" should be corrected to "Pull-down".

Corrected.

(21) At lines 790 and 796, "bluestaining" should be corrected to "blue staining".

Deleted.

(22) At line 825, "SCRV and infection" should be corrected to "SCRV infection".

Corrected.

(23) At lines 826-827, "SCRV (H) and poly(I:C) (I) infection" should be corrected to "SCRV infection (H) and poly(I:C) stimulation (I)".

Corrected.

eLife assessment

The authors provide compelling evidence that MSP1 inhibition (leading to chromosomal instability or CIN in the cancer cells) increases phagocytosis and that tumors with CIN respond better to macrophage therapeutics. In this important study, they demonstrate particularly impressive survival rates for mouse models of CIN B16 tumors treated with adoptively transferred macrophages, CD47-SIRPα blockade, and anti-Tyrp1 IgG.

eLife assessment

This article presents important results describing how the gathering, integration, and broadcasting of information in the brain changes when consciousness is lost either through anesthesia or injury. They provide convincing evidence to support their conclusions, although the paper relies on a single analysis tool (partial information decomposition) and could benefit from a clearer explication of its conceptual basis, methodology, and results. The work will be of interest to both neuroscientists and clinicians interested in basic and clinical aspects of consciousness

eLife assessment

This article presents important results describing how the gathering, integration, and broadcasting of information in the brain changes when consciousness is lost either through anesthesia or injury. They provide convincing evidence to support their conclusions, although the paper relies on a single analysis tool (partial information decomposition) and could benefit from a clearer explication of its conceptual basis, methodology, and results. The work will be of interest to both neuroscientists and clinicians interested in basic and clinical aspects of consciousness.

Author response:

We thank the reviewers for their help and their suggestions to make this manuscript more rigorous. We would like to post provisional author responses when eLife publish the reviewed preprint, and the more detailed responses will be supplemented with the revised manuscript.

• There are questions about choices made in the computational approach (architecture and type of generative model, training set).

We will train a new generator model based on the current GAN architecture, but with ‘hybrid’ AMP/AVP training sets (Reviewer 1 and 3). Hence, we can directly compare the performances of two generators. Based on our preliminary data, providing GAN with more AVP sequences during training helped the designed peptides pass the AVP filter, at the cost of reducing the average AMPredicgtor scores. The new generator also elevated the diversity of designed sequences.

We also perturbed the detailed architecture of our deep learning models, including fully-connected graph edge encodings and different versions of ESM (e.g. esm1b_t33_650M_UR50S, esm2_t48_15B_UR50D, Reviewer 2). In the revised manuscript, we will report the effects of these modifications and suggest the overall construct of GCN and GAN are suitable for a light-weight sequence label model, as demonstrated in Author response table 1 and 2. For the generator, we suggest that using our approach, we may have reached a plateau for the GAN sampling (Author response table 3).

Author response table 1.

Results of AMPredictor with different graph edge encodings

Author response table 2.

Results of AMPredictor with different ESM versions

Author response table 3.

Evaluation of generated sequences with different sampling numbers

• There is an important concern about the small number of antimicrobial peptides tested, compared to other studies, and the origin of antiviral activities.

We will address this concern by increasing the number of peptides tested in anti-microbial and anti-viral experiments. As reported in current version of our manuscript, the first generation of GAN generated 128 unique designs and the top 2% (3 designs) was tested experimentally. The second generation of GAN will produce ~1024 designs (1-2 weeks) and the top 2% (~ 20 new sequences) will be tested. We are in the process of synthesize (2-3 weeks) and MIC measurement (1 week). The overall size of tested sample will reach 20-30 sequences. We will focus on sequences with low similarity (< 30%) to any known AMPs, thus expanding the universe functional peptides. We estimated the collection of these new data in 6 weeks.

eLife assessment

This study presents a useful pipeline for de novo design of antimicrobial peptides active both against bacteria and viruses. The method is based on deep learning, using a GAN generator and a regression tasked to predict antimicrobial activity. The evidence supporting the conclusions is promising but incomplete: three generated peptides are studied experimentally in vitro, and one is then tested in vivo in mice; the comparisons to other design methods could also be strengthened. This work will be of interest to the community working on machine learning for biomedical applications and specifically on antimicrobial peptides.

Reviewer #1 (Public Review):

This manuscript presents a pipeline incorporating a deep generative model and peptide property predictors for the de novo design of peptide sequences with dual antimicrobial/antiviral functions. The authors synthesized and experimentally validated three peptides designed by the pipeline, demonstrating antimicrobial and antiviral activities, with one leading peptide exhibiting antimicrobial efficacy in animal models. However, the manuscript as it stands, has several major limitations on the computational side.

Major issues:

(1) The choice of GAN as the generative model. There are multiple deep generative frameworks (e.g., language models, VAEs, and diffusion models), and GANs are known for their training difficulty and mode collapse. Could the authors elaborate on the specific rationale behind choosing GANs for this task?

(2) The pipeline is supposed to generate peptides showing dual properties. Why were antiviral peptides not used to train the GAN? Would adding antiviral peptides into the training lead to a higher chance of getting antiviral generations?

(3) For the antimicrobial peptide predictor, where were the contact maps of peptides sourced from?

(4) Morgan fingerprint can be used to generate amino acid features. Would it be better to concatenate ESM features with amino acid-level fingerprints and use them as node features of GNN?

(5) Although the number of labeled antiviral peptides may be limited, the input features (ESM embeddings) should be predictive enough when coupled with shallow neural networks. Have the authors tried simple GNNs on antiviral prediction and compared the prediction performance to those of existing tools?

(6) Instead of using global alignment to get match scores, the authors should use local alignment.

(7) How novel are the validated peptides? The authors should run a sequence alignment to get the most similar known AMP for each validated peptide, and analyze whether they are similar.

(8) Only three peptides were synthesized and experimentally validated. This is too few and unacceptable in this field currently. The standard is to synthesize and characterize several dozens of peptides at the very least to have a robust study.

Reviewer #2 (Public Review):

Summary:

This study marks a noteworthy advance in the targeted design of AMPs, leveraging a pioneering deep-learning framework to generate potent bifunctional peptides with specificity against both bacteria and viruses. The introduction of a GAN for generation and a GCN-based AMPredictor for MIC predictions is methodologically robust and a major stride in computational biology. Experimental validation in vitro and in animal models, notably with the highly potent P076 against a multidrug-resistant bacterium and P002's broad-spectrum viral inhibition, underpins the strength of their evidence. The findings are significant, showcasing not just promising therapeutic candidates, but also demonstrating a replicable means to rapidly develop new antimicrobials against the threat of drug-resistant pathogens.

Strengths:

The de novo AMP design framework combines a generative adversarial network (GAN) with an AMP predictor (AMPredictor), which is a novel approach in the field. The integration of deep generative models and graph-encoding activity regressors for discovering bifunctional AMPs is cutting-edge and addresses the need for new antimicrobial agents against drug-resistant pathogens. The in vitro and in vivo experimental validations of the AMPs provide strong evidence to support the computational predictions. The successful inhibition of a spectrum of pathogens in vitro and in animal models gives credibility to the claims. The discovery of effective peptides, such as P076, which demonstrates potent bactericidal activity against multidrug-resistant A. baumannii with low cytotoxicity, is noteworthy. This could have far-reaching implications for addressing antibiotic resistance. The demonstrated activity of the peptides against both bacterial and viral pathogens suggests that the discovered AMPs have a wide therapeutic potential and could be effective against a range of pathogens.

Reviewer #3 (Public Review):

Summary:

Dong et al. described a deep learning-based framework of antimicrobial (AMP) generator and regressor to design and rank de novo antimicrobial peptides (AMPs). For generated AMPs, they predicted their minimum inhibitory concentration (MIC) using a model that combines the Morgan fingerprint, contact map, and ESM language model. For their selected AMPs based on predicted MIC, they also use a combination of antiviral peptide (AVP) prediction models to select AMPs with potential antiviral activity. They experimentally validated 3 candidates for antimicrobial activity against S. aureus, A. baumannii, E. coli, and P. aeruginosa, and their toxicity on mouse blood and three human cell lines. The authors select their most promising AMP (P076) for in vivo experiments in A. baumannii-infected mice. They finally test the antiviral activity of their 3 AMPs against viruses.

Strengths:

-The development of de novo antimicrobial peptides (AMPs) with the novelty of being bifunctional (antimicrobial and antiviral activity).

-Novel, combined approach to AMP activity prediction from their amino acid sequence.

Weaknesses:

-I missed justification on why training AMPs without information of their antiviral activity would generate AMPs that could also have antiviral activity with such high frequency (32 out of 104).

-The justification for AMP predictor advantages over previous tools lacks rationale, comparison with previous tools (e.g., with the very successful AMP prediction approach described by Ma et al. 10.1038/s41587-022-01226-0), and proper referencing.

-Experimental validation of three de novo AMPs is a very low number compared to recent similar studies.

-I have concerns regarding the in vivo experiments including i) the short period of reported survival compared to recent studies (0.1038/s41587-022-01226-0, 10.1016/j.chom.2023.07.001, 0.1038/s41551-022-00991-2) and ii) although in Figure 2 f and g statistics have been provided, log scale y-axis would provide a better comparative representation of different conditions.

-I had difficulty reading the story because of the use of acronyms without referring to their full name for the first time, and incomplete annotation in figures and captions.

eLife assessment

This fundamental study addresses discrepancies in determining bacterial burden in osteomyelitis as determined by culture and enumeration using DNA. The authors present compelling data demonstrating the emergence of discrepancies between CFU counts and genome copy numbers detected by PCR in Staphylococcus aureus strains infecting osteocyte-like cells. The observations represent a substantial addition to the field of musculoskeletal infection, with possible broad applicability and clinical benefit to other infectious diseases.

Reviewer #1 (Public Review):

Summary:

This work shows, based on basic laboratory investigations of in vitro grown bacteria as well as human bone samples, that conventional bacterial culture can substantially underrepresent the quantity of bacteria in infected tissues. This has often been mentioned in the literature, however, relatively limited data has been provided to date. This manuscript compares culture to a digital droplet PCR approach, which consistently showed greater levels of bacteria across the experiments (and for two different strains).

Strengths:

Consistency of findings across in vitro experiments and clinical biopsies. There are real-world clinical implications for the findings of this study.

Weaknesses:<br /> No major weaknesses. Only 3 human samples were analyzed, although the results are compelling.

Reviewer #2 (Public Review):

In this study, the authors address discrepancies in determining the local bacterial burden in osteomyelitis between that determined by culture and enumeration by DNA-directed assay. Discrepancies between culture and other means of bacterial enumeration are long established and highlighted by Staley and Konopka's classic, "The great plate count anomaly" (1985). Here, the authors first present data demonstrating the emergence of discrepancies between CFU counts and genome copy numbers detected by PCR in S. aureus strains infecting osteocyte-like cells. They go on to demonstrate PCR evidence that S. aureus can be detected in bone samples from sites meeting a widely accepted clinico-pathological definition of osteomyelitis. They conclude their approach offers advantages in quantifying intracellular bacterial load in their in vitro "co-culture" system.

WEAKNESSES

(A) My main concern here is the significance of these results outside the model osteocyte system used by this group. Although they carefully avoid over-interpreting their results, there is a strong undercurrent suggesting their approach could enhance aetiologic diagnosis in osteomyelitis and that enumeration of the infecting pathogen might have clinical value. In the first place molecular diagnostics such as 16S rDNA-directed PCR are well established in identifying pathogens that don't grow. Secondly, it is hard to see how enumeration could have value beyond in vitro and animal model studies since serial samples will rarely be available from clinical cases.

(B) I have further concerns regarding interpretation of the combined bacterial and host cell-directed PCRs against the CFU results. Significance is attached to the relatively sustained genome counts against CFU declines. On the one hand it must be clearly recognised that detection of bacterial genomes does not equate to viable bacterial cells with potential for further replication or production of pathogenic factors. Of equal importance is the potential contribution of extracellular DNA from lysed bacteria and host cells to these results. The authors must clarify what steps, if any, they have taken to eliminate such contributions for both bacteria and host cells. Even the treatment with lysotaphin may have coated their osteocyte cultures with bacterial DNA, contributing downstream to the ddPCR results presented.

STRENGTHS

(C) On the positive side, the authors provide clear evidence for the value of the direct buffer extraction system they used as well as confirming the utility of ddPCR for quantification. In addition, the successful application of MinION technology to sequence the EF-Tu amplicons from clinical samples is of interest.

(D) Moreover, the phenomenology of the infection studies indicating greater DNA than CFU persistence and differences between the strains and the different MOI inoculations are interesting and well-described, although I have concerns regarding interpretation.

Author response:

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

Reviewer #1 (Public Review):

Summary:

This work shows, based on basic laboratory investigations of invitro-grown bacteria as well as human bone samples, that conventional bacterial culture can substantially underrepresent the quantity of bacteria in infected tissues. This has often been mentioned in the literature, however, relatively limited data has been provided to date. This manuscript compares culture to a digital droplet PCR approach, which consistently showed greater levels of bacteria across the experiments (and for two different strains).

Strengths:

Consistency of findings across in vitro experiments and clinical biopsies. There are real-world clinical implications for the findings of this study.

Weaknesses:

No major weaknesses. Only three human samples were analyzed, although the results are compelling.

We only put in three examples of clinical diagnosis to showcase the application of this method particularly to osteomyelitis. For further validation, larger cohort studies are required, which are currently underway.

Reviewer #2 (Public Review):

In this study, the authors address discrepancies in determining the local bacterial burden in osteomyelitis between that determined by culture and enumeration by DNA-directed assay. Discrepancies between culture and other means of bacterial enumeration are long established and highlighted by Staley and Konopka's classic, "The great plate count anomaly" (1985). Here, the authors first present data demonstrating the emergence of discrepancies between CFU counts and genome copy numbers detected by PCR in S. aureus strains infecting osteocyte-like cells. They go on to demonstrate PCR evidence that S. aureus can be detected in bone samples from sites meeting a widely accepted clinicopathological definition of osteomyelitis. They conclude their approach offers advantages in quantifying intracellular bacterial load in their in vitro "co-culture" system.

The publication related to “The great plate count anomaly (1985)” has been added to revised version as new reference #2.

Weaknesses

- My main concern here is the significance of these results outside the model osteocyte system used by this group. Although they carefully avoid over-interpreting their results, there is a strong undercurrent suggesting their approach could enhance aetiologic diagnosis in osteomyelitis and that enumeration of the infecting pathogen might have clinical value. In the first place, molecular diagnostics such as 16S rDNA-directed PCR are well established in identifying pathogens that don't grow. Secondly, it is hard to see how enumeration could have value beyond in vitro and animal model studies since serial samples will rarely be available from clinical cases.

Indeed, we initiated this study for the purpose of trying to improve the diagnostic outcomes for osteomyelitis, in particular that associated with prosthetic joint infection (PJI) but also all other forms, as the current gold-standard diagnostic approaches for this type of infection, either bacterial culture or whole genome sequencing, are very time consuming and costly, and yet are not necessarily accurate. Our method has the benefits (not limited to) of achieving absolute quantification of bacterial load in a shortened time period (in the order of hours) in clinical bone specimens from infected patients. Many of the identified bacterial species in patients were not able to be diagnosed by standard bacterial culturing. Moreover, one of the problematic features of treating bone infection is that repetitive surgeries are usually needed, particularly in PJI, hence, serial clinical bone specimens from the same patient are in fact often available. Therefore, our method of being able to quantify bacterial load offers the advantage of monitoring the infected status throughout the treatment journey. In this study, we chose the tuf gene as the targeting sequence to amplify the bacterial signal instead of the well-established 16S PCR for the reason that tuf provides much better sequence discrimination between bacterial species. Therefore, the short PCR amplicon of just 271 bp used in our study, is able to give us a highly accurate taxonomic readout. By this approach, we again shorten the time required for diagnosis. In the last paragraph of the Discussion in the revised manuscript, extra text, a figure demonstrating the strong sequence diversity in tuf (Supplementary Figure 2) and an additional reference have been added to address the Reviewer’s concerns.

- I have further concerns regarding the interpretation of the combined bacterial and host cell-directed PCRs against the CFU results. Significance is attached to the relatively sustained genome counts against CFU declines. On the one hand, it must be clearly recognised that the detection of bacterial genomes does not equate to viable bacterial cells with the potential for further replication or production of pathogenic factors. Of equal importance is the potential contribution of extracellular DNA from lysed bacteria and host cells to these results. The authors must clarify what steps, if any, they have taken to eliminate such contributions for both bacteria and host cells. Even the treatment with lysotaphin may have coated their osteocyte cultures with bacterial DNA, contributing downstream to the ddPCR results presented.

We agree that concerns around the interpretation of any molecular readout need to be taken into account. We have yet to find a method that can definitively identify bacterial viability in a clinical setting in the absence of culture. However, PJI and osteomyelitis in general is characterised by a high percentage of culture-negative infection cases, calling for such molecular approaches. Commercially available, so called “live/dead” bacterial PCR reagents exist that act as PCR signal inhibitors by penetrating the cell wall of compromised cells to prevent the PCR signal being generated from those cells. In our experience, while these can provide a certain level of added scrutiny in an experimental setting, they are not definitive because the reaction is often incomplete in an idealised situation and also the reagent may cancel signal from viable bacteria growing under conditions of stress, such as during antimicrobial treatment and host-derived stress imparted in intracellular or intra-tissue environments. Indeed, such stresses are likely contributors to clinical non-culturability. Whole genome sequencing would provide more certainty of bacterial viability to demonstrate genomic intactness but as we discuss herein, this a lengthy and costly process, and one which may prove difficult from host tissue with a low pathogen load. It should be noted that the significance of any diagnostic readout, including from culture, WGS or our method reported here would need to be interpreted by the treating clinical team. We would argue that a rapid, practical molecular diagnostic method in the absence or even presence of culture would provide treating clinicians with an improved rationale for tailoring antimicrobial treatments. 

Strengths

- On the positive side, the authors provide clear evidence for the value of the direct buffer extraction system they used as well as confirming the utility of ddPCR for quantification. In addition, the successful application of MinION technology to sequence the EF-Tu amplicons from clinical samples is of interest.

- Moreover, the phenomenology of the infection studies indicating greater DNA than CFU persistence and differences between the strains and the different MOI inoculations are interesting and well-described, although I have concerns regarding interpretation.

eLife assessment

This manuscript by Vuong and colleagues reports on the kinetics of viremia in a large set of individuals from Vietnam. In the large cohort, all 4 dengue serotypes are represented and the authors try to correlate viraemia measured at various days from illness onset with thrombocytopaenia and severe dengue, according to the WHO 2009 classification scheme. These are fundamental findings that provide compelling evidence of the importance of measuring viremia early in the phase of the disease. These data will help to inform the design of studies of antiviral drugs against dengue.

Reviewer #1 (Public Review):

Summary:

This manuscript by Vuong and colleagues reports a study that pooled data from 3 separate longitudinal study that collectively spanned an observation period of over 15 years. The authors examined for correlation between viraemia measured at various days from illness onset with thrombocytopaenia and severe dengue, according to the WHO 2009 classification scheme. The motivation for this study is both to support the use of viraemia measurement as a prognostic indicator of dengue and also to, when an antiviral drug becomes licensed for use, guide the selection of patients for antiviral therapy. They found that the four DENVs show differences in peak and duration of viraemia and that viraemia levels before day 5 but not those after from illness onset correlated with platelet count and plasma leakage at day 7 onwards. They concluded that the viraemia kinetics call for early measurement of viraemia levels in the early febrile phase of illness.

Strengths:

This is a unique study due to the large sample size and longitudinal viraemia measurements in the study subjects. The data addresses a gap in information in the literature, where although it has been widely indicated that viraemia levels are useful when collected early in the course of illness, this is the first time anyone has systematically examined this notion. The inclusion of correlation between rate of viraemia decline and risk of severe dengue/plasma leakage further strengthens the relevance of this paper to those interested in anti-dengue therapeutic research and development.

Weaknesses:

The study only analysed data from dengue patients in Vietnam. Moreover, the majority of these patients had DENV-1 infection; few had DENV-4 infection. The data could thus be skewed by the imbalance in the prevalence of the different types of DENV during the period of observation. The use of patient-reported time of symptom onset as a reference point for viraemia measurement is pragmatic although there is subjectivity and thus noise in the data.

Reviewer #2 (Public Review):

Summary:

The authors have carried out a comprehensive analysis regarding the kinetics of viraemia and clinical disease severity.

Strengths:

The manuscript provides important information, especially regarding the time of clearance of the virus and disease severity.

Weaknesses:

Due to the lower number of patients with primary dengue, cannot get an idea regarding viraemia kinetics and disease severity for different serotypes during primary infection.

Author response:

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

Reviewer #1 (Public Review):

Summary:

This manuscript by Vuong and colleagues reports a study that pooled data from 3 separate longitudinal studies that collectively spanned an observation period of over 15 years. The authors examined for correlation between viraemia measured at various days from illness onset with thrombocytopaenia and severe dengue, according to the WHO 2009 classification scheme. The motivation for this study is both to support the use of viraemia measurement as a prognostic indicator of dengue and also when an antiviral drug becomes licensed for use, to guide the selection of patients for antiviral therapy. They found that the four DENVs show differences in peak and duration of viraemia and that viraemia levels before day 5 but not those after from illness onset correlated with platelet count and plasma leakage at day 7 onwards. They concluded that the viraemia kinetics call for early measurement of viraemia levels in the early febrile phase of illness.

Strengths:

This is a unique study due to the large sample size and longitudinal viraemia measurements in the study subjects. The data addresses a gap in information in the literature, where although it has been widely indicated that viraemia levels are useful when collected early in the course of illness, this is the first time anyone has systematically examined this notion.

Weaknesses:

The study only analysed data from dengue patients in Vietnam. Moreover, the majority of these patients had DENV-1 infection; few had DENV-4 infection. The data could thus be skewed by the imbalance in the prevalence of the different types of DENV during the period of observation. The use of patient-reported time of symptom onset as a reference point for viraemia measurement is pragmatic although there is subjectivity and thus noise in the data.

We acknowledge and appreciate your comments regarding the limitations of our study, including the pooled data from Vietnam and the use of symptom onset as a reference point for viremia kinetics. These points have been incorporated into the “Limitations” section.

Reviewer #2 (Public Review):

Summary:

This manuscript highlights very important findings in the field, especially in designing clinical trials for the evaluation of antivirals.

Strengths:

The study shows significant differences between the kinetics of viral loads between serotypes, which is very interesting and should be taken into account when designing trials for antivirals.

Weaknesses:

The kinetics of the viral loads based on disease severity throughout the illness are not described, and it would be important if this could be analyzed.

In response to your suggestion, we have expanded our analysis to investigate the relationship between the rate of viremia decline and clinical outcomes. Our findings demonstrate that a faster rate of viremia decline is associated with a reduced risk of severe clinical outcomes. We have incorporated this new analysis into the revised manuscript, providing further details in the “Statistical Analysis” section (page 7) and presenting the results on pages 15 and in Figure 6.

Reviewer #1 (Recommendations For The Authors):

Several areas require additional attention. I have limited my comments on the findings as I am not a mathematician and cannot knowledgeably comment on the statistical modelling methods.

Comment #1: Lines 83-84. Although viraemia level shows declining trends from illness onset and thus lessens its prognostic value, it remains unknown if a more rapid rate of decline in viraemia is associated with a reduced risk of severe dengue. This is the fundamental premise of antiviral drug development for the treatment of dengue. The authors are uniquely poised to show if this logic that underpins antiviral development is likely correct and perhaps even estimate the extent to which a decline in viraemia needs to occur for a measurable reduction in the risk of severe dengue. Could the authors consider such an analysis?

We appreciate your valuable suggestion. In response, we have expanded our analysis to investigate the relationship between the rate of viremia decline and clinical outcomes Utilizing a model of viremia kinetics with the assumption of a linear log-10 viremia decrease over time, we calculated the rate of decline for each patient. Our findings demonstrate that a faster rate of viremia decline is associated with a significantly reduced risk of severe clinical outcomes. We have incorporated this new analysis into the revised manuscript, providing further details in the “Statistical Analysis” section (page 7) and presenting the results on pages 15 and in Figure 6.

Comment #2: Lines 101-102. Studies A and B were conducted in parallel, and several patients enrolled in study A from primary healthcare clinics were eventually also enrolled in study B upon hospitalization. It would be helpful to know how many patients from study A were included in study B. It would also be useful for the authors to indicate if such inclusion would constitute double-counting at any point in their analyses.

To address potential confusion regarding patient overlap between studies A and B, we have provided further clarification in the revised manuscript’s Legend of Figure 1. Among confirmed dengue patients, 31 individuals enrolled in study A were later included in study B upon hospitalization. Of these, 9 had viremia measurements available in both studies and were consequently analysed in study A only. The remaining 22 lacked viremia data in study A but had measurements in study B, leading to their inclusion in study B in the analysis. We have taken meticulous care to ensure no patient data is double-counted.

Comment #3: Lines 126-127. The definition of probable primary and secondary dengue from IgG measurements needs more detail. How was the anti-DENV IgG ELISA data from paired sera interpreted?

To ensure clarity, we have moved the definitions of probable primary and secondary infections from the supplementary file (Appendix 2) to the main text of the revised manuscript (Methods section – Plasma viremia measurement, dengue diagnostics, and clinical endpoints – page 6): “A probable primary infection was defined by two negative/equivocal IgG results on separate samples taken at least two days apart within the first ten days of symptom onset, with at least one sample during the convalescent phase (days 6-10). A probable secondary infection was defined by at least one positive IgG result during the first ten days. Cases without time-appropriate IgG results were classified as indeterminate.”

Comment #4: Lines 230-232 and Figure 4. The findings reported in Figure 4 are curious. Why is the platelet count highest (significantly?) for DENV-1 compared to other DENV-type infections at low viraemia levels on LM days 1-3? Does that also mean that DENV-3 and -4 infections have a greater impact on platelet counts at days 7-10 than DENV-1 and -2?

In our analyses, we allowed the relation between viremia and platelet count to differ by serotype. Figure 4 shows the highest platelet counts for DENV-1 compared to other serotypes, especially at low viremia levels. Apparently, while DENV-1 on average has higher viremia (Figure 3), the same viremia level in DENV-1 compared to other serotypes is associated with a less severe disease course and higher platelet count. This does not necessarily imply that platelet count overall, uncorrected for viremia level, differs by genotype. Indeed, our unpublished analysis (shown below) indicates a modest influence of serotype on platelet count.

Author response image 1.

Comment #5: Figure 5. In a recent paper (Vuong et al, Clin Infect Dis 2021), the authors show elegantly that the viraemia levels on admission correlated with severe dengue. However, these correlations were different for each of the four DENV types and whether the infection was primary or secondary. Why wasn't the analysis in Figure 5 further stratified by their probable primary or secondary dengue status?

We appreciate your feedback and have stratified Figure 5 by serotype and immune status as suggested. Please note that due to the limited number of severe dengue in primary infections (only 1 case in DENV-1) and plasma leakage in primary DENV-4 (see Appendix 4-table 1), the estimated probability of having these outcomes is nearly zero across all viremia levels within these subgroups.

Comment #6: Line 279. The description in this line is at odds with the data in Figure 3A, which shows that DENV-2 could be detected over a longer period than DENV-1 as the one-step RT-qPCR assay has a lower detection limit than DENV-1.

In response to your feedback, we have revised the description to clarify that DENV-1 exhibits higher viremia levels compared to DENV-2 and DENV-3 in the revised manuscript (page 18).

Reviewer #2 (Recommendations For The Authors):

Introduction

Comment #1: Line 56: the authors state that viraemia is associated with dengue disease severity and cite their previous results. They then summarize the results of this study and others. The highlights of this paper should be described in more detail. It is important that the authors state the conclusions of their own paper, including that the association was not very strong and that the viral loads were lowest with DENV2, but DENV2 was associated with more severe disease.

Thank you for your comment. To improve the introduction’s flow, we have removed that sentence in line 56 of the manuscript and have added the weak association in the next paragraph (pages 3-4).

Comment #2: It would be important to cite smaller studies that show a delay in clearance of the virus being associated with more severe disease outcomes.

Thanks for your suggestion. We have added information to the introduction (page 4), highlighting a study which found a slower rate of viral clearance to be associated with more severe outcomes (Wang et al., 2008). However, other studies have shown no association (Vaughn et al., 2000; Fox et al., 2011). This lack of conclusive evidence underscores the need for further research.

Methods

Comment #3: The authors highlight the possible discrepancies in comparing viral kinetics of two RT-PCR methods. Although it is not ideal to combine such results, the authors have analyzed them separately, providing valuable data.

We appreciate your comment.

Comment #4: Which tests were used to define the immune status as primary and secondary? What were the definitions?

We have moved the definitions of probable primary and secondary infections from the supplementary file (Appendix 2) to the main text of the revised manuscript (Methods section – Plasma viremia measurement, dengue diagnostics, and clinical endpoints – page 6): “A probable primary infection was defined by two negative/equivocal IgG results on separate samples taken at least two days apart within the first ten days of symptom onset, with at least one sample during the convalescent phase (days 6-10). A probable secondary infection was defined by at least one positive IgG result during the first ten days. Cases without time-appropriate IgG results were classified as indeterminate.”

Results

Comment #5: It is interesting that DENV2 showed the slowest decline, but yet associated with overall lower viral loads during early illness and more severe disease outcomes. Could delayed clearance of the virus be associated with disease severity?

We have expanded our analysis to investigate the relationship between the rate of viremia decline and clinical outcomes Utilizing a model of viremia kinetics with the assumption of a linear log-10 viremia decrease over time, we calculated the rate of decline for each patient. Our findings demonstrate that a faster rate of viremia decline is associated with a significantly reduced risk of severe clinical outcomes. We have incorporated this new analysis into the revised manuscript, providing further details in the “Statistical Analysis” section (page 7) and presenting the results on pages 15 and in Figure 6.

Comment #6: Were there any differences in the kinetics of viral loads in children vs adults? I.e. children, young adults and older adults (>60 or 50?). Or were there insufficient numbers for this comparison?

To address this point, we have modified the reported results of Figure 3-D by ages of 5, 10, 15, 25, and 50 years, represented children, adolescents, young adults, and older adults. Our analysis shows that viremia kinetics are largely similar across ages.

Comment #7: Did any patients have comorbidities such as diabetes, obesity etc... if so, were there any differences in the viral loads?

We appreciate your interest in the potential impact of comorbidities on viral loads. However, due to data limitations, we were unable to analyze this association. Only 6 patients had documented diabetes in the pooled dataset. In study C, 39 patients had obesity, whereas body mass index data is not available for studies A and B, although reports suggest a lower prevalence of obesity compared to study C.

Comment #8: Were there any differences in the kinetics of the overall viral loads between DF/DHF/DSS or dengue with warning signs, without warning signs and severe dengue? Especially related to the time for viral clearance?

Thank you for your suggestion. Such analysis reverses time and the causal direction, while we are more interested in looking forward. Therefore, instead of analyzing viremia kinetics based on disease severity, we have added an analysis to investigate the relationship between the rate of decline in viremia and clinical outcomes, as shown in the response to your comment #5. Results show that a more rapid rate of viremia decline is associated with a reduced risk of more severe clinical outcomes. In addition, in this study, we selected two clinical outcomes severe dengue and plasma leakage. The definitions are based on the WHO 2009 guidelines and standard endpoint definitions for dengue trials (Tomashek et al., 2018).

Reviewer #1 (Public Review):

Summary:

Authors previously demonstrated that species-specific variation in primate CD4 impacts its ability to serve as a functional receptor for diverse SIVs. Here, Warren and Barbachano-Guerrero et al. perform population genetics analyses and functional characterization of great ape CD4 with a particular focus on gorillas, which are natural hosts of SIVgor. They first used ancestral reconstruction to derive the ancestral hominin and hominid CD4. Using pseudotyped viruses representing a panel of envelopes from SIVcpz and HIV strains, they find that these ancestral reconstructions of CD4 are more similar to human CD4 in terms of being a broadly susceptible entry receptor (in the context of mediating entry into Cf2Th cells stably expressing human CCR5). In contrast, extant gorilla and chimpanzee CD4 are functional entry receptors for a narrower range of HIV and SIVcpz isolates. Based on these differences, authors next surveyed gorilla sequences and identified several CD4 haplotypes, specifically in the region encoding the CD4 D1 domain, which directly contacts the viral glycoprotein and thus may impact the interaction. Consistent with this possibility, authors demonstrated that gorilla CD4 haplotypes are, on average, less capable of supporting entry than human CD4, and that some are largely unable to function as SIV entry receptors. Interestingly, individual residues found at key positions in the gorilla CD4 D1 when tested in the context of human CD4 reduce entry of some virions pseudotyped with diverse SIVcpz envelopes, suggesting that individual amino acids can in part explain the observed differences across gorilla CD4 haplotypes. Finally, the authors perform statistical tests to infer that CD4 from great apes with endemic SIV (i.e., chimpanzees and gorillas) but not non-reservoirs (i.e., orangutans, bonobos) or recent spillover hosts (i.e., humans), have been subject to selection as a result of pressure from endemic SIV.

The conclusions of this paper are mostly well supported by data.

Strengths:

(1) The functional assays are appropriate to test the stated hypothesis, and the authors use a broad diversity of envelopes from HIV and SIVcpz strains. Authors also partially characterize one potential mechanism of gorilla CD4 resistance - receptor glycosylation at the derived N15 found in 5/6 gorilla haplotypes.

(2) Ancestral reconstruction provides a particularly interesting aspect of the study, allowing authors to infer the ancestral state of hominid CD4 relative to modern CD4 from gorillas and chimpanzees. This, coupled with evidence supporting SIV-driven selection of gorilla CD4 diversity and the characterization of functional diversity of extant haplotypes provides several interesting findings.

Weaknesses:

(3). The major inference of the work is that SIV infection of gorillas drove the observed diversity in gorilla CD4. This is supported by the majority of SNPs being localized to the CD4 D1, which directly interacts with envelope, and the demonstrated functional consequences of that diversity for viral entry. However, SIVgor (to the best of my knowledge) only infects Western lowland gorillas (Gorilla gorilla gorilla), and one Gorilla gorilla diehli and three Gorilla beringei graueri individuals were included in the haplotype and allele frequency analyses. The presence of these haplotypes or the presence of similar allele frequencies in Eastern lowland and mountain gorillas would impact this conclusion. It would be helpful for the authors to clarify this point.

(4) The authors appear to use a somewhat atypical approach to assess intra-population selection to compensate for relatively small numbers of NHP sequences (Fig. 6). However, they do not cite precedence for the robustness of the approach or the practice of grouping sequences from multiple species for the endemic vs other comparison. They also state in the methods that some genes encoded in the locus were removed from the analysis "because they have previously been shown to directly interact with a viral protein." This seems to undercut the analysis, and prevents alternative explanations for the observed diversity in CD4 (e.g., passenger mutations from selection at a neighboring locus).

(5) Data in Figure 5 is graphed as % infected cells instead of virus titer (TDU/mL). It's unclear why this is the case, and prevents a comparison to data in Figure 2 and Figure 4.

(6) The lack of pseudotyping with SIVgor envelope is a surprising omission from this study, that would help to contextualize the findings. Similarly, building gorilla CD4 haplotype SNPs onto the hominin ancestor (as opposed to extant human CD4) may provide additional insights that are meaningful towards understanding the evolutionary trajectory of gorilla CD4.

Comments on revised version:

In the revised manuscript, the authors more appropriately contextualize conclusions that can be made based on their data versus inferences, which are now much more clearly described in the discussion. The authors also included more references to substantiate claims, additional description of methodology, and provided well-reasoned responses to the weaknesses described in my primary review.

Re: #3. As the authors point out, we do not know if eastern gorillas were at one time exposed to SIV. The authors use a variety of phylogenetic and functional approaches to infer that SIVcpz is the selective pressure-shaping gorilla CD4. While I agree this is a highly likely scenario, the allelic diversity of CD4 across gorilla subpopulations raises multiple evolutionary scenarios consistent with the data.

Re: #4. The explanation provided by the authors is reasonable. However, a demonstration that this approach is robust to potential factors that might skew the data (e.g., recombination) is argued but not tested. Part of the concern here is that the study is limited by very small sample sizes, and to the best of my knowledge, grouping sequences from multiple species to make claims about selection is not an established practice. The authors note in their response that they confirmed the existence of CD4 alleles in this study with those identified in 100 gorilla individuals from Russell et al. 2021 (unavailable to the authors at the time of submission) - a re-analysis that includes that data from Russell et al. 2021 would have strengthened the analyses.

eLife assessment

This study presents an important finding on how lentiviral infection has driven the diversification of the HIV/SIV entry receptor CD4. Using a combination of molecular evolution approaches coupled with functional testing of extant and ancestral reconstructions of great ape CD4, the authors provide solid evidence to support the idea that endemic simian immunodeficiency virus infection in gorillas have selected for gorilla CD4 alleles that are more resistant to SIV infection. Expanding the study to interrogate the evolution and function of additional primate CD4 sequences could yield more convincing evidence.

Reviewer #2 (Public Review):

Lentiviral infection of primate species has been linked to the rapid mutational evolution of numerous primate genes that interact with these viruses, including genes that inhibit lentiviruses as well as genes required for viral infection. In this manuscript, Warren et al. provide further support for the diversification of CD4, the lentiviral entry receptor, to resist lentiviral infection in great ape populations. This work builds on their prior publication (Warren et al. 2019, PMCID: PMC6561292 ) and that of other groups (e.g., Russell et al. 2021, PMCID: PMC8020793; Bibollet-Ruche et al. 2019, PMCID: PMC6386711) documenting both sequence and functional diversity in CD4, specifically within (1) the CD4 domain that binds to the lentiviral envelope and (2) great ape populations with endemic lentiviruses. Thus, the paper's finding that gorilla populations exhibit diverse CD4 alleles that differ in their susceptibility to lentiviral infection is well demonstrated both here and in a prior publication.

Strengths:

By reconstructing the CD4 sequence from the ancestor of gorillas and chimpanzees, the authors document that modern species have evolved more resistance to (admittedly modern) lentiviruses. They also deconstruct the molecular basis of this resistance by showing that one mutation, which adds a glycosylation site to CD4, is sufficient to confer lentiviral resistance to the susceptible human allele.

Weaknesses:

Warren et al. also pursue two novel lines of evidence to suggest that lentiviruses are the causative driver of great ape CD4 diversification, which seems likely from a logical perspective but is difficult to prove. First, they demonstrate that resistance to lentiviral infection is a derived trait in chimpanzees and gorillas, which have been co-evolving with endemic lentiviruses, but not in humans, which only recently acquired HIV. Nevertheless, these three examples are insufficient to prove that derived resistance is not stochastic or due to drift. The argument would be strengthened by demonstrating that bonobo and orangutan CD4, which also do not have endemic lentiviruses, resemble the ancestral and human susceptibility to great-ape-infecting lentiviruses.

Second, Warren et al. provide a population genetic argument that only endemically infected primates exhibit diversifying selection, again arguing for endemic lentiviruses being the evolutionary driver. The authors compare SNP occurrence in CD4 to neighboring genes, demonstrating that non-synonymous SNP frequency is only elevated in endemically infected species. Moreover, these amino-acid-coding changes are significantly concentrated in the CD4 domain that binds the lentiviral envelope. This is a creative analysis to overcome the problem of very small sample sizes, with very few great ape individuals sequenced. However, the small number of species compared (2-4 in each group) also limits the power of the analysis. Expanding the analysis to Old World Monkey species that do or do not have endemic lentiviruses, as well as great apes, would strengthen this argument.

Overall, this manuscript lends additional support to a well-documented example of a host-virus arms race: that of lentiviruses and the viral entry receptor.

Author response:

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

We are thankful for the comments and suggestions from the Editor and Reviewers about our manuscript submitted to the eLife Journal. We have addressed all the comments, and we think these modifications will help bring clarity to our message and be helpful to your readership. Here we include an outline of the corrections performed, as well as a detailed response to each of the reviewer’s comments.

As per the Editor and Reviewers suggestions, outline of corrections:

·        The title of the manuscript has been changed to reflect a more conservative conclusion.

·        Changes in the main manuscript text were made to enhance clarity, including the use genetic terminology and naming.

·        Specific responses to some comments from the reviewers are included in this document. We combined some comments that would be better addressed together.

Accompanied to this letter is an updated version of our manuscript with the track changes feature enabled. Again, we are thankful of the comments and suggestions we received, and we hope this revised version of our manuscript will be accompanied by an updated assessment and public reviews and a final eLife Version of Record.

Response to the public review and minor recommendations.

From Reviewer #1:

The major inference of the work is that SIV infection of gorillas drove the observed diversity in gorilla CD4. This is supported by the majority of SNPs being localized to the CD4 D1, which directly interacts with the envelope, and the demonstrated functional consequences of that diversity for viral entry. However, SIVgor (to the best of my knowledge) only infects Western lowland gorillas (Gorilla gorilla gorilla), and one Gorilla gorilla diehli and three Gorilla beringei graueri individuals were included in the haplotype and allele frequency analyses. The presence of these haplotypes or the presence of similar allele frequencies in Eastern lowland and mountain gorillas would impact this conclusion. It would be helpful for the authors to clarify this point.

From Reviewer #1 (minor comment):

Which subspecies of gorilla are the nsSNPs coming from? Gorilla gorilla diehli [n =1]; Gorilla beringei graueri [n = 3]) are not extant reservoirs of SIV and to my knowledge are not thought to have been, and so it's important to point out where the diversity is coming from if the authors are asserting that SIVgor drove this population-level diversity in gorilla CD4.

We initially included genomic data from all the gorilla individuals available to maximize sensitivity to identify allelic variants. Although evidence points to eastern gorillas not being currently infected with SIV, our results show that all allelic variants identified have differential susceptibility to the HIV-1 and SIVcpz strains tested. The allelic variants we identified with this genomic data set match the variants identified by Russell et al (doi.org/10.1073/pnas.2025914118), including the ones found in eastern gorillas, and recapitulate that those variants have differential susceptibility to lentiviral entry, similar to the variants of western populations. Whether eastern gorillas have been exposed to lentiviruses in the past remains unknown.

From Reviewer #1:

The authors appear to use a somewhat atypical approach to assess intra-population selection to compensate for relatively small numbers of NHP sequences (Fig. 6). However, they do not cite precedence for the robustness of the approach or the practice of grouping sequences from multiple species for the endemic vs other comparison. They also state in the methods that some genes encoded in the locus were removed from the analysis "because they have previously been shown to directly interact with a viral protein." This seems to undercut the analysis and prevents alternative explanations for the observed diversity in CD4 (e.g., passenger mutations from selection at a neighboring locus).

Given the nature of our samples, to detect any influence of natural selection acting on CD4, we chose to compare patterns of molecular evolution of CD4 to its neighboring loci. Comparisons of molecular evolution signatures across genomic regions are the basis of methods to detect positive selection (e.g., Sabeti DOI: 10.1038/nature01140). For our comparison, the neighboring loci represent our neutral standard for the genomic region CD4 resides. Our rationale is that demographic and neutral influences on the number and frequency of polymorphic sites in a region would equally affect all loci in a genomic region. Because these neighboring loci are our neutral benchmark, we excluded before analysis other genes in this genomic region that interact with viruses. The logic is that these loci may be evolving under the influence of positive selection and would decrease the power of our comparison. None of the excluded loci are direct neighbors to CD4. This, and given that the CD4 genomic region in humans is of average recombination rate, dampens the possibility that what we are observing at CD4 is due to selection acting at a neighboring locus. In addition, the classic population genetic method to detect positive selection, the McDonald-Kreitman test (McDonald DOI: 10.1038/351652a0), was originally presented combining polymorphism data across species. We assume that any effect on levels of diversity created by combining variability between species would equally affect all loci included in the study, not just CD4.

From Reviewer #1:

Data in Figure 5 is graphed as % infected cells instead of virus titer (TDU/mL). It's unclear why this is the case, and prevents a comparison to data in Figure 2 and Figure 4.

From Reviewer #1 (minor comment):

Figure 5: the data presentation is now shown as % infected cells instead of viral titer. This makes it difficult to compare data from Figure 5 to other figures. Can the authors please either justify this change, display data consistently or provide matched data displays as a Supplemental Figure?

For the experiments presented in figures 2 and 4 we used different volumes of infecting pseudoviruses, which allowed us to identify the linear range of infection. Then, based on the number of cells plated per experimental replicate, we calculated a virus titer. In follow-up experiments (Fig. 5), we used fixed volumes of virus that would infect ~10-20% of control (wild-type; wt) CD4-expressing cells. Comparisons were then made between wt and mutated CD4s, and these data are best presented in their raw forms as percent cells infected.  Although this change in method prevents direct comparison between the figures, we focused on the differences observed between the experimental conditions per experimental panel.

From Reviewer #1:

The lack of pseudotyping with SIVgor envelope is a surprising omission from this study, that would help to contextualize the findings.

From Reviewer #2 (minor comment):

The inclusion of HIV-1 but not SIVgor strains in Figures 2D/E is somewhat conspicuous since chimpanzee alleles certainly differ in susceptibility to SIVcpz (and SIVgor) strains per Russell et al. 2021. The authors should either test some SIVgor infections, cite published data on at least extant human/chimpanzee/gorilla CD4 susceptibility to SIVgor, or address why they did not include it.

We agree the data of host susceptibility to SIVgor strains would have been an interesting question to explore. However, we opted to focus on the transmission of SIVcpz strains into gorilla populations for this study. It is worth mentioning that we have cloned SIVgor envelope genes from some strains into our expression system, but we were unable to recover infectious pseudoviruses using an HIV-1DEnv-GFP backbone. This suggests that HIV-1 may be incompatible with incorporating SIVgor Env into virus particles. Recently, Russell et al (DOI: 10.1073/pnas.2025914118) managed to generate SIVgor Env pseudotyped virions using a different backbone (SIVcpzDEnv-GFP) that was unavailable to us at the time of this study.

From Reviewer #1:

Similarly, building gorilla CD4 haplotype SNPs onto the hominin ancestor (as opposed to extant human CD4) may provide additional insights that are meaningful toward understanding the evolutionary trajectory of gorilla CD4.

We decided to use the extant human CD4 as a backbone to test the effects on the individual amino acid variants found in the allelic diversity of the gorilla population since the human protein is highly susceptible to all the HIV-1 and SIV strains tested, and the expected phenotype is a loss-of-function. Since the D1 of the human and ancestral sequences for CD4 are almost identical (except for a change that is fixed in gorillas), and they showed similar levels of susceptibility to lentivirus entry, we expect that the phenotypes found would be the same if the gorilla SNPs were built into the ancestral CD4 backbone.

From Reviewer #2:

To bolster the argument that lentiviruses are indeed the causative driver of this diversification, which seems likely from a logical perspective but is difficult to prove, Warren et al. pursue two novel lines of evidence. First, the authors reconstruct ancestral CD4 genes that predate lentiviral infection of hominid populations. They then demonstrate that resistance to lentiviral infection is a derived trait in chimpanzees and gorillas, which have been co-evolving with endemic lentiviruses, but not in humans, which only recently acquired HIV. Nevertheless, the derived resistance could be stochastic or due to drift. This argument would be strengthened by demonstrating that bonobo and orangutan CD4, which also do not have endemic lentiviruses, resemble the ancestral and human susceptibility to great-ape-infecting lentiviruses.

From Reviewer #2 (minor comment):

The data presented in Figure 2, showing that chimp and gorilla (but not human) CD4 resistance to lentiviral infection is a derived trait, is very intriguing for suggesting that endemic lentiviruses are the causative driver of CD4 evolution. Nevertheless, this could be stochastic or due to genetic drift. Given the later emphasis on several other non-endemically infected species, the authors should at the very least include the sequences for bonobo and orangutan CD4 in the presented alignment (Fig 2B). Ideally, they would also test these orthologs to demonstrate that they are not resistant to lentiviruses infecting great apes (SIVcpz / HIV-1 / SIVgor). If they have also derived resistance, this would suggest a possible other evolutionary driver or genetic drift.

Based on our analysis on polymorphic sites using available data from populations of apes, we strongly believe the accumulation of resistant polymorphisms in CD4 did not arise in a stochastic manner. The frequency and accumulation of these changes strongly correlate with the function of CD4 as a receptor for lentivirus entry. We agree that experimentally testing the CD4 protein from bonobo and orangutan would strengthen our conclusions; however, based on our genomic analyses, we decided to focus on the species that would present a higher level of variability of susceptibility to the lentivirus tested, namely gorillas and chimpanzees.

From Reviewer #2:

Warren et al. provide a population genetic argument that only endemically infected primates exhibit diversifying selection, again arguing for endemic lentiviruses being the evolutionary driver. The authors compare SNP occurrence in CD4 to neighboring genes, demonstrating that non-synonymous SNP frequency is only elevated in endemically infected species. Moreover, these amino-acid-coding changes are significantly concentrated in the CD4 domain that binds the lentiviral envelope. This is a creative analysis to overcome the problem of very small sample sizes, with very few great ape individuals sequenced. The additional small number of species compared (2-3 in each group) also limits the power of the analysis; the authors could consider expanding their analysis to Old World Monkey species that do or do not have endemic lentiviruses, as well as great apes.

The scope of this project was to evaluate the differential phenotype of the accumulated polymorphisms found in the ape branch of the primates. Although evaluating the accumulation of polymorphisms in a broader range of primates would generate interesting observations, this would likely require increasing the total number of primate species to include sampling along the speciation tree, many of which lack population level data.

From Reviewer #1 (minor comment):

Ancestral reconstruction methods and associated data tables should be included to indicate statistical support for assigned codons. A comment on ambiguity at relevant positions is needed. Similarly, given the polymorphic nature of gorilla and chimpanzee CD4, how confident are the authors in their ancestral reconstructions based on a single representative genome per species? Does this change when you include the broader panel of gorilla sequences? Is the ancestral reconstruction robust to other methods besides PAML?

We used the PAML software package to reconstruct the ancestral hominin and hominid sequence of CD4 because it is a standard and well recognized method for this purpose. For this analysis, we used the set of primate sequences selected for positive selection analyses (see methods), namely the longest isoform sequences for each of the available species that best aligned with human CD4. We feel that the best way to perform to the ancestral state reconstruction was to use only these curated sequences instead of the population level sequences, removing potential biases introduced by having different numbers of variants per species. 

From Reviewer #1 (minor comment):

Page 10: "It seems that allele 2, which doesn't have this glycan, would be at a fitness disadvantage. In support of this, allele 2 is one of the least frequent alleles in the gorilla population that we surveyed (Figure 3B)." - this inference depends on the gorilla species that encode allele 2 and allele frequencies. There are statistical tests to address this inference.

Population genetic statistics that test for skews in sample allele frequencies are not appropriate here due to the nature of the samples in this study. However, the reviewer is correct that our inference in allele frequency is dependent on the gorilla species that we find this allele in. Allele 2 is found in the Gorilla beringei graueri subspecies of gorilla included in this study.  We only have data for three individuals (six alleles) from this subspecies compared to 51 individual (102 alleles) from Gorilla gorilla gorilla. As such, genetic subdivision between the gorilla subspecies could also produce the low frequency of allele 2 observed in our sample.

From Reviewer #1 (minor comment):

Page 11: "These results imply that the resistance to SIVcpz found in gorilla individuals is not dependent on single amino acids, but rather the cumulative effect of multiple SNPs." Would it be more relevant (or relevant in other ways) to test this statement by putting those mutations into the hominid ancestor? Testing individual residues in the context of human CD4 may be subject to epistasis or several other factors.

We agree that constructing multiple of the resistant SNPs in the susceptible human background would have strengthened our hypothesis, as all these amino acid changes are associated with increased resistance to at least one of the lentiviruses tested. However, the number of CD4 variants to test would increase significantly and we feel that this approach was out of the scope of this manuscript.

From Reviewer #1 (minor comment):

Figure 6: If you perform this analysis on chimpanzee CD4 alone do you get the same result? Just gorillas? If you remove eastern/mountain gorillas? The very small numbers of non-human non-SIV-reservoir great apes may preclude a strong conclusion.

We agree that our study is limited by the small number of available sequences from individuals of the studied species. If we remove a whole species or subspecies the statistical power would be greatly reduced. Removing all chimpanzees or gorillas (or a subspecies) would still show that only each of those species accumulate SNPs in the D1 region of CD4, although with less statistical significance.

From Reviewer #2 (minor comment):

Related to Figure 2: It would strengthen the argument that resistance is a derived trait if the authors mapped the causative mutations from gorilla CD4 onto the ancestral hominin CD4. However, this experiment is not particularly critical, merely a suggestion.

We appreciate this suggestion. We decided to use the human CD4 backbone as it is widely susceptible to lentiviral entry. The hominid and hominin ancestral sequences are almost identical to the human sequence in domain 1, except for a fixed mutation shared with the gorilla CD4. We expect that the SNPs observed in the gorilla population would also reduce susceptibility to lentivirus entry in the ancestral CD4 reconstructions.

From Reviewer #2 (minor comment):

Related to Figure 3B: It is difficult to make much of the allele frequency for 8 alleles in 32 individuals. Can the authors collate this with allele frequency for the referenced 100 individuals from Russell et al. 2021, to give a better sense of population frequency? This may allow the authors to better correlate allele frequency with SIVcpz resistance patterns in Figure 4, strengthening their argument that more resistant alleles should be over-represented in the population.

At the time of our analysis the data from Russell (DOI: 10.1073/pnas.2025914118) was not available to collate or compare. When that data became available, we immediately compared the existence of the alleles found and confirmed that the ones we found were also detected in the samples used in that study.

From Reviewer #2 (minor comment):

Related to Figure 6: As written, several methodological details should be clarified. How were human genomes selected to limit the sample size to 50?

We selected a total of 50 human individuals in order to size-match the sample size of the largest group in Fig 6B (chimpanzee, n=50). We randomly selected 10 individuals for each of the 5 superpopulations [Africans (AFR), Admixed Americans (AMR), East Asians (EAS), Europeans (EUR) and South Asians (SAS)] defined by the 1000 Genome Project.

From Reviewer #2 (minor comment):

Related to Figure 6: What comparison is being reported for the Mann-Whitney U test (CD4 vs. which gene)? Are the means shown in A an average of 2 (endemic) or 3 (non-endemic) species - if so, the authors should show the individual data points to give a clearer depiction of the data spread. In addition, it is not clear that a statistical test with sample sizes of 2 is meaningful, since Mann Whitney typically assumes n > 5. To strengthen this statistical argument, it may be necessary to include additional species that have (a) multiple genomes (or at least this locus) sequenced, and (b) have or lack lentiviral sequences. This may necessitate expanding the analysis to include Old World Monkeys (e.g. Rhesus Macaque Genome Project).

In the Figure 6 we use the Mann-Whitney U test to compare variation between CD4 and the neighboring loci. The average and SEM are for two endemic and four non-endemic species (two orangutan datasets are from two distinct species vs the gorilla subspecies). It is true our sample size is small for any statistical testing. For the Mann-Whitney U-test it is generally preferred to have n > 5 in each group. So, we do run into problems with the endemically infected comparisons as we only have two data points (chimpanzee and gorilla) for the CD4 group. For the uninfected species, CD4 has four data points.

From Reviewer #1 (minor comment):

Page 6. "This suggests that the ancestral versions of CD4 in apes were susceptible to primate lentivirus entry" - The data show that tested virus pseudotyped with SIV/HIV envs can engage ancestral CD4 in the context of a canine cell line expressing human CCR5, but not necessarily that this interaction was sufficient for the process of entry per se, especially in the context of a gorilla (or hominid) cell. Some additional context would be useful for a broad readership.

From Reviewer #1 (minor comment):

Page 6: "but that selective pressures exerted by SIVs in the chimpanzee and gorilla lineages have led to the retention of mutations that confer resistance to primate lentivirus infection. This has not happened in humans where selective pressure by HIV-1 is too new" - this cannot be concluded from the data in Figure 1. It would be more appropriate as a Discussion point.

From Reviewer #1 (minor comment):

Page 14: "Natural tolerance is often required before a virus can establish itself long term in a host reservoir, and thus understanding it is key to understanding virus reservoirs in nature" - please provide a reference. This is one among several theories of long-term host-virus evolution dynamics/outcomes, and further discussion may benefit the broad readership of eLife.

From Reviewer #1 (minor comment):

Page 15: "There is a surprising outcome of virus-driven host evolution in that the divergence and diversity of these host genes ultimately comes at a detriment to the very viruses that drove this evolution." - it is not clear to this reviewer why this is surprising.

From Reviewer #2 (minor comment):

Related to Figure 5A: The authors suggest that the gorilla glycosylation site provides resistance to SIVcpz, based on TAN1.910, but in fact the glycosylated allele is no more resistant than the un-glycosylated allele to most SIVcpz strains (in Figure 4). The authors should acknowledge this more clearly in the text.

From Reviewer #2 (minor comment):

The title of this article (that infection "has driven selection") is somewhat overstated - though it seems very likely that lentiviruses are driving CD4 diversification, this is difficult to prove. The arguments presented here rely on very few data points: modern chimp and gorilla compared to ancestral CD4, and a population genetic analysis relying on 2 or 3 species with 10-50 individuals each. The authors should either bolster these arguments (see the above suggestions) and/or soften the claim in the title.

Modifications to the main text of the manuscript have been made to enhance clarity on the subjects stated above.

eLife assessment

The humanized model of EAE represents a valuable model in which to evaluate mechanisms that may drive EAE-like processes in vivo. The data are solid given the revisions and expansion of numbers of mice to yield more statistical rigor. This model will be used by the greater community studying EAE.

Joint Public Review:

The premise of this work carries great potential. Namely, developing a humanized mouse system in which features of adaptive immunity that contribute to inflammatory demyelination can be interrogated will allow for traction into therapeutics currently unavailable to the field. Immediate questions stemming from the current study include the potential effect of ex vivo activation of PBMCs (or individual T and B cells) in vitro prior to transfer as well as the TCR and BCR repertoire of CNS vs peripheral lymphocytes before and after immunization. This group has been thoughtful and clever about their approach (e.g. use of subjects treated with natalizumab), which gives hope that fundamental aspects of pathogenesis will be uncovered by this form of modeling MS disease.

Multiple sclerosis is an inflammatory and demyelinating disease of the central nervous system where immune cells play an important role in disease pathobiology. Increased incidence of disease in individuals carrying certain HLA class-II genes plus studies in animal models suggests that HLA-DRB1*15 restricted CD4 T cells might be responsible for disease initiation, and other immune cells such as B cells, CD8 T cells, monocytes/macrophages, and dendritic cells (DC) also contribute to disease pathology. However, a direct role of human immune cells in disease is lacking to a lag between immune activation and the first sign of clinical disease. Therefore, there is an emphasis on understanding whether immune cells from HLA-DR15+ MS patients differ from HLA-DR15+ healthy controls in their phenotype and pro-inflammatory capacity. To overcome this, authors have used severely immunodeficient B2m-NOG mice that lack B, T cells and NK cells and have defective innate immune responses and engrafted PBMCs from 3 human donors (HLA-DR15+ MS and HI donors, HLA-DR13+ MS donor) in these B2m-NOG mice to determine whether they can induce CNS inflammation and demyelination like MS.

The study's strength is the use of PBMCs from HLADRB1-typed MS subjects and healthy control, the use of NOG mice, the characterization of immune subsets (revealing some interesting observations), CNS pathology etc. Weaknesses are lack of phenotype in mice and no disease phenotype even in humanized mice immunized for disease using standard disease induction protocol employed in an animal model of MS, and lack of mechanistic data on why CD8 T cells are more enriched than CD4+ T cells. The last point is important as postmortem human MS patients' brain tissue had been shown to have more CD8+ T cells than CD4+ T cells.

Thus, this work is an important step in the right direction as previous humanized studies have not used HLA-DRB1 typed PBMCs however the weaknesses as highlighted above are limitations in the model.

Author response:

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

We provide below a point-by-point reply to the Reviewers, and hope that our new manuscript will now meet the Reviewers’ concerns and the requirements for publication in eLife. 

In summary, we have performed a new set of mouse humanization experiments using a new cohort of 4 additional HLA-DRB1*15-typed MS patients as donors, all presenting with highly active disease and under treatment with natalizumab. The new experiments aim to strengthen and further extend the findings of the original paper that HLA restriction rather than disease status plays an important role in the development of CNS inflammation. Additionally, we performed EAE using a revised protocol using lower amounts of peptide antigens to reduce the possibility of immune tolerance. Indeed, our original observations were further enriched with the finding that immunization increases infiltration of the CNS by human CD4 T cells, a finding consistent with EAE pathology, and that these human CD4 T cells co-localize with human CD8 T cells in the brain lesions. Further, we provide more detailed information concerning the EBV infection status of the PBMC donors used for humanization and find some first indications of relationships between the B cell engraftment in humanized mice, EBV status  of the donors and the development of brain lesions that might stimulate further investigation in future studies.   

Point-by-point reply to reviewers:

Reviewer 1:

We thank Reviewer 1 for their valuable comments, and for their support of the overall approach as a model system. We have addressed the comments by providing additional requested information, as well as performing a EAE with a revised protocol, as suggested. We believe the new results significantly upgrades the information gained from this study.

(1) Throughout their paper, the authors never quantify the difference in CD4 vs CD8 T cell infiltration into the CNS. While repeatedly claiming that there are fewer CD4 T cells present than CD8 T cells within the CNS, this data is not included. Further, spinal cord numbers of CD4 and CD8 are not provided in lieu of CD3 T cell characterization.

Reply: We have now included quantitative data for the differences in CD4 vs CD8 T cells in the brain and spinal cord of non-immunized and EAE immunized mice. Thus, in brain (Fig. 2E) and spinal cord (Fig. 3D) of non-immunized mice, and brain (Fig. 4D, E, L) and spinal cord (Fig. 5D) of immunized mice we show data for numbers of hCD8 and hCD4 T cells, and ratios of CD4 to CD8 in at borders and parenchyma. Notably, using a revised EAE protocol in the second set of experiments, we observed a marked increase in hCD4 T cell infiltration at the CNS borders and parenchyma, an observation consistent with successful EAE immunization.

B cells don't make up any significant component of the cells transferred from HLA-DR15 donors. While the cells transferred from the HLA-DR13 donor are composed of a considerable number of B cells, the mice that received these cells didn't develop any signs of neurologic disease.

In the second experiment using new DR15 MS donors, we observed significant B cell engraftment also in several groups of DR15 MS mice. With the additional groups of mice, we were able to see a relationship between B cell engraftment in DR13 and DR15 MS mice with indicators of recent or ongoing reactivation of EBV. This is an interesting preliminary observation that might be tested in future larger studies. 

(2) Incomplete exploration of potential experimental autoimmune encephalomyelitis (EAE) modeling. Comparison of the susceptibility of B2m-NOG mice to EAE dependent on various peptide doses would be highly informative. Given that the number of hCD45+ in the periphery of NOG mice decreases following this immunization it would be prudent for the authors to determine if such a high peptide dose is truly ideal for EAE development in this mouse model.

Reply: We thank the reviewer for this critical comment. In the second group of experiments (DR15 MS2-5), we revised the EAE protocol to use lower amounts of peptides in a single immunization, thereby greatly reducing the exposure of human T cells to antigen and risk of tolerance/anergy. This resulted in (i), by-pass of the reduction in proportions of peripheral hCD45 cells following immunization in the peripheral blood (Fig. 1A), and (ii), increased numbers of hCD4 T cells and hCD4/hCD8 T cell ratios at the borders and infiltrating the parenchyma of brain (Fig. 4D,E) and spinal cord (Fig. 5D). 

(3) The degree of myelin injury is not presented. The statement is repeatedly made that "demyelination was not observed in the brain or spinal cord" but no quantification of myelin staining is shown.  

Reply: The reviewer refers to a pivotal feature (and limitation) of this particular humanized model. Despite significant T cell infiltration of white and grey matter regions of brain and spinal cord, there is no detectable demyelination. This has also been reported by in independent study using a similar humanized system (Zayoud et al., 2013). We have supplemented the figures with photomicrographs showing the presence of unperturbed myelin in the corpus callosum white T cell lesions (Fig. 4F, inset stained with Luxol fast blue), and a confocal micrograph in the same region double-immunostained for hCD45 immune cells and MBP (Fig. 4G). 

Minor points:

Method of quantification (e.g. cells per brain slice in figures 2E; 4E) is not very quantitative and should be justified or more appropriately updated to be more rigorous in methodology.

Reply: In the new figures, we have changed the method of quantification of brain parenchyma infiltrating cells from per brain slice, to cells per tissue area mm2 (Fig. 2D, Fig. 4D).

Fig. 4 data should be shown from un-immunized DR15 MS and DR15 HI mice.

Reply: We now include the quantitative data from un-immunized mice compared to immunized mice in all groups (Fig. 4 C-E). 

Reviewer 2:

We thank Reviewer 2 for their very pertinent comments and overall for highlighting the importance of humanized mice as an approach for further understanding the pathobiology of MS. We also thank this reviewer for their positive comments concerning the study design, specifically the use of fresh PBMC isolated from HLADRB1-typed MS individuals and healthy control. The reviewer highlights 4 major weaknesses of the study that we have tried to address in order to increase the value of the study.

(i) Lack of sufficient sample size (n=1 in each group) to make any conclusion.

Reply: We have increased the sample size for the DR15 MS group from n=1 to n=5 by generating new humanized mice using PBMC freshly isolated from additional MS donors, all HLA-DRB1*5 with active RRMS and under treatment with natalizumab. Here we were able to maximize on our excellent collaboration with neurologists at the neighboring University Hospital, which runs a large organized MS outpatient clinic, with HLADRB1-typed MS individuals that are closely monitored over the course of their disease and therapy. In this way, we were able to address the engraftment success of human immune cells and variability in CNS lesion development across mice generated from 5 different DR15 MS patients. We also monitored markers for EBV activation status in all the patients used for mouse humanization in this study. 

(ii) Lack of phenotype in mice.

Reply: As already described in the results and address in the discussion, the B2m-NOG immunodeficient mouse strain used here is a state-of-the-art experimental tool for humanization studies, but unfortunately fails to support engraftment by human monocytes. We and previous groups (Zayoud et al., 2013) show that CNS lesions in humanized mice contain high numbers of hCD4 and CD8 T cells, accompanied by locally activated murine microglia and astrocytes, but lack human monocytes. The humanized mice contain large proportions of immature mouse CD11b+Ly6Chi monocytes in the periphery (Suppl. Table 4) but these cells are not recruited into the CNS in non-immunized or immunized humanized mice, potentially due to incompatible chemokine signals across mouse/human. The absence of human monocyte engraftment in this model is the most likely reason that lesions do not demyelinate and this limitation of the currently available host mouse strains is one that needs to be addressed before full modelling of CNS demyelination by human immune cells can be achieved.

(iii) No disease phenotype even in humanized mice immunized for disease using standard disease induction protocol employed in an animal model of MS.

Reply: As described above, following the suggestion of reviewer 1 (point 2) we revised the EAE protocol to use lower amounts of peptides given as a single immunization. This resulted in increased numbers of hCD4 T cells and the hCD4/hCD8 T cell ratios at the borders and infiltrating the parenchyma of brain ((Fig. 1E, Fig. 2D) and spinal cord (Fig. 5D), all indicative of a successful EAE immunization. Although immunized mice showed lesions with mixed populations of hCD4 and hCD8 T cells, demyelination and therefore clinical symptoms were again not observed. As outlined in (ii) above, successful human monocyte engraftment would be fundamental for the development of demyelination and clinical symptoms in PBMC humanized mice, and new immunodeficient animal strains should be developed to achieve this.  

(iv) Mechanistic data on why CD8 T cells are more enriched than CD4+ T cells.

Reply: The question of why hCD8 T cells are more enriched in the CNS than hCD4 cells is answered at least in part by the results from our new EAE experiments, which clearly show that immunization increases CNS infiltration by hCD4 T cells versus hCD8 T cells. In general, EAE protocols are designed to activate antigen-specific CD4 T cells and this is verified in the CNS of immunized humanized mice, where hCD4 T cells infiltrate to join hCD8T cells in lesion areas. The predilection of hCD8 T cells for CNS is obvious in non-immunized humanized mice, especially in the parenchyma (see Fig. 2E) and MS patients, while hCD4 infiltration becomes important after EAE immunization. The humanized model system might therefore represent a unique tool for studying mechanisms underlying preferential hCD8 T cell involvement in MS neuroinflammaton, a system that is not accurately modelled in current EAE models. As this reviewer correctly points out, this is very important point as postmortem MS patients’ brains have more CD8 T cells than CD4 T cells.

eLife assessment

In this valuable study, the authors use a computational model to investigate how recurrent connections influence the firing patterns of grid cells, which are thought to play a role in encoding an animal's position in space. The work suggests that a one-dimensional network architecture may be sufficient to generate the hexagonal firing patterns of grid cells, a possible alternative to attractor models based on recurrent connectivity between grid cells. However, the support for this proposal was incomplete, as some conclusions for how well the model dynamics are necessary to generate features of grid cell organization were not well supported.

Reviewer #1 (Public Review):

I'll begin by summarizing what I understand from the results presented, and where relevant how my understanding seems to differ from the authors' claims. I'll then make specific comments with respect to points raised in my previous review (below), using the same numbering. Because this is a revision I'll try to restrict comments here to the changes made, which provide some clarification, but leave many issues incompletely addressed.

As I understand it the main new result here is that certain recurrent network architectures promote emergence of coordinated grid firing patterns in a model previously introduced by Kropff and Treves (Hippocampus, 2008). The previous work very nicely showed that single neurons that receive stable spatial input could 'learn' to generate grid representations by combining a plasticity rule with firing rate adaptation. The previous study also showed that when multiple neurons were synaptically connected their grid representations could develop a shared orientation, although with the recurrent connectivity previously used this substantially reduced the grid scores of many of the neurons. The advance here is to show that if the initial recurrent connectivity is consistent with that of a line attractor then the network does a much better job of establishing grid firing patterns with shared orientation.

Beyond this point, things become potentially confusing. As I understand it now, the important influence of the recurrent dynamics is in establishing the shared orientation and not in its online generation. This is clear from Figure S3, but not from an initial read of the abstract or main text. This result is consistent with Kropff and Treves' initial suggestion that 'a strong collateral connection... from neuron A to neuron B... favors the two neurons to have close-by fields... Summing all possible contributions would result in a field for neuron B that is a ring around the field of neuron A.' This should be the case for the recurrent connections now considered, but the evidence provided doesn't convincingly show that attractor dynamics of the circuit are a necessary condition for this to arise. My general suggestion for the authors is to remove these kind of claims and to keep their interpretations more closely aligned with what the results show.

Major (numbered according to previous review)

(1) Does the network maintain attractor dynamics after training? Results now show that 'in a trained network without feedforward Hebbian learning the removal of recurrent collaterals results in a slight increase in gridness and spacing'. This clearly implies that the recurrent collaterals are not required for online generation of the grid patterns. This point needs to be abundantly clear in the abstract and main text so the reader can appreciate that the recurrent dynamics are important specifically during learning.<br /> (2) Additional controls for Figure 2 to test that it is connectivity rather than attractor dynamics (e.g. drawing weights from Gaussian or exponential distributions). The authors provide one additional control based on shuffling weights. However, this is far from exhaustive and it seems difficult on this basis to conclude that it is specifically the attractor dynamics that drive the emergence of coordinated grid firing.<br /> (3) What happens if recurrent connections are turned off? The new data clearly show that the recurrent connections are not required for online grid firing, but this is not clear from the abstract and is hard to appreciate from the main text.<br /> (4) This is addressed, although the legend to Fig. S2D could provide an explanation / definition for the y-axis values.<br /> (5) Given the 2D structure of the network input it perhaps isn't surprising that the network generates 2D representations and this may have little to do with its 1D connectivity. The finding that the networks maintain coordinated grids when recurrent connections are switched off supports my initial concern and the authors explanation, to me at least, remain confusing. I think it would be helpful to consider that the connectivity is specifically important for establishing the coordinated grid firing, but that the online network does not require attractor dynamics to generate coordinated grid firing.<br /> (6) Clarity of the introduction. This is somewhat clearer, but I wonder if it would be hard for someone not familiar with the literature to accurately appreciate the key points.<br /> (7) Remapping. I'm not sure why this is ill posed. It seems the proposed model can not account for remapping results (e.g. Fyhn et al. 2007). Perhaps the authors could just clearly state this as a limitation of the model (or show that it can do this).

Previous review:

This study investigates the impact of recurrent connections on grid fields generated in networks trained by adjusting the strength of feedforward spatial inputs. The main result is that if the recurrent connections in the network are given a 1D continuous attractor architecture, then aligned grid firing patterns emerge in the network following training. Detailed analyses of the low dimensional dynamics of the resulting networks are then presented. The simulations and analyses appear carefully carried out.

The feedforward model investigated by the authors (previously introduced by Kropff & Treves, 2008) is an interesting and important alternative to models that generate grid firing patterns through 2-dimensional continuous attractor network (CAN) dynamics. However, while both classes of model generate grid fields, in making comparisons the manuscript is insufficiently clear about their differences. In particular, in the CAN models grid firing is a direct result of their 2-D architecture, either a torus structure with a single activity bump (e.g. Guanella et al. 2007, Pastoll et al. 2013), or sheet with multiple local activity bumps (Fuhs & Touretzky, Burak & Fiete, 2009). In these models, spatial input can anchor the grid representations but is not necessary for grid firing. By contrast, in the feedforward models neurons transform existing spatial inputs into a grid representation. Thus, the two classes of model implement different computations; CANs path integrate, while the feedforward models transform spatial representations. A demonstration that a 1D CAN generates coordinated 2D grid fields would be surprising and important, but its less clear why coordination between grids generated by the feedforward mechanism would be surprising. As written, it's unclear which of these claims the study is trying to make. If the former, then the conclusion doesn't appear well supported by the data as presented, if the latter then the results are perhaps not so unexpected, and the imposed attractor dynamics may still not be relevant.

Whichever claim is being made, it could be helpful to more carefully evaluate the model dynamics given predictions expected for the different classes of model. Key questions that are not answered by the manuscript include:

- At what point is the 1D attractor architecture playing a role in the models presented here? Is it important specifically for training or is it also contributing to computation in the fully trained network?

- Is an attractor architecture required at all for emergence of population alignment and gridness? Key controls missing from Figure 2 include training on networks with other architectures. For example, one might consider various architectures with randomly structured connectivity (e.g. drawing weights from exponential or Gaussian distributions).

- In the trained models do the recurrent connections substantially influence activity in the test conditions? Or after training are the 1D dynamics drowned out by feedforward inputs?

- What is the low dimensional structure of the input to the network? Can the apparent discrepancy between dimensionality of architecture and representation be resolved by considering structure of the inputs, e.g. if the input is a 2 dimensional representation of location then is it surprising that the output is too?

- What happens to representations in the trained networks presented when place cells remap? Is the 1D manifold maintained as expected for CAN models, or does it reorganise?

Reviewer #3 (Public Review):

Summary:

The paper proposes an alternative to the attractor hypothesis, as an explanation for the fact that grid cell population activity patterns (within a module) span a toroidal manifold. The proposal is based on a class of models that were extensively studied in the past, in which grid cells are driven by synaptic inputs from place cells in the hippocampus. The synapses are updated according to a Hebbian plasticity rule. Combined with an adaptation mechanism, this leads to patterning of the inputs from place cells to grid cells such that the spatial activity patterns are organized as an array of localized firing fields with hexagonal order. I refer to these models below as feedforward models.

It has already been shown by Si, Kropff, and Treves in 2012 that recurrent connections between grid cells can lead to alignment of their spatial response patterns. This idea was revisited by Urdapilleta, Si, and Treves in 2017. Thus, it should already be clear that in such models, the population activity pattern spans a manifold with toroidal topology. The main new contributions in the present paper are (i) in considering a form of recurrent connectivity that was not directly addressed before. (ii) in applying topological analysis to simulations of the model. (iii) in interpreting the results as a potential explanation for the observations of Gardner et al.

Strengths:

The exploration of learning in a feedforward model, when recurrent connectivity in the grid cell layer is structured in a ring topology, is interesting. The insight that this not only align the grid cells in a common direction but also creates a correspondence between their intrinsic coordinate (in terms of the ring-like recurrent connectivity) and their tuning on the torus is interesting as well, and the paper as a whole may influence future theoretical thinking on the mechanisms giving rise to the properties of grid cells.

Weaknesses:

(1) In Si, Kropff and Treves (2012) recurrent connectivity was dependent on the head direction tuning, in addition to the location on a 2d plane, and therefore involved a ring structure. Urdapilleta, Si, and Treves considered connectivity that depends on the distance on a 2d plane. The novelty here is that the initial connectivity is structured uniquely according to latent coordinates residing on a ring.

(2) The paper refers to the initial connectivity within the grid cell layer as one that produces an attractor. However, it is not shown that this connectivity, on its own, indeed sustains persistent attractor states. Furthermore, it is not clear whether this is even necessary to obtain the results of the model. It seems possible that (possibly weaker) connections with ring topology, that do not produce attractor dynamics but induce correlations between neurons with similar locations on the ring would be sufficient to align the spatial response patterns during the learning of feedforward weights.

(3) Given that all the grid cells are driven by an input from place cells that span a 2d manifold, and that the activity in the grid cell network settles on a steady state which is uniquely determined by the inputs, it is expected that the manifold of activity states in the grid cell layer, corresponding to inputs that locally span a 2d surface, would also locally span a 2d plane. The result is not surprising. My understanding is that this result is derived as a prerequisite for the topological analysis, and it is therefore quite technical.

(4) The modeling is all done in planar 2d environments, where the feedforward learning mechanism promotes the emergence of a hexagonal pattern in the single neuron tuning curve. Under the scenario in which grid cell responses are aligned (i.e. all neurons develop spatial patterns with the same spacing and orientation) it is already quite clear, even without any topological analysis that the emerging topology of the population activity is a torus.

However, the toroidal topology of grid cells in reality has been observed by Gardner et al also in the wagon wheel environment, in sleep, and close to boundaries (whereas here the analysis is restricted to the a sub-region of the environment, far away from the walls). There is substantial evidence based on pairwise correlations that it persists also in various other situations, in which the spatial response pattern is not a hexagonal firing pattern. It is not clear that the mechanism proposed in the present paper would generate toroidal topology of the population activity in more complex environments. In fact, it seems likely that it will not do so, and this is not explored in the manuscript.

(5) Moreover, the recent work of Gardner et al. demonstrated much more than the preservation of the topology in the different environments and in sleep: the toroidal tuning curves of individual neurons remained the same in different environments. Previous works, that analyzed pairwise correlations under hippocampal inactivation and various other manipulations, also pointed towards the same conclusion. Thus, the same population activity patterns are expressed in many different conditions. In the present model, this preservation across environments is not expected. Moreover, the results of Figure 6 suggest that even across distinct rectangular environments, toroidal tuning curves will not be preserved, because there are multiple possible arrangements of the phases on the torus which emerge in different simulations.

(6) In real grid cells, there is a dense and fairly uniform representation of all phases (see the toroidal tuning of grid cells measured by Gardner et al). Thus, the highly clustered phases obtained in the model (Fig. S1) seem incompatible with the experimental reality. I suspect that this may be related to the difficulty in identifying the topology of a torus in persistent homology analysis based on the transpose of the matrix M.

(7) The motivations stated in the introduction came across to me as weak. As now acknolwledged in the manuscript, attractor models can be fully compatible with distortions of the hexagonal spatial response patterns - they become incompatible with this spatial distortions only if one adopts a highly naive and implausible hypothesis that the attractor state is updated only by path integration. While attractor models are compatible with distortions of the spatial response pattern, it is very difficult to explain why the population activity patterns are tightly preserved across multiple conditions without a rigid two-dimentional attractor structure. This strong prediction of attractor models withstood many experimental tests - in fact, I am not aware of any data set where substantial distortions of the toroidal activity manifold were observed, despite many attempts to challenge the model. This is the main motivation for attractor models. The present model does not explain these features, yet it also does not directly offer an explanation for distortions in the spatial response pattern.

(8). There is also some weakness in the mathematical description of the dynamics. Mathematical equations are formulated in discrete time steps, without a clear interpretation in terms of biophysically relevant time scales. It appears that there are no terms in the dynamics associated with an intrinsic time scale of the neurons or the synapses (a leak time constant and/or synaptic time constants). I generally favor simple models without lots of complexity, yet within this style of modelling, the formulation adopted in this manuscript is unconventional, introducing a difficulty in interpreting synaptic weights as being weak or strong, and a difficulty in interpreting the model in the context of other studies.

In my view, the weaknesses discussed above limit the ability of the model, as it stands, to offer a compelling explanation for the toroidal topology of grid cell population activity patterns, and especially the rigidity of the manifold across environments and behavioral states. Still, the work offers an interesting way of thinking on how the toroidal topology might emerge.

eLife assessment

This study presents a valuable insight into a computational mechanism of pain perception. The evidence supporting the authors' claims is compelling. The work will be of interest to pain researchers working on computational models and cognitive mechanisms of pain in a Bayesian framework.

Reviewer #1 (Public Review):

Summary:

This study examined the role of statistical learning in pain perception, suggesting that individuals' expectations about a sequence of events influence their perception of pain intensity. They incorporated the components of volatility and stochasticity into their experimental design and asked participants (n = 27) to rate the pain intensity, their prediction, and their confidence level. They compared two different inference strategies: Bayesian inference vs. heuristic-employing Kalman filters and model-free reinforcement learning. They showed that the expectation-weighted Kalman filter best explained the temporal pattern of participants' ratings. These results provide evidence for a Bayesian inference perspective on pain, supported by a computational model that elucidates the underlying process.

Strengths:

- Their experimental design included a wide range of input intensities and the levels of volatility and stochasticity. With elaborated computational models, they provide solid evidence that statistical learning shapes pain.

Weaknesses:

- Relevance to clinical pain: While the authors underscore the relevance of their findings to chronic pain, they did not include data pertaining to clinical pain.

Reviewer #3 (Public Review):

The study investigated how statistical aspects of temperature sequences, such as manipulations of stochasticity (i.e., randomness of a sequence) and volatility (i.e., speed at which a sequence unfolded) influenced pain perception. Using an innovative stimulation paradigm and computational modelling of perceptual variables, this study demonstrated that perception is weighted by expectations. Overall, the findings support the conclusion that pain perception is mediated by expectations in a Bayesian manner. The provision of additional details during the review process strengthens the reliability of this conclusion. The methods presented offer tools and frameworks for further research in pain perception and can be extended to investigations into chronic pain processes.

Author response:

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

eLife assessment

This study presents a valuable insight into a computational mechanism of pain perception. The evidence supporting the authors’ claims is solid, although the inclusion of 1) more diverse candidate computational models, 2) more systematic analysis of the temporal regularity effects on the model fit, and 3) tests on clinical samples would have strengthened the study. The work will be of interest to pain researchers working on computational models and cognitive mechanisms of pain in a Bayesian framework.

Thank you very much again for considering the manuscript and judging it as a valuable contribution to understanding mechanisms of pain perception. We recognise the above-mentioned points of improvement and elaborate on them in the initial response to the reviewers.

Response to the reviewers

Reviewer 1:

Reviewer Comment 1.1 — Selection of candidate computational models: While the paper juxtaposes the simple model-free RL model against a Kalman Filter model in the context of pain perception, the rationale behind this choice remains ambiguous. It prompts the question: could other RL-based models, such as model-based RL or hierarchical RL, offer additional insights? A more detailed explanation of their computational model selection would provide greater clarity and depth to the study.

Initial reply: Thank you for this point. Our models were selected a-priori, following the modelling strategy from Jepma et al. (2018) and hence considered the same set of core models for clear extension of the analysis to our non-cue paradigm. The key question for us was whether expectations were used to weight the behavioural estimates, so our main interest was to compare expectation vs non-expectation weighted models.

Model-based and hierarchical RL are very broad terms that can be used to refer to many different models, and we are not clear about which specific models the reviewer is referring to. Our Bayesian models are generative models, i.e. they learn the generative statistics of the environment (which is characterised by inherent stochasticity and volatility) and hence operate model-based analyses of the stimulus dynamics. In our case, this happened hierarchically and it was combined with a simple RL rule.

Revised reply: We clarified our modelling choices in the ”Modelling strategy” subsection of the results section.

Reviewer Comment 1.2 — Effects of varying levels of volatility and stochasticity: The study commendably integrates varying levels of volatility and stochasticity into its experimental design. However, the depth of analysis concerning the effects of these variables on model fit appears shallow. A looming concern is whether the superior performance of the expectation-weighted Kalman Filter model might be a natural outcome of the experimental design. While the non-significant difference between eKF and eRL for the high stochasticity condition somewhat alleviates this concern, it raises another query: Would a more granular analysis of volatility and stochasticity effects reveal fine-grained model fit patterns?

Initial reply: We are sorry that the reviewer finds shallow ”the depth of analysis concerning the effects of these variables on model fit”. We are not sure which analysis the reviewer has in mind when suggesting a ”more granular analysis of volatility and stochasticity effects” to ”reveal fine-grained model fit patterns”. Therefore, we find it difficult to improve our manuscript in this regard. We are happy to add analyses to our paper but we would be greatful for some specific pointers. We have already provided:

•    Analysis of model-naive performance across different levels of stochasticity and volatility (section 2.3, figure 3, supplementary information section 1.1 and tables S1-2)

•    Model fitting for each stochasticity/volatility condition (section 2.4.1, figure 4, supplementary table S5)

•    Group-level and individual-level differences of each model parameter across stochasticity/volatility conditions (supplementary information section 7, figures S4-S5).

•    Effect of confidence on scaling factor for each stochasticity/volatility condition (figure 5)

Reviewer Comment 1.3 — Rating instruction: According to Fig. 1A, participants were prompted to rate their responses to the question, ”How much pain DID you just feel?” and to specify their confidence level regarding their pain. It is difficult for me to understand the meaning of confidence in this context, given that they were asked to report their *subjective* feelings. It might have been better to query participants about perceived stimulus intensity levels. This perspective is seemingly echoed in lines 100-101, ”the primary aim of the experiment was to determine whether the expectations participants hold about the sequence inform their perceptual beliefs about the intensity of the stimuli.”

Initial reply: Thank you for raising this question, which allows us to clarify our paradigm. On half of the trials, participants were asked to report the perceived intensity of the previous stimulus; on the remaining trials, participants were requested to predict the intensity of the next stimulus. Therefore, we did query ”participants about perceived stimulus intensity levels”, as described at lines 49-55, 296-303, and depicted in figure 1.

The confidence refers to the level of confidence that participants have regarding their rating - how sure they are. This is done in addition to their perceived stimulus intensity and it has been used in a large body of previous studies in any sensory modality.

Reviewer Comment 1.4 — Relevance to clinical pain: While the authors underscore the relevance of their findings to chronic pain, they did not include data pertaining to clinical pain. Notably, their initial preprint seemed to encompass data from a clinical sample (https://www.medrxiv.org /content/10.1101/2023.03.23.23287656v1), which, for reasons unexplained, has been omitted in the current version. Clarification on this discrepancy would be instrumental in discerning the true relevance of the study’s findings to clinical pain scenarios.

Initial reply: The preprint that the Reviewer is referring to was an older version of the manuscript in which we combined two different experiments, which were initially born as separate studies: the one that we submitted to eLife (done in the lab, with noxious stimuli in healthy participants) and an online study with a different statistical learning paradigm (without noxious stimuli, in chronic back pain participants). Unfortunately, the paradigms were different and not directly comparable. Indeed, following submission to a different journal, the manuscript was criticised for this reason. We therefore split the paper in two, and submitted the first study to eLife. We are now planning to perform the same lab-based experiment with noxious stimuli on chronic back pain participants. Progress on this front has been slowed down by the fact that I (Flavia Mancini) am on maternity leave, but it remains top priority once back to work.

Reviewer Comment 1.5 — Paper organization: The paper’s organization appears a little bit weird, possibly due to the removal of significant content from their initial preprint. Sections 2.12.2 and 2.4 seem more suitable for the Methods section, while 2.3 and 2.4.1 are the only parts that present results. In addition, enhancing clarity through graphical diagrams, especially for the experimental design and computational models, would be quite beneficial. A reference point could be Fig. 1 and Fig. 5 from Jepma et al. (2018), which similarly explored RL and KF models.

Initial reply: Thank you for these suggestions. We will consider restructuring the paper in the revised version.

Revised reply: We restructured introduction, results and parts of the methods. We followed the reviewer’s suggestion regarding enhancing clarity through graphical diagrams. We have visualised the experimental design in Figure 1D. Furthemore, we have visualised the two main computational models (eRL and eKF) in Figure 2, following from Jepma et al. (2018). As a result, we have updated the notation in Section 4.4 to be clearer and consistent with the graphical representation (rename the variable referring to observed thermal input from Ot to Nt).

Reviewer Comment 1.6 — In lines 99-100, the statement ”following the work by [23]” would be more helpful if it included a concise summary of the main concepts from the referenced work.

- It would be helpful to have descriptions of the conditions that Figure 1C is elaborating on.

- In line 364, the ”N {t}” in the sentence ”The observation on trial t, N {t}”, should be O {t}.

Initial reply: Thank you for spotting these and for providing the suggestions. We will include the correction in the revised version.

Revised reply: We have added the following regarding the lines 99-100:

”We build on the work by [23], who show that pain perception is strongly influenced by expectations as defined by a cue that predicts high or low pain. In contrast to the cue-paradigm from [23], the primary aim of our experiment was to determine whether the expectations participants hold about the sequence itself inform their perceptual beliefs about the intensity of the stimuli.”

See comment in the previous reply, regarding the notation change from Ot to Nt.

Reviewer 2:

Reviewer Comment 2.1 — This is a highly interesting and novel finding with potential implications for the understanding and treatment of chronic pain where pain regulation is deficient. The paradigm is clear, the analysis is state-of-the-art, the results are convincing, and the interpretation is adequate.

Initial reply: Thank you very much for these positive comments.

Reviewer 3:

Summary:

I am pleased to have had the opportunity to review this manuscript, which investigated the role of statistical learning in the modulation of pain perception. In short, the study showed that statistical aspects of temperature sequences, with respect to specific manipulations of stochasticity (i.e., randomness of a sequence) and volatility (i.e., speed at which a sequence unfolded) influenced pain perception. Computational modelling of perceptual variables (i.e., multi-dimensional ratings of perceived or predicted stimuli) indicated that models of perception weighted by expectations were the best explanation for the data. My comments below are not intended to undermine or question the quality of this research. Rather, they are offered with the intention of enhancing what is already a significant contribution to the pain neuroscience field. Below, I highlight the strengths and weaknesses of the manuscript and offer suggestions for incorporating additional methodological details.

Strengths:

The manuscript is articulate, coherent, and skilfully written, making it accessible and engaging.

- The innovative stimulation paradigm enables the exploration of expectancy effects on perception without depending on external cues, lending a unique angle to the research.

- By including participants’ ratings of both perceptual aspects and their confidence in what they perceived or predicted, the study provides an additional layer of information to the understanding of perceptual decision-making. This information was thoughtfully incorporated into the modelling, enabling the investigation of how confidence influences learning.

- The computational modelling techniques utilised here are methodologically robust. I commend the authors for their attention to model and parameter recovery, a facet often neglected in previous computational neuroscience studies.

- The well-chosen citations not only reflect a clear grasp of the current research landscape but also contribute thoughtfully to ongoing discussions within the field of pain neuroscience.

Initial reply: We are really grateful for reviewer’s insightful comments and for providing useful guidance regarding our methodology. We are also thankful for highlighting the strengths of our manuscript. Below we respond to individual weakness mentioned in the reviews report.

Reviewer Comment 3.1 — In Figure 1, panel C, the authors illustrate the stimulation intensity, perceived intensity, and prediction intensity on the same scale, facilitating a more direct comparison. It appears that the stimulation intensity has been mathematically transformed to fit a scale from 0 to 100, aligning it with the intensity ratings corresponding to either past or future stimuli. Given that the pain threshold is specifically marked at 50 on this scale, one could logically infer that all ratings falling below this value should be deemed non-painful. However, I find myself uncertain about this interpretation, especially in relation to the term ”arbitrary units” used in the figure. I would greatly appreciate clarification on how to accurately interpret these units, as well as an explanation of the relationship between these values and the definition of pain threshold in this experiment.

Initial reply: Indeed, as detailed in the Methods section 4.3, the stimulation intensity was originally transformed from the 1-13 scale to 0-100 scale to match the scales in the participant response screens.

Following the method used to establish the pain threshold, we set the stimulus intensity of 7 as the threshold on the original 1-13 scale. However, during the rating part of the experiment, several of the participants never or very rarely selected a value above 50 (their individually defined pain threshold), despite previously indicating a moment during pain threshold procedure when a stimulus becomes painful. This then results in the re-scaled intensity values as well the perception rating, both on the same 0-100 scale of arbitrary units, to never go above the pain threshold. Please see all participant ratings and inputs in the Figure below. We see that it would be more illustrative to re-plot Figure 1 with a different exemplary participant, whose ratings go above the pain threshold, perhaps with an input intensity on the 1-13 scale on the additional right-hand-side y-axis. We will add this in the revised version as well as highlight the fact above.

Importantly, while values below 50 are deemed non-painful by participants, the thermal stimulation still activates C-fibres involved in nociception, and we would argue that the modelling framework and analysis still applies in this case.

Revised reply: We re-plotted Figure 1E-F with a different exemplary participant, whose rating go above the pain threshold. We also included all participant pain perception and prediction ratings, noxious input sequences and confidence ratings in the supplement in Figures S1-S3.

Reviewer Comment 3.2 — The method of generating fluctuations in stimulation temperatures, along with the handling of perceptual uncertainty in modelling, requires further elucidation. The current models appear to presume that participants perceive each stimulus accurately, introducing noise only at the response stage. This assumption may fail to capture the inherent uncertainty in the perception of each stimulus intensity, especially when differences in consecutive temperatures are as minimal as 1°C.

Initial reply: We agree with the reviewer that there are multiple sources of uncertainty involved in the process of rating the intensity of thermal stimuli - including the perception uncertainty. In order to include an account of inaccurate perception, one would have to consider different sources that contribute to this, which there may be many. In our approach, we consider one, which is captured in the expectation weighted model, more clearly exemplified in the expectation-weighted Kalman-Filter model (eKF). The model assumes participants perception of input as an imperfect indicator of the true level of pain. In this case, it turns out that perception is corrupted as a result of the expectation participants hold about the upcoming stimuli. The extent of this effect is partly governed by a subjective level of noise ϵ, which may also subsume other sources of uncertainty beyond the expectation effect. Moreover, the response noise ξ, could also subsume any other unexplained sources of noise.

Author response image 1.

Stimulis intensity transformation

Revised reply: We clarified our modelling choices in the ”2.2 Modelling strategy” subsection.

Reviewer Comment 3.3 — A key conclusion drawn is that eKF is a better model than eRL. However, a closer examination of the results reveals that the two models behave very similarly, and it is not clear that they can be readily distinguished based on model recovery and model comparison results.

Initial reply: While, the eKF appears to rank higher than the eRL in terms of LOOIC and sigma effects, we don’t wish to make make sweeping statements regarding significance of differences between eRL and eKF, but merely point to the trend in the data. We shall make this clearer in the revised version of the manuscript. However, the most important result is that the models involving expectation-weighing are arguably better capturing the data.

Revised reply: We elaborated on the significance statements in the ”Modelling Results” subsection:

• We considered at least a 2 sigma effect as indication of a significant difference. In each condition, the expectation weighted models (eKF and eRL) provided better fit than models without this element (KF and RL; approx. 2-4 sigma difference, as reported in Figure 5A-D). This suggests that regardless of the levels of volatility and stochasticity, participants still weigh perception of the stimuli with their expectation.

and in the first paragraph of the Discussion:

• When varying different levels of inherent uncertainty in the sequences of stimuli (stochasticity and volatility), the expectation and confidence weighted models fitted the data better than models weighted for confidence but not for expectations (Figure 5A-D). The expectation-weighted bayesian (KF) model offered a better fit than the expectation-weighted, model-free RL model, although in conditions of high stochasticity this difference was short of significance. Overall, this suggests that participants’ expectations play a significant role in the perception of sequences of noxious stimuli.

We are aware of the limitations and lack of clear guidance regarding using sigma effects to establish significance (as per reviewer’s suggestion: https://discourse.mc-stan.org/t/loo-comparison-in-referenceto-standard-error/4009). Here we decided to use the above-mentioned threshold of 2-sigma as an indication of significance, but note the potential limitations of the inferences - especially when distinguishing between eRL/eKF models.

Reviewer Comment 3.4 — Regarding model recovery, the distinction between the eKF and eRL models seems blurred. When the simulation is based on the eKF, there is no ability to distinguish whether either eKF or eRL is better. When the simulation is based on the eRL, the eRL appears to be the best model, but the difference with eKF is small. This raises a few more questions. What is the range of the parameters used for the simulations?

Initial reply: We agree that the distinction between eKF and eRL in the model recovery is not that clean-cut, which may in turn point to the similarity between the two models. To simulate the data for the model and parameter recovery analysis, we used the group means and variances estimated on the participant data to sample individual parameter values.

Reviewer Comment 3.5 — Is it possible that either eRL or eKF are best when different parameters are simulated? Additionally, increasing the number of simulations to at least 100 could provide more convincing model recovery results.

Initial reply: It could be a possibility, but would require further investigation and comparison of fits for different bins/ranges of parameters to see if there is any consistent advantage of one model over another is each bin. We will consider adding this analysis, and provide an additional 50 simulations to paint a more convincing picture.

Revised reply: We increased the number of simulations per model pair to ≈ 100 (after rejecting fits based on diagnostics criteria - E-BFMI and divergent transitions) and updated the confusion matrix (Table S4). Although the confusion between eRL and eKF remains, the model recovery shows good distinction between expectation weighted vs non-expectation weighted (and Random) models, which supports our main conclusion in the paper.

Reviewer Comment 3.6 — Regarding model comparison, the authors reported that ”the expectation-weighted KF model offered a better fit than the eRL, although in conditions of high stochasticity, this difference was short of significance against the eRL model.” This interpretation is based on a significance test that hinges on the ratio between the ELPD and the surrounding standard error (SE). Unfortunately, there’s no agreed-upon threshold of SEs that determines significance, but a general guideline is to consider ”several SEs,” with a higher number typically viewed as more robust. However, the text lacks clarity regarding the specific number of SEs applied in this test. At a cursory glance, it appears that the authors may have employed 2 SEs in their interpretation, while only depicting 1 SE in Figure 4.

Initial reply: Indeed, we considered 2 sigma effect as a threshold, however we recognise that there is no agreed-upon threshold, and shall make this and our interpretation clearer regarding the trend in the data, in the revision.

Revised reply: We clarify this further, as per our revised response to Comment 3.3 above. We have also added the following statement in section 4.5.1 (Methods, Model comparison): ”There’s no agreed-upon threshold of SEs that determines significance, but the higher the sigma difference, the more robust is the effect.”

Reviewer Comment 3.7 — With respect to parameter recovery, a few additional details could be included for completeness. Specifically, while the range of the learning rate is understandably confined between 0 and 1, the range of other simulated parameters, particularly those without clear boundaries, remains ambiguous. Including scatter plots with the simulated parameters on the xaxis and the recovered parameters on the y-axis would effectively convey this missing information.

Furthermore, it would be beneficial for the authors to clarify whether the same priors were used for both the modelling results presented in the main paper and the parameter recovery presented in the supplementary material.

Initial reply: Thanks for this comment and for the suggestions. To simulate the data for the model and parameter recovery analysis, we used the group means and variances estimated on the participant data to sample individual parameter values. The priors on the group and individual-level parameters in the recovery analysis where the same as in the fitting procedure. We will include the requested scatter plots in the next iteration of the manuscript.

Revised reply: We included parameter recovery scatter plots for each model and parameter in the Supplement Figures S7-S11.

Reviewer Comment 3.8 — While the reliance on R-hat values for convergence in model fitting is standard, a more comprehensive assessment could include estimates of the effective sample size (bulk ESS and/or tail ESS) and the Estimated Bayesian Fraction of Missing Information (EBFMI), to show efficient sampling across the distribution. Consideration of divergences, if any, would further enhance the reliability of the results.

Initial reply: Thank you very much for this suggestion, we will aim to include these measures in the revised version.

Revised reply: We have considered the suggested diagnostics and include bulk and tail ESS values for each condition, model, parameter in the Supplement Tables S6-S9. We also report number of chain with low E-BFMI (0), number of divergent transitions (0) and the E-BFMI values per chain in Table S10.

Reviewer Comment 3.9 — The authors write: ”Going beyond conditioning paradigms based in cuing of pain outcomes, our findings offer a more accurate description of endogenous pain regulation.” Unfortunately, this statement isn’t substantiated by the results. The authors did not engage in a direct comparison between conditioning and sequence-based paradigms. Moreover, even if such a comparison had been made, it remains unclear what would constitute the gold standard for quantifying ”endogenous pain regulation.”

Initial reply: This is valid point, indeed we do not compare paradigms in our study, and will remove this statement in the future version.

Revised reply: We have removed this statement from the revised version.

Reviewer Comment 3.10 — In relation to the comment on model comparison in my public review, I believe the following link may provide further insight and clarify the basis for my observation. It discusses the use of standard error in model comparison and may be useful for the authors in addressing this particular point: https://discourse.mc-stan.org/t/loo-comparison-in-referenceto-standard-error/4009

Initial reply: Thank you for this suggestion, we will consider the forum discussion in our manuscript.

eLife assessment

This useful study reports how neuronal activity in the prefrontal cortex maps time intervals during which animals have to wait until reaching a reward and how this mapping is preserved across days. However, the evidence supporting the claims is incomplete as these sequential neuronal patterns do not necessarily represent time but instead may be correlated with stereotypical behavior and restraint from impulsive decision, which would require further controls (e.g. behavioral analysis) to clarify the main message. The study will be of interest to neuroscientists interested in decision making and motor control.

Reviewer #1 (Public Review):

Summary:

This paper investigates the neural population activity patterns of the medial frontal cortex in rats performing a nose poking timing task using in vivo calcium imaging. The results showed neurons that were active at the beginning and end of the nose poking and neurons that formed sequential patterns of activation that covaried with the timed interval during nose poking on a trial-by-trial basis. The former were not stable across sessions, while the latter tended to remain stable over weeks. The analysis on incorrect trials suggests the shorter non-rewarded intervals were due to errors in the scaling of the sequential pattern of activity.

Strengths:

This study measured stable signals using in vivo calcium imaging during experimental sessions that were separated by many days in animals performing a nose poking timing task. The correlation analysis on the activation profile to separate the cells in the three groups was effective and the functional dissociation between beginning and end, and duration cells was revealing. The analysis on the stability of decoding of both the nose poking state and poking time was very informative. Hence, this study dissected a neural population that formed sequential patterns of activation that encoded timed intervals.

Weaknesses:

It is not clear whether animals had enough simultaneously recorded cells to perform the analyzes of Figures 2-4. In fact, rat 3 had 18 responsive neurons which probably is not enough to get robust neural sequences for the trial-by-trial analysis and the correct and incorrect trial analysis. In addition, the analysis of behavioral errors could be improved. The analysis in Figure 4A could be replaced by a detailed analysis on the speed, and the geometry of neural population trajectories for correct and incorrect trials. In the case of Figure 4G is not clear why the density of errors formed two clusters instead of having a linear relation with the produce duration. I would be recommendable to compute the scaling factor on neuronal population trajectories and single cell activity or the computation of the center of mass to test the type III errors.

Due to the slow time resolution of calcium imaging, it is difficult to perform robust analysis on ramping activity. Therefore, I recommend downplaying the conclusion that: "Together, our data suggest that sequential activity might be a more relevant coding regime than the ramping activity in representing time under physiological conditions."

Reviewer #2 (Public Review):

In this manuscript, Li and collaborators set out to investigate the neuronal mechanisms underlying "subjective time estimation" in rats. For this purpose, they conducted calcium imaging in the prefrontal cortex of water-restricted rats that were required to perform an action (nosepoking) for a short duration to obtain drops of water. The authors provided evidence that animals progressively improved in performing their task. They subsequently analyzed the calcium imaging activity of neurons and identify start, duration, and stop cells associated with the nose poke. Specifically, they focused on duration cells and demonstrated that these cells served as a good proxy for timing on a trial-by-trial basis, scaling their pattern of actvity in accordance with changes in behavioral performance. In summary, as stated in the title, the authors claim to provide mechanistic insights into subjective time estimation in rats, a function they deem important for various cognitive conditions.

This study aligns with a wide range of studies in system neuroscience that presume that rodents solve timing tasks through an explicit internal estimation of duration, underpinned by neuronal representations of time. Within this framework, the authors performed complex and challenging experiments, along with advanced data analysis, which undoubtedly merits acknowledgement. However, the question of time perception is a challenging one, and caution should be exercised when applying abstract ideas derived from human cognition to animals. Studying so-called time perception in rats has significant shortcomings because, whether acknowledged or not, rats do not passively estimate time in their heads. They are constantly in motion. Moreover, rats do not perform the task for the sake of estimating time but to obtain their rewards are they water restricted. Their behavior will therefore reflects their motivation and urgency to obtain rewards. Unfortunately, it appears that the authors are not aware of these shortcomings. These alternative processes (motivation, sensorimotor dynamics) that occur during task performance are likely to influence neuronal activity. Consequently, my review will be rather critical. It is not however intended to be dismissive. I acknowledge that the authors may have been influenced by numerous published studies that already draw similar conclusions. Unfortunately, all the data presented in this study can be explained without invoking the concept of time estimation. Therefore, I hope the authors will find my comments constructive and understand that as scientists, we cannot ignore alternative interpretations, even if they conflict with our a priori philosophical stance (e.g., duration can be explicitly estimated by reading neuronal representation of time) and anthropomorphic assumptions (e.g., rats estimate time as humans do). While space is limited in a review, if the authors are interested, they can refer to a lengthy review I recently published on this topic, which demonstrates that my criticism is supported by a wide range of timing experiments across species (Robbe, 2023). In addition to this major conceptual issue that cast doubt on most of the conclusions of the study, there are also several major statistical issues.

Main Concerns

(#1) The authors used a task in which rats must poke for a minimal amount of time (300 ms and then 1500 ms) to be able to obtain a drop of water delivered a few centimeters right below the nosepoke. They claim that their task is a time estimation task. However, they forget that they work with thirsty rats that are eager to get water sooner than later (there is a reason why they start by a short duration!). This task is mainly probing the animals ability to wait (that is impulse control) rather than time estimation per se. Second, the task does not require to estimate precisely time because there appear to be no penalties when the nosepokes are too short or when they exceed. So it will be unclear if the variation in nosepoke reflects motivational changes rather than time estimation changes. The fact that this behavioral task is a poor assay for time estimation and rather reflects impulse control is shown by the tendency of animals to perform nose-pokes that are too short, the very slow improvement in their performance (Figure 1, with most of the mice making short responses), and the huge variability. Not only do the behavioral data not support the claim of the authors in terms of what the animals are actually doing (estimating time), but this also completely annhilates the interpretation of the Ca++ imaging data, which can be explained by motivational factors (changes in neuronal activity occurring while the animals nose poke may reflect a growing sens of urgency to check if water is available).

(#2) A second issue is that the authors seem to assume that rats are perfectly immobile and perform like some kind of robots that would initiate nose pokes, maintain them, and remove them in a very discretized manner. However, in this kind of task, rats are constantly moving from the reward magazine to the nose poke. They also move while nose-poking (either their body or their mouth), and when they come out of the nose poke, they immediately move toward the reward spout. Thus, there is a continuous stream of movements, including fidgeting, that will covary with timing. Numerous studies have shown that sensorimotor dynamics influence neural activity, even in the prefrontal cortex. Therefore, the authors cannot rule out that what the records reflect are movements (and the scaling of movement) rather than underlying processes of time estimation (some kind of timer). Concretely, start cells could represent the ending of the movement going from the water spout to the nosepoke, and end cells could be neurons that initiate (if one can really isolate any initiation, which I doubt) the movement from the nosepoke to the water spout. Duration cells could reflect fidgeting or orofacial movements combined with an increasing urgency to leave the nose pokes.

(#3) The statistics should be rethought for both the behavioral and neuronal data. They should be conducted separately for all the rats, as there is likely interindividual variability in the impulsivity of the animals.

(#4) The fact that neuronal activity reflects an integration of movement and motivational factors rather than some abstract timing appears to be well compatible with the analysis conducted on the error trials (Figure 4), considering that the sensorimotor and motivational dynamics will rescale with the durations of the nose poke.

(#5) The authors should mention upfront in the main text (result section) the temporal resolution allowed by their Ca+ probe and discuss whether it is fast enough in regard of behavioral dynamics occurring in the task.

Author response:

eLife assessment

This useful study reports how neuronal activity in the prefrontal cortex maps time intervals during which animals have to wait until reaching a reward and how this mapping is preserved across days. However, the evidence supporting the claims is incomplete as these sequential neuronal patterns do not necessarily represent time but instead may be correlated with stereotypical behavior and restraint from impulsive decision, which would require further controls (e.g. behavioral analysis) to clarify the main message. The study will be of interest to neuroscientists interested in decision making and motor control. 

We thank the editors and reviewers for the constructive comments. In light of the questions mentioned by the reviewers, we plan to perform additional analyses in our revision, particularly aiming to address issues related to single-cell scalability, and effects of motivation and movement. We believe these additional data will greatly improve the rigor and clarity of our study. We are grateful for the review process of eLife.

Public Reviews:

Reviewer #1 (Public Review):

Summary:

This paper investigates the neural population activity patterns of the medial frontal cortex in rats performing a nose poking timing task using in vivo calcium imaging. The results showed neurons that were active at the beginning and end of the nose poking and neurons that formed sequential patterns of activation that covaried with the timed interval during nose poking on a trial-by-trial basis. The former were not stable across sessions, while the latter tended to remain stable over weeks. The analysis on incorrect trials suggests the shorter non-rewarded intervals were due to errors in the scaling of the sequential pattern of activity. 

Strengths:

This study measured stable signals using in vivo calcium imaging during experimental sessions that were separated by many days in animals performing a nose poking timing task. The correlation analysis on the activation profile to separate the cells in the three groups was effective and the functional dissociation between beginning and end, and duration cells was revealing. The analysis on the stability of decoding of both the nose poking state and poking time was very informative. Hence, this study dissected a neural population that formed sequential patterns of activation that encoded timed intervals. 

We thank the reviewer for the positive comments.

Weaknesses: 

It is not clear whether animals had enough simultaneously recorded cells to perform the analyzes of Figures 2-4. In fact, rat 3 had 18 responsive neurons which probably is not enough to get robust neural sequences for the trial-by-trial analysis and the correct and incorrect trial analysis. 

We thank the reviewer for the comment. We would like to mention that the 18 cells plotted in Supplementary figure 1 were only from the duration cell category. To improve the clarity of our results, we are going to provide information regarding the number of cells from each rat in our revision. In general, we imaged more than 50 cells from each rat. We would also like to point to the data from individual trials in Supplementary figure 1B showing robust sequentiality.

In addition, the analysis of behavioral errors could be improved. The analysis in Figure 4A could be replaced by a detailed analysis on the speed, and the geometry of neural population trajectories for correct and incorrect trials.

We thank the reviewer for the suggestions. We are going to conduct the analysis as the reviewer recommended. We agree with the reviewer that better presentation of the neural activity will be helpful for the readers.

In the case of Figure 4G is not clear why the density of errors formed two clusters instead of having a linear relation with the produce duration. I would be recommendable to compute the scaling factor on neuronal population trajectories and single cell activity or the computation of the center of mass to test the type III errors. 

We would like to mention that the prediction errors plotted in this graph were calculated from two types of trials. The correct trials tended to show positive time estimation errors while the incorrect trials showed negative time estimation errors. We believe that the polarity switch between these two types suggested a possible use of this neural mechanism to time the action of the rats.

In addition, we are going to perform the analysis suggested by the reviewer in our revision. We agree that different ways of analyzing the data would provide better characterization of the scaling effect.

Due to the slow time resolution of calcium imaging, it is difficult to perform robust analysis on ramping activity. Therefore, I recommend downplaying the conclusion that: "Together, our data suggest that sequential activity might be a more relevant coding regime than the ramping activity in representing time under physiological conditions." 

We agree with the reviewer and we have mentioned this caveat in our original manuscript. We are going to rephrase the sentence as the reviewer suggested during our revision.

Reviewer #2 (Public Review):

In this manuscript, Li and collaborators set out to investigate the neuronal mechanisms underlying "subjective time estimation" in rats. For this purpose, they conducted calcium imaging in the prefrontal cortex of water-restricted rats that were required to perform an action (nosepoking) for a short duration to obtain drops of water. The authors provided evidence that animals progressively improved in performing their task. They subsequently analyzed the calcium imaging activity of neurons and identify start, duration, and stop cells associated with the nose poke. Specifically, they focused on duration cells and demonstrated that these cells served as a good proxy for timing on a trial-by-trial basis, scaling their pattern of actvity in accordance with changes in behavioral performance. In summary, as stated in the title, the authors claim to provide mechanistic insights into subjective time estimation in rats, a function they deem important for various cognitive conditions. 

This study aligns with a wide range of studies in system neuroscience that presume that rodents solve timing tasks through an explicit internal estimation of duration, underpinned by neuronal representations of time. Within this framework, the authors performed complex and challenging experiments, along with advanced data analysis, which undoubtedly merits acknowledgement. However, the question of time perception is a challenging one, and caution should be exercised when applying abstract ideas derived from human cognition to animals. Studying so-called time perception in rats has significant shortcomings because, whether acknowledged or not, rats do not passively estimate time in their heads. They are constantly in motion. Moreover, rats do not perform the task for the sake of estimating time but to obtain their rewards are they water restricted. Their behavior will therefore reflects their motivation and urgency to obtain rewards. Unfortunately, it appears that the authors are not aware of these shortcomings. These alternative processes (motivation, sensorimotor dynamics) that occur during task performance are likely to influence neuronal activity. Consequently, my review will be rather critical. It is not however intended to be dismissive. I acknowledge that the authors may have been influenced by numerous published studies that already draw similar conclusions. Unfortunately, all the data presented in this study can be explained without invoking the concept of time estimation. Therefore, I hope the authors will find my comments constructive and understand that as scientists, we cannot ignore alternative interpretations, even if they conflict with our a priori philosophical stance (e.g., duration can be explicitly estimated by reading neuronal representation of time) and anthropomorphic assumptions (e.g., rats estimate time as humans do). While space is limited in a review, if the authors are interested, they can refer to a lengthy review I recently published on this topic, which demonstrates that my criticism is supported by a wide range of timing experiments across species (Robbe, 2023). In addition to this major conceptual issue that cast doubt on most of the conclusions of the study, there are also several major statistical issues. 

Main Concerns 

(1) The authors used a task in which rats must poke for a minimal amount of time (300 ms and then 1500 ms) to be able to obtain a drop of water delivered a few centimeters right below the nosepoke. They claim that their task is a time estimation task. However, they forget that they work with thirsty rats that are eager to get water sooner than later (there is a reason why they start by a short duration!). This task is mainly probing the animals ability to wait (that is impulse control) rather than time estimation per se. Second, the task does not require to estimate precisely time because there appear to be no penalties when the nosepokes are too short or when they exceed. So it will be unclear if the variation in nosepoke reflects motivational changes rather than time estimation changes. The fact that this behavioral task is a poor assay for time estimation and rather reflects impulse control is shown by the tendency of animals to perform nose-pokes that are too short, the very slow improvement in their performance (Figure 1, with most of the mice making short responses), and the huge variability. Not only do the behavioral data not support the claim of the authors in terms of what the animals are actually doing (estimating time), but this also completely annhilates the interpretation of the Ca++ imaging data, which can be explained by motivational factors (changes in neuronal activity occurring while the animals nose poke may reflect a growing sens of urgency to check if water is available). 

We would like to respond to the reviewer’s comments 1, 2 and 4 together since they all focus on the same issue. We thank the reviewer for the very thoughtful comments and for sharing his detailed reasoning from a recently published review (Robbe, 2023). A lot of the discussion goes beyond the scope of this study and we agree that whether there is an explicit representation of time (an internal clock) in the brain is a difficult question to answer, particularly by using animal behaviors. In fact, even with fully conscious humans and elaborated task design, we think it is still questionable to clearly dissociate the neural substrate of “timing” from “motor”. In the end, it may as well be that as the reviewer cited from Bergson’s article, the experience of time cannot be measured.

Studying the neural representation of any internal state may suffer from the same ambiguity. With all due respect, however, we would like to limit our response in the scope of our results. According to the reviewer, two alternative interpretations of the task-related sequential activity exist: 1, duration cells may represent fidgeting or orofacial movements and 2, duration cells may represent motivation or motion plan of the rats. To test the first alternative interpretation, we will perform a more comprehensive analysis of the behavior data at all the limbs and visible body parts of the rat during nose poke and analyze its periodicity among different trials, although the orofacial movements may not be visible to us.

Regarding the second alternative interpretation, we think our data in the original Figure 4G argues against it. In this graph, we plotted the decoding error of time using the duration cells’ activity against the actual duration of the trials. If the sequential activity of durations cells only represents motivation, then the errors should distribute evenly across different trial times, or linearly modulated by trial durations. The unimodal distribution we observed (Figure 4G and see Author response image 1 below for a re-plot without signs) suggests that the scaling factor of the sequential activity represents information related to time. And the fact that this unimodal distribution centered at the time threshold of the task provides strong evidence for the active use of scaling factor for time estimation. In order to further test the relationship to motivation, we will measure the time interval between exiting nose poke to the start of licking water reward as an independent measurement of motivation for each trial. We will analyze and report whether this measurement correlates with the nose poking durations in our data in the revision.

Author response image 1.

Furthermore, whether the scaling sequential activity we report represents behavioral timing or true time estimation, the reviewer would agree that these activities correlate with the animal’s nose poking durations, and a previous study has showed that PFC silencing led to disruption of the mouse’s timing behavior (PMID: 24367075). The main surprising finding of the paper is that these duration cells are different from the start and end cells in terms of their coding stability. Thus, future studies dissecting the anatomical microcircuit of these duration cells may provide further clue regarding whether they receive inputs from thirst or reward-related brain regions. This may help partially resolve the “time” vs. “motor” debate the reviewer mentioned.

(2) A second issue is that the authors seem to assume that rats are perfectly immobile and perform like some kind of robots that would initiate nose pokes, maintain them, and remove them in a very discretized manner. However, in this kind of task, rats are constantly moving from the reward magazine to the nose poke. They also move while nose-poking (either their body or their mouth), and when they come out of the nose poke, they immediately move toward the reward spout. Thus, there is a continuous stream of movements, including fidgeting, that will covary with timing. Numerous studies have shown that sensorimotor dynamics influence neural activity, even in the prefrontal cortex. Therefore, the authors cannot rule out that what the records reflect are movements (and the scaling of movement) rather than underlying processes of time estimation (some kind of timer). Concretely, start cells could represent the ending of the movement going from the water spout to the nosepoke, and end cells could be neurons that initiate (if one can really isolate any initiation, which I doubt) the movement from the nosepoke to the water spout. Duration cells could reflect fidgeting or orofacial movements combined with an increasing urgency to leave the nose pokes.

(3)The statistics should be rethought for both the behavioral and neuronal data. They should be conducted separately for all the rats, as there is likely interindividual variability in the impulsivity of the animals.

We thank the reviewer for the comment, yet we are not quite sure what specifically was asked by the reviewer. There is undoubtedly variance among individual animals. One of the core reasons for statistical comparison is to compare the group difference with the variance due to sampling. It appears that the reviewer would like to require we conduct our analysis using each rat individually. We will conduct and report analysis with individual rat in Figure 1C, Figure 2C, G, K, Figure 4F in our revised manuscript.

(4) The fact that neuronal activity reflects an integration of movement and motivational factors rather than some abstract timing appears to be well compatible with the analysis conducted on the error trials (Figure 4), considering that the sensorimotor and motivational dynamics will rescale with the durations of the nose poke. 

(5) The authors should mention upfront in the main text (result section) the temporal resolution allowed by their Ca+ probe and discuss whether it is fast enough in regard of behavioral dynamics occurring in the task. 

We thank the reviewer for the suggestion. We have originally mentioned the caveat of calcium imaging in the interpretation of our results. We will incorporate more texts for this purpose during our revision. In terms of behavioral dynamics (start and end of nose poke in this case), we think calcium imaging could provide sufficient kinetics. However, the more refined dynamics related to the reproducibility of the sequential activity or the precise representation of individual cells on the scaled duration may be benefited from improved time resolution.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

(1) Please refer explicitly to the three types of cells in the abstract. 

We will modify the abstract as suggested during revision.

(2) Please refer to the work of Betancourt et al., 2023 Cell Reports, where a trial-by-trail analysis on the correlation between neural trajectory dynamics in MPC and timing behavior is reported. In that same paper the stability of neural sequences across task parameters is reported. 

We will cite and discuss this study in our revised paper.

(3) Please state the number of studied animals at the beginning of the results section. 

We will provide this information as requested. The number of animals were also plotted in Figure 1D for each analysis.

(4) Why do the middle and right panels of Figure 2E show duration cells. 

Figure 2E was intended to show examples of duration cells’ activity. We included different examples of cells that peak at different points in the scaled duration. We believe these multiple examples would give the readers a straight forward impression of these cells’ activity patterns.

(5) Which behavioral sessions of Figure 1B were analyzed further. 

We will label the analyzed sessions in Figure 1B during our revision.

(6) In Figure 3A-C please increase the time before the beginning of the trial in order to visualize properly the activation patterns of the start cells. 

We thank the reviewer for the suggestion and will modify the figure accordingly during revision.

(7) Please state what could be the behavioral and functional effect of the ablation of the cortical tissue on top of mPFC. 

We thank the reviewer for the question. In our experience, mice with lens implanted in mPFC did not show observable different to mice without surgery regarding the acquisition of the task and the distribution of the nose-poke durations. Although we could not rule out the effect on other cognitive process, the mice appeared to be intact in the scope of our task. We will provide these behavior data during our revision.

eLife assessment

This study presents a useful exploration of the complex relationship between structure and function in the developing human brain using a large-scale imaging dataset from the Human Connectome Project in Development and gene expression profiles from the Allen Brain Atlas. The evidence supporting the claims of the authors is solid, although the inclusion of more systematic analyses of structural and functional connectivity with respect to myelin measures and oligodendrocyte-related genes, and also more details regarding the imaging analyses, cognitive scores, and design and validation strategies, would have strengthened the paper. The work will be of interest to developmental biologists and neuroscientists seeking to elucidate structure-function relationships in the human brain.

Reviewer #1 (Public Review):

Summary:

This work studies spatio-temporal patterns of structure-function coupling in developing brains, using a large set of imaging data acquired from children aged 5-22. Magnetic resonance imaging data of brain structure and function were obtained from a publicly available database, from which structural and functional features and measures were derived. The authors examined the spatial patterns of structure-function coupling and how they evolve with brain development. This work further sought correlations of brain structure-function coupling with behavior and explored evolutionary, microarchitectural and genetic bases that could potentially account for the observed patterns.

Strength:

The strength of this work is the use of currently available state-of-the-art analysis methods, along with a large set of high-quality imaging data, and comprehensive examinations of structure-function coupling in developing brains. The results are comprehensive and illuminating.

Weakness:

As with most other studies, transcriptomic and cellular architectures of structure-function coupling were characterized only on the basis of a common atlas in this work.

The authors have achieved their aims in this study, and the findings provide mechanistic insights into brain development, which will inspire further basic and clinical studies along this line.

Author response:

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

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

(1) Lines 40-42: The sentence "The coupling of structural connectome (SC) and functional connectome (FC) varies greatly across different cortical regions reflecting anatomical and functional hierarchies as well as individual differences in cognitive function, and is regulated by genes" is a misstatement. Regional variations of structure-function coupling do not really reflect differences in cognitive function among individuals, but inter-subject variations do.

Thank you for your comment. We have made revisions to the sentence to correct its misstatement. Please see lines 40-43: “The coupling of structural connectome (SC) and functional connectome (FC) varies greatly across different cortical regions reflecting anatomical and functional hierarchies[1, 6-9] and is regulated by genes[6, 8], as well as its individual differences relates to cognitive function[8, 9].”

(2) In Figure 1, the graph showing the relation between intensity and cortical depth needs explanation.

Thank you for your comment. We have added necessary explanation, please see lines 133-134: “The MPC was used to map similarity networks of intracortical microstructure (voxel intensity sampled in different cortical depth) for each cortical node.”

(3) Line 167: Change "increased" to "increase".

We have corrected it, please see lines 173-174: “…networks significantly increased with age and exhibited greater increase.”

(4) Line 195: Remove "were".

We have corrected it, please see line 204: “…default mode networks significantly contributed to the prediction…”

(5) Lines 233-240, Reproducibility analyses: Comparisons of parcellation templates were not made with respect to gene weights. Is there any particular reason?

Thank you for your comment. We have quantified the gene weights based on HCPMMP using the same procedures. We identified a correlation (r \= 0.25, p<0.001) between the gene weights in HCPMMP and BNA. Given that this is a relatively weak correlation, we need to clarify the following points.

Based on HCPMMP, we produced an averaged gene expression profile for 10,027 genes covering 176 left cortical regions[1]. The excluding 4 cortical regions that had an insufficient number of assigned samples may lead to different templates having a relatively weak correlation of gene associations. Moreover, the effect of different template resolutions on the results of human connectome-transcriptome association is still unclear.

In brain connectome analysis, the choice of parcellation templates can indeed influence the subsequent findings to some extent. A methodological study[2] provided referenced correlations about 0.4~0.6 for white matter connectivity and 0.2~0.4 for white matter nodal property between two templates (refer to Figure 4 and 5 in [2]). Therefore, the age-related coupling changes as a downstream analysis was calculated using multimodal connectome and correlated with gene expression profiles, which may be influenced by the choice of templates. 

We have further supplemented gene weights results obtained from HCPMMP to explicitly clarify the dependency of parcellation templates.

Please see lines 251-252: “The gene weights of HCPMMP was consistent with that of BNA (r = 0.25, p < 0.001).”

Author response image 1.

The consistency of gene weights between HCPMMP and BNA.

Please see lines 601-604: “Finally, we produced an averaged gene expression profile for 10,027 genes covering 176 left cortical regions based on HCPMMP and obtained the gene weights by PLS analysis. We performed Pearson's correlation analyses to assess the consistency of gene weights between HCPMMP and BNA.”

Reviewer #2 (Recommendations For The Authors):

Your paper is interesting to read and I found your efforts to evaluate the robustness of the results of different parcellation strategies and tractography methods very valuable. The work is globally easy to navigate and well written with informative good-quality figures, although I think some additional clarifications will be useful to improve readability. My suggestions and questions are detailed below (I aimed to group them by topic which did not always succeed so apologies if the comments are difficult to navigate, but I hope they will be useful for reflection and to incorporate in your work).

* L34: 'developmental disorder'

** As far as I understand, the subjects in HCP-D are mostly healthy (L87). Thus, while your study provides interesting insights into typical brain development, I wonder if references to 'disorder' might be premature. In the future, it would be interesting to extend your approach to the atypical populations. In any case, it would be extremely helpful and appreciated if you included a figure visualising the distribution of behavioural scores within your population and in relationship to age at scan for your subjects (and to include a more detailed description of the assessment in the methods section) given that large part of your paper focuses on their prediction using coupling inputs (especially given a large drop of predictive performance after age correction). Such figures would allow the reader to better understand the cognitive variability within your data, but also potential age relationships, and generally give a better overview of your cohort.

We agree with your comment that references to 'disorder' is premature. We have made revisions in abstract and conclusion. 

Please see lines 33-34: “This study offers insight into the maturational principles of SC-FC coupling in typical development.”

Please see lines 395-396: “Further investigations are needed to fully explore the clinical implications of SC-FC coupling for a range of developmental disorders.”

In addition, we have included a more detailed description of the cognitive scores in the methods section and provided a figure to visualize the distributions of cognitive scores and in relationship to age for subjects. Please see lines 407-413: “Cognitive scores. We included 11 cognitive scores which were assessed with the National Institutes of Health (NIH) Toolbox Cognition Battery (https://www.healthmeasures.net/exploremeasurement-systems/nih-toolbox), including episodic memory, executive function/cognitive flexibility, executive function/inhibition, language/reading decoding, processing speed, language/vocabulary comprehension, working memory, fluid intelligence composite score, crystal intelligence composite score, early child intelligence composite score and total intelligence composite score. Distributions of these cognitive scores and their relationship with age are illustrated in Figure S12.”

Author response image 2.

Cognitive scores and age distributions of scans.

* SC-FC coupling

** L162: 'Regarding functional subnetworks, SC-FC coupling increased disproportionately with age (Figure 3C)'.

*** As far as I understand, in Figure 3C, the points are the correlation with age for a given ROI within the subnetwork. Is this correct? If yes, I am not sure how this shows a disproportionate increase in coupling. It seems that there is great variability of SC-FC correlation with age across regions within subnetworks, more so than the differences between networks. This would suggest that the coupling with age is regionally dependent rather than network-dependent? Maybe you could clarify?

The points are the correlation with age for a given ROI within the subnetwork in Figure 3C. We have revised the description, please see lines 168-174: “Age correlation coefficients distributed within functional subnetworks were shown in Figure 3C. Regarding mean SC-FC coupling within functional subnetworks, the somatomotor (𝛽𝑎𝑔𝑒\=2.39E-03, F=4.73, p\=3.10E-06, r\=0.25, p\=1.67E07, Figure 3E), dorsal attention (𝛽𝑎𝑔𝑒\=1.40E-03, F=4.63, p\=4.86E-06, r\=0.24, p\=2.91E-07, Figure 3F), frontoparietal (𝛽𝑎𝑔𝑒 =2.11E-03, F=6.46, p\=2.80E-10, r\=0.33, p\=1.64E-12, Figure 3I) and default mode (𝛽𝑎𝑔𝑒 =9.71E-04, F=2.90, p\=3.94E-03, r\=0.15, p\=1.19E-03, Figure 3J) networks significantly increased with age and exhibited greater increase.” In addition, we agree with your comment that the coupling with age is more likely region-dependent than network-dependent. We have added the description, please see lines 329-332: “We also found the SC-FC coupling with age across regions within subnetworks has more variability than the differences between networks, suggesting that the coupling with age is more likely region-dependent than network-dependent.” This is why our subsequent analysis focused on regional coupling.  

*** Additionally, we see from Figure 3C that regions within networks have very different changes with age. Given this variability (especially in the subnetworks where you show both positive and negative correlations with age for specific ROIs (i.e. all of them)), does it make sense then to show mean coupling over regions within the subnetworks which erases the differences in coupling with age relationships across regions (Figures 3D-J)?

Considering the interest and interpretation for SC-FC coupling, showing the mean coupling at subnetwork scales with age correlation is needed, although this eliminates variability at regional scale. These results at different scales confirmed that coupling changes with age at this age group are mainly increased.

*** Also, I think it would be interesting to show correlation coefficients across all regions, not only the significant ones (3B). Is there a spatially related tendency of increases/decreases (rather than a 'network' relationship)? Would it be interesting to show a similar figure to Figure S7 instead of only the significant regions?

As your comment, we have supplemented the graph which shows correlation coefficients across all regions into Figure 3B. Similarly, we supplemented to the other figures (Figure S3-S6).

Author response image 3.

Aged-related changes in SC-FC coupling. (A) Increases in whole-brain coupling with age. (B) Correlation of age with SC-FC coupling across all regions and significant regions (p<0.05, FDR corrected). (C) Comparisons of age-related changes in SC-FC coupling among functional networks. The boxes show the median and interquartile range (IQR; 25–75%), and the whiskers depict 1.5× IQR from the first or third quartile. (D-J) Correlation of age with SC-FC coupling across the VIS, SM, DA, VA, LIM, FP and DM. VIS, visual network; SM, somatomotor network; DA, dorsal attention network; VA, ventral attention network; LIM, limbic network; FP, frontoparietal network; DM, default mode network.

*** For the quantification of MPC.

**** L421: you reconstructed 14 cortical surfaces from the wm to pial surface. If we take the max thickness of the cortex to be 4.5mm (Fischl & Dale, 2000), the sampling is above the resolution of your anatomical images (0.8mm). Could you expand on what the interest is in sampling such a higher number of surfaces given that the resolution is not enough to provide additional information?

The surface reconstruction was based on state-of-the-art equivolumetric surface construction techniques[3] which provides a simplified recapitulation of cellular changes across the putative laminar structure of the cortex. By referencing a 100-μm resolution Merkerstained 3D histological reconstruction of an entire post mortem human brain (BigBrain: https://bigbrain.loris.ca/main.php), a methodological study[4] systematically evaluated MPC stability with four to 30 intracortical surfaces when the resolution of anatomical image was 0.7 mm, and selected 14 surfaces as the most stable solution. Importantly, it has been proved the in vivo approach can serve as a lower resolution yet biologically meaningful extension of the histological work[4]. 

**** L424: did you aggregate intensities over regions using mean/median or other statistics?

It might be useful to specify.

Thank you for your careful comment. We have revised the description in lines 446-447: “We averaged the intensity profiles of vertices over 210 cortical regions according to the BNA”.

**** L426: personal curiosity, why did you decide to remove the negative correlation of the intensity profiles from the MPC? Although this is a common practice in functional analyses (where the interpretation of negatives is debated), within the context of cortical correlations, the negative values might be interesting and informative on the level of microstructural relationships across regions (if you want to remove negative signs it might be worth taking their absolute values instead).

We agree with your comment that the interpretation of negative correlation is debated in MPC. Considering that MPC is a nascent approach to network modeling, we adopted a more conservative strategy that removing negative correlation by referring to the study [4] that proposed the approach. As your comment, the negative correlation might be informative. We will also continue to explore the intrinsic information on the negative correlation reflecting microstructural relationships.

**** L465: could you please expand on the notion of self-connections, it is not completely evident what this refers to.

We have revised the description in lines 493-494: “𝑁𝑐 is the number of connection (𝑁𝑐 = 245 for BNA)”.

**** Paragraph starting on L467: did you evaluate the multicollinearities between communication models? It is possibly rather high (especially for the same models with similar parameters (listed on L440-444)). Such dependence between variables might affect the estimates of feature importance (given the predictive models only care to minimize error, highly correlated features can be selected as a strong predictor while the impact of other features with similarly strong relationships with the target is minimized thus impacting the identification of reliable 'predictors').

We agree with your comment. The covariance structure (multicollinearities) among the communication models have a high probability to lead to unreliable predictor weights. In our study, we applied Haufe's inversion transform[5] which resolves this issue by computing the covariance between the predicted FC and each communication models in the training set. More details for Haufe's inversion transform please see [5]. We further clarified in the manuscript, please see in lines 497-499: “And covariance structure among the predictors may lead to unreliable predictor weights. Thus, we applied Haufe's inversion transform[38] to address these issues and identify reliable communication mechanisms.”

**** L474: I am not completely familiar with spin tests but to my understanding, this is a spatial permutation test. I am not sure how this applies to the evaluation of the robustness of feature weight estimates per region (if this was performed per region), it would be useful to provide a bit more detail to make it clearer.

As your comment, we have supplemented the detail, please see lines 503-507: “Next, we generated 1,000 FC permutations through a spin test[86] for each nodal prediction in each subject and obtained random distributions of model weights. These weights were averaged over the group and were investigated the enrichment of the highest weights per region to assess whether the number of highest weights across communication models was significantly larger than that in a random discovery.”

**** L477: 'significant communication models were used to represent WMC...', but in L103 you mention you select 3 models: communicability, mean first passage, and flow graphs. Do you want to say that only 3 models were 'significant' and these were exactly the same across all regions (and data splits/ parcellation strategies/ tractography methods)? In the methods, you describe a lot of analysis and testing but it is not completely clear how you come to the selection of the final 3, it would be beneficial to clarify. Also, the final 3 were selected on the whole dataset first and then the pipeline of SC-FC coupling/age assessment/behaviour predictions was run for every (WD, S1, S2) for both parcellations schemes and tractography methods or did you end up with different sets each time? It would be good to make the pipeline and design choices, including the validation bit clearer (a figure detailing all the steps which extend Figure 1 would be very useful to understand the design/choices and how they relate to different runs of the validation).

Thank you for your comment. In all reproducibility analyses, we used the same 3 models which was selected on the main pipeline (probabilistic tractography and BNA parcellation). According to your comment, we produced a figure that included the pipeline of model selection as the extend of Figure 1. And the description please see lines 106-108: “We used these three models to represent the extracortical connectivity properties in subsequent discovery and reproducibility analyses (Figure S1).” 

Author response image 4.

Pipeline of model selection and reproducibility analyses.

**** Might the imbalance of features between structural connectivity and MPC affect the revealed SC-FC relationships (3 vs 1)? Why did you decide on this ratio rather than for example best WM structural descriptor + MPC?

We understand your concern. The WMC communication models represent diverse geometric, topological, or dynamic factors. In order to describe the properties of WMC as best as possible, we selected three communication models after controlling covariance structure that can significantly predict FC from the 27 models. Compared to MPC, this does present a potential feature imbalance problem. However, this still supports the conclusion that coupling models that incorporate microarchitectural properties yield more accurate predictions of FC from SC[6, 7]. The relevant experiments are shown in Figure S2 below. If only the best WM structural descriptor is used, this may lose some communication properties of WMC.

**** L515: were intracranial volume and in-scanner head motion related to behavioural measures? These variables likely impact the inputs, do you expect them to influence the outcome assessments? Or is there a mistake on L518 and you actually corrected the input features rather than the behaviour measures?

The in-scanner head motion and intracranial volume are related to some age-adjusted behavioural measures, as shown in the following table. The process of regression of covariates from cognitive measures was based on these two cognitive prediction studies [8, 9]. Please see lines 549-554: “Prior to applying the nested fivefold cross-validation framework to each behaviour measure, we regressed out covariates including sex, intracranial volume, and in-scanner head motion from the behaviour measure[59, 69]. Specifically, we estimated the regression coefficients of the covariates using the training set and applied them to the testing set. This regression procedure was repeated for each fold.”

Author response table 1.

** Additionally, in the paper, you propose that the incorporation of cortical microstructural (myelin-related) descriptors with white-matter connectivity to explain FC provides for 'a more comprehensive perspective for characterizing the development of SC-FC coupling' (L60). This combination of cortical and white-matter structure is indeed interesting, however the benefits of incorporating different descriptors could be studied further. For example, comparing results of using only the white matter connectivity (assessed through selected communication models) ~ FC vs (white matter + MPC) ~ FC vs MPC ~ FC. Which descriptors better explain FC? Are the 'coupling trends' similar (or the same)? If yes, what is the additional benefit of using the more complex combination? This would also add strength to your statement at L317: 'These discrepancies likely arise from differences in coupling methods, highlighting the complementarity of our methods with existing findings'. Yes, discrepancies might be explained by the use of different SC inputs. However, it is difficult to see how discrepancies highlight complementarity - does MCP (and combination with wm) provide additional information to using wm structural alone?~

According to your comment, we have added the analyses based on different models using only the myelin-related predictor or WM connectivity to predict FC, and further compared the results among different models. please see lines 519-521: “In addition, we have constructed the models using only MPC or SCs to predict FC, respectively. Spearman’s correlation was used to assess the consistency between spatial patterns based on different models.” 

Please see lines 128-130: “In addition, the coupling pattern based on other models (using only MPC or only SCs to predict FC) and the comparison between the models were shown in Figure S2A-C.” Please see lines 178-179: “The age-related patterns of SC-FC coupling based other coupling models were shown in Figure S2D-F.”

Although we found that there were spatial consistencies in the coupling patterns between different models, the incorporation of MPC with SC connectivity can improve the prediction of FC than the models based on only MPC or SC. For age-related changes in coupling, the differences between the models was further amplified. We agree with you that the complementarity cannot be explicitly quantified and we have revised the description, please see line 329: “These discrepancies likely arise from differences in coupling methods.”

Author response image 5.

Comparison results between different models. Spatial pattern of mean SC-FC coupling based on MPC ~ FC (A), SCs ~ FC (B), and MPC + SCs ~ FC (C). Correlation of age with SC-FC coupling across cortex based on MPC ~ FC (D), SCs ~ FC (E), and MPC + SCs ~ FC (F).

** For the interpretation of results: L31 'SC-FC coupling is positively associated with genes in oligodendrocyte-related pathways and negatively associated with astrocyte-related gene'; L124: positive myelin content with SC-FC coupling...and similarly on L81, L219, L299, L342, and L490:

***You use a T1/T2 ratio which is (in large part) a measure of myelin to estimate the coupling between SC and FC. Evaluation with SC-FC coupling with myeline described in Figure 2E is possibly biased by the choice of this feature. Similarly, it is possible that reported positive associations with oligodendrocyte-related pathways and SC-FC coupling in your work could in part result from a bias introduced by the 'myelin descriptor' (conversely, picking up the oligodendrocyte-related genes is a nice corroboration for the T1/T2 ration being a myelin descriptor, so that's nice). However, it is possible that if you used a different descriptor of the cortical microstructure, you might find different expression patterns associated with the SCFC coupling (for example using neurite density index might pick up neuronal-related genes?). As mentioned in my previous suggestions, I think it would be of interest to first use only the white matter structural connectivity feature to assess coupling to FC and assess the gene expression in the cortical regions to see if the same genes are related, and subsequently incorporate MPC to dissociate potential bias of using a myelin measure from genetic findings.

Thank you for your insightful comments. In this paper, however, the core method of measuring coupling is to predict functional connections using multimodal structural connections, which may yield more information than a single modal. We agree with your comment that separating SCs and MPC to look at the genes involved in both separately could lead to interesting discoveries. We will continue to explore this in the future.

** Generally, I find it difficult to understand the interpretation of SC-FC coupling measures and would be interested to hear your thinking about this. As you mention on L290-294, how well SC predicts FC depends on which input features are used for the coupling assessment (more complex communication models, incorporating additional microstructural information etc 'yield more accurate predictions of FC' L291) - thus, calculated coupling can be interpreted as a measure of how well a particular set of input features explain FC (different sets will explain FC more or less well) ~ coupling is related to a measure of 'missing' information on the SC-FC relationship which is not contained within the particular set of structural descriptors - with this approach, the goal might be to determine the set that best, i.e. completely, explains FC to understand the link between structure and function. When you use the coupling measures for comparisons with age, cognition prediction etc, the 'status' of the SC-FC changes, it is no longer the amount of FC explained by the given SC descriptor set, but it's considered a descriptor in itself (rather than an effect of feature selection / SC-FC information overlap) - how do you interpret/argue for this shift of use?

Thank you for your comment. In this paper, we obtain reasonable SC-FC coupling by determining the optimal set of structural features to explain the function. The coupling essentially measures the direct correspondence between structure and function. To study the relationship between coupling and age and cognition is actually to study the age correlation and cognitive correlation of this direct correspondence between structure and function. 

** In a similar vein to the above comment, I am interested to hear what you think: on L305 you mention that 'perfect SC-FC coupling may be unlikely'. Would this reasoning suggest that functional activity takes place through other means than (and is therefore somehow independent of) biological (structural) substrates? For now, I think one can only say that we have imperfect descriptors of the structure so there is always information missing to explain function, this however does not mean the SC and FC are not perfectly coupled (only that we look at insufficient structural descriptors - limitations of what imaging can assess, what we measure etc). This is in line with L305 where you mention that 'Moreover, our results suggested that regional preferential contributions across different SCs lead to variations in the underlying communication process'. This suggests that locally different areas might use different communication models which are not reflected in the measures of SC-FC coupling that was employed, not that the 'coupling' is lower or higher (or coupling is not perfect). This is also a change in approach to L293: 'This configuration effectively releases the association cortex from strong structural constraints' - the 'release' might only be in light of the particular structural descriptors you use - is it conceivable that a different communication model would be more appropriate (and show high coupling) in these areas.

Thank you for your insightful comments. We have changed the description, please see lines 315317: “SC-FC coupling is dynamic and changes throughout the lifespan[7], particularly during adolescence[6,9], suggesting that perfect SC-FC coupling may require sufficient structural descriptors.” 

*Cognitive predictions:

** From a practical stand-point, do you think SC-FC coupling is a better (more accurate) indicator of cognitive outcomes (for example for future prediction studies) than each modality alone (which is practically easier to obtain and process)? It would be useful to check the behavioural outcome predictions for each modality separately (as suggested above for coupling estimates). In case SC-FC coupling does not outperform each modality separately, what is the benefit of using their coupling? Similarly, it would be useful to compare to using only cortical myelin for the prediction (which you showed to increase in importance for the coupling). In the case of myelin->coupling-> intelligence, if you are able to predict outcomes with the same performance from myelin without the need for coupling measures, what is the benefit of coupling?

From a predictive performance point of view, we do not believe that SC-FC coupling is a better indicator than a single mode (voxel, network or other indicator). Our starting point is to assess whether SC-FC coupling is related to the individual differences of cognitive performances rather than to prove its predictive power over other measures. As you suggest, it's a very interesting perspective on the predictive power of cognition by separating the various modalities and comparing them. We will continue to explore this issue in the future study.

** The statement on L187 'suggesting that increased SC-FC coupling during development is associated with higher intelligence' might not be completely appropriate before age corrections (especially given the large drop in performance that suggests confounding effects of age).

According to your comment, we have removed the statement.

** L188: it might be useful to report the range of R across the outer cross-validation folds as from Figure 4A it is not completely clear that the predictive performance is above the random (0) threshold. (For the sake of clarity, on L180 it might be useful for the reader if you directly report that other outcomes were not above the random threshold).

According to your comment, we have added the range of R and revised the description, please see lines 195-198: “Furthermore, even after controlling for age, SC-FC coupling remained a significant predictor of general intelligence better than at chance (Pearson’s r\=0.11±0.04, p\=0.01, FDR corrected, Figure 4A). For fluid intelligence and crystal intelligence, the predictive performances of SC-FC coupling were not better than at chance (Figure 4A).”

In a similar vein, in the text, you report Pearson's R for the predictive results but Figure 4A shows predictive accuracy - accuracy is a different (categorical) metric. It would be good to homogenise to clarify predictive results.

We have made the corresponding changes in Figure 4.

Author response image 6.

Encoding individual differences in intelligence using regional SC-FC coupling. (A) Predictive accuracy of fluid, crystallized, and general intelligence composite scores. (B) Regional distribution of predictive weight. (C) Predictive contribution of functional networks. The boxes show the median and interquartile range (IQR; 25–75%), and the whiskers depict the 1.5× IQR from the first or third quartile.

*Methods and QC:

-Parcellations

** It would be useful to mention briefly how the BNA was applied to the data and if any quality checks were performed for the resulting parcellations, especially for the youngest subjects which might be most dissimilar to the population used to derive the atlas (healthy adults HCP subjects) ~ question of parcellation quality.

We have added the description, please see lines 434-436: “The BNA[31] was projected on native space according to the official scripts (http://www.brainnetome.org/resource/) and the native BNA was checked by visual inspection.” 

** Additionally, the appropriateness of structurally defined regions for the functional analysis is also a topic of important debate. It might be useful to mention the above as limitations (which apply to most studies with similar focus).

We have added your comment to the methodological issues, please see lines 378-379: “Third, the appropriateness of structurally defined regions for the functional analysis is also a topic of important debate.”

- Tractography

** L432: it might be useful to name the method you used (probtrackx).

We have added this name to the description, please see lines 455-456: “probabilistic tractography (probtrackx)[78, 79] was implemented in the FDT toolbox …”

** L434: 'dividing the total fibres number in source region' - dividing by what?

We have revised the description, please see line 458: “dividing by the total fibres number in source region.”

** L436: 'connections in subcortical areas were removed' - why did you trace connections to subcortical areas in the first place if you then removed them (to match with cortical MPC areas I suspect)? Or do you mean there were spurious streamlines through subcortical regions that you filtered?

On the one hand we need to match the MPC, and on the other hand, as we stated in methodological issues, the challenge of accurately resolving the connections of small structures within subcortical regions using whole-brain diffusion imaging and tractography techniques[10, 11]. 

** Following on the above, did you use any exclusion masks during the tracing? In general, more information about quality checks for the tractography would be useful. For example, L437: did you do any quality evaluations based on the removed spurious streamlines? For example, were there any trends between spurious streamlines and the age of the subject? Distance between regions/size of the regions?

We did not use any exclusion masks. We performed visual inspection for the tractography quality and did not assess the relationship between spurious streamlines and age or distance between regions/size of the regions.

** L439: 'weighted probabilistic network' - this was weighted by the filtered connectivity densities or something else?

The probabilistic network is weighted by the filtered connectivity densities.

** I appreciate the short description of the communication models in Text S1, it is very useful.

Thank you for your comment.

** In addition to limitations mentioned in L368 - during reconstruction, have you noticed problems resolving short inter-hemispheric connections?

We have not considered this issue, we have added it to the limitation, please see lines 383-384: “In addition, the reconstruction of short connections between hemispheres is a notable challenge.”

- Functional analysis:

** There is a difference in acquisition times between participants below and above 8 years (21 vs 26 min), does the different length of acquisition affect the quality of the processed data?

We have made relatively strict quality control to ensure the quality of the processed data.  

** L446 'regressed out nuisance variables' - it would be informative to describe in more detail what you used to perform this.

We have provided more detail about the regression of nuisance variables, please see lines 476-477: “The nuisance variables were removed from time series based on general linear model.”

** L450-452: it would be useful to add the number of excluded participants to get an intuition for the overall quality of the functional data. Have you checked if the quality is associated with the age of the participant (which might be related to motion etc). Adding a distribution of remaining frames across participants (vs age) would be useful to see in the supplementary methods to better understand the data you are using.

We have supplemented the exclusion information of the subjects during the data processing, and the distribution and aged correlation of motion and remaining frames. Please see lines 481-485: “Quality control. The exclusion of participants in the whole multimodal data processing pipeline was depicted in Figure S13. In the context of fMRI data, we computed Pearson’s correlation between motion and age, as well as between the number of remaining frames and age, for the included participants aged 5 to 22 years and 8 to 22 years, respectively. These correlations were presented in Figure S14.”

Author response image 7.

Exclusion of participants in the whole multimodal data processing pipeline.  

Author response image 8.

Figure S14. Correlations between motion and age and number of remaining frames and age.

** L454: 'Pearson's correlation's... ' In contrast to MPC you did not remove negative correlations in the functional matrices. Why this choice?

Whether the negative correlation connection of functional signal is removed or not has always been a controversial issue. Referring to previous studies of SC-FC coupling[12-14], we find that the practice of retaining negative correlation connections has been widely used. In order to retain more information, we chose this strategy. Considering that MPC is a nascent approach to network modeling, we adopted a more conservative strategy that removing negative correlation by referring to the study [4] that proposed the approach.

- Gene expression:

** L635, you focus on the left cortex, is this common? Do you expect the gene expression to be fully symmetric (given reported functional hemispheric asymmetries)? It might be good to expand on the reasoning.

An important consideration regarding sample assignment arises from the fact that only two out of six brains were sampled from both hemispheres and four brains have samples collected only in the left. This sparse sampling should be carefully considered when combining data across donors[1]. We have supplemented the description, please see lines 569-571: “Restricting analyses to the left hemisphere will minimize variability across regions (and hemispheres) in terms of the number of samples available[40].”

** Paragraph of L537: you use evolution of coupling with age (correlation) and compare to gene expression with adults (cohort of Allen Human Brain Atlas - no temporal evolution to the gene expressions) and on L369 you mention that 'relative spatial patterns of gene expressions remain stable after birth'. Of course this is not a place to question previous studies, but would you really expect the gene expression associated with the temporary processes to remain stable throughout the development? For example, myelination would follow different spatiotemporal gradient across brain regions, is it reasonable to expect that the expression patterns remain the same? How do you then interpret a changing measure of coupling (correlation with age) with a gene expression assessed statically?

We agree with your comment that the spatial expression patterns is expected to vary at different periods. We have revised the previous description, please see lines 383-386: “Fifth, it is important to acknowledge that changes in gene expression levels during development may introduce bias in the results.”

- Reproducibility analyses:

** Paragraph L576: are we to understand that you performed the entire pipeline 3 times (WD, S1, S2) for both parcellations schemes and tractography methods (~12 times) including the selection of communication models and you always got the same best three communication models and gene expression etc? Or did you make some design choices (i.e. selection of communication models) only on a specific set-up and transfer to other settings?

The choice of communication model is established at the beginning, which we have clarified in the article, please see lines 106-108: “We used these three models to represent the extracortical connectivity properties in subsequent discovery and reproducibility analyses (Figure S1).” For reproducibility analyses (parcellation, tractography, and split-half validation), we fixed other settings and only assessed the impact of a single factor.

** Paragraph of L241: I really appreciate you evaluated the robustness of your results to different tractography strategies. It is reassuring to see the similarity in results for the two approaches. Did you notice any age-related effects on tractography quality for the two methods given the wide age range (did you check?)

In our study, the tractography quality was checked by visual inspection. Using quantifiable tools to tractography quality in future studies could answer this question objectively.

** Additionally, I wonder how much of that overlap is driven by the changes in MPC which is the same between the two methods... especially given its high weight in the SC-FC coupling you reported earlier in the paper. It might be informative to directly compare the connectivity matrices derived from the two tracto methods directly. Generally, as mentioned in the previous comments, I think it would be interesting to assess coupling using different input settings (with WM structural and MPC separate and then combined).

As your previous comment, we have examined the coupling patterns, coupling differences, coupling age correlation, and spatial correlations between the patterns based on different models, as shown in Figure S2. Please see our response to the previous comment for details.

** L251 - I also wonder if the random splitting is best adapted to validation in your case given you study relationships with age. Would it make more sense to make stratified splits to ensure a 'similar age coverage' across splits?

In our study, we adopt the random splitting process which repeated 1,000 times to minimize bias due to data partitioning. The stratification you mentioned is a reasonable method, and keeping the age distribution even will lead to higher verification similarity than our validation method. However, from the validation results of our method, the similarity is sufficient to explain the generalization of our findings.

Minor comments

L42: 'is regulated by genes'

** Coupling (if having a functional role and being regulated at all) is possibly resulting from a complex interplay of different factors in addition to genes, for example, learning/environment, it might be more cautious to use 'regulated in part by genes' or similar.

We have corrected it, please see line 42.

L43 (and also L377): 'development of SC-FC coupling'

** I know this is very nitpicky and depends on your opinion about the nature of SC-FC coupling, but 'development of SC-FC coupling' gives an impression of something maturing that has a role 'in itself' (for example development of eye from neuroepithelium to mature organ etc.). For now, I am not sure it is fully certain that SC-FC coupling is more than a byproduct of the comparison between SC and FC, using 'changes in SC-FC coupling with development' might be more apt.

We have corrected it, please see lines 43-44.

L261 'SC-FC coupling was stronger ... [] ... and followed fundamental properties of cortical organization.' vs L168 'No significant correlations were found between developmental changes in SC-FC coupling and the fundamental properties of cortical organization'.

**Which one is it? I think in the first you refer to mean coupling over all infants and in the second about correlation with age. How do you interpret the difference?

Between the ages of 5 and 22 years, we found that the mean SC-FC coupling pattern has become similar to that of adults, consistent with the fundamental properties of cortical organization. However, the developmental changes in SC-FC coupling are heterogeneous and sequential and do not follow the mean coupling pattern to change in the same magnitude.

L277: 'temporal and spatial complexity'

** Additionally, communication models have different assumptions about the flow within the structural network and will have different biological plausibility (they will be more or less

'realistic').

Here temporal and spatial complexity is from a computational point of view.

L283: 'We excluded a centralized model (shortest paths), which was not biologically plausible' ** But in Text S1 and Table S1 you specify the shortest paths models. Does this mean you computed them but did not incorporate them in the final coupling computations even if they were predictive?

** Generally, I find the selection of the final 3 communication models confusing. It would be very useful if you could clarify this further, for example in the methods section.

We used all twenty-seven communication models (including shortest paths) to predict FC at the node level for each participant. Then we identified three communication models that can significantly predict FC. For the shortest path, he was excluded because he did not meet the significance criteria. We have further added methodological details to this section, please see lines 503-507.

L332 'As we observed increasing coupling in these [frontoparietal network and default mode network] networks, this may have contributed to the improvements in general intelligence, highlighting the flexible and integrated role of these networks' vs L293 'SC-FC coupling in association areas, which have lower structural connectivity, was lower than that in sensory areas. This configuration effectively releases the association cortex from strong structural constraints imposed by early activity cascades, promoting higher cognitive functions that transcend simple sensori-motor exchanges'

** I am not sure I follow the reasoning. Could you expand on why it would be the decoupling promoting the cognitive function in one case (association areas generally), but on the reverse the increased coupling in frontoparietal promoting the cognition in the other (specifically frontoparietal)?

We tried to explain the problem, for general intelligence, increased coupling in frontoparietal could allow more effective information integration enable efficient collaboration between different cognitive processes.

* Formatting errors etc.

L52: maybe rephrase?

We have rephrased, please see lines 51-53: “The T1- to T2-weighted (T1w/T2w) ratio of MRI has been proposed as a means of quantifying microstructure profile covariance (MPC), which reflects a simplified recapitulation in cellular changes across intracortical laminar structure[6, 1215].”

L68: specialization1,[20].

We have corrected it.

L167: 'networks significantly increased with age and exhibited greater increased' - needs rephrasing.

We have corrected it.

L194: 'networks were significantly predicted the general intelligence' - needs rephrasing.

We have corrected it, please see lines 204-205: “we found that the weights of frontoparietal and default mode networks significantly contributed to the prediction of the general intelligence.”

L447: 'and temporal bandpass filtering' - there is a verb missing.

We have corrected it, please see line 471: “executed temporal bandpass filtering.”

L448: 'greater than 0.15' - unit missing.

We have corrected it, please see line 472: “greater than 0.15 mm”.

L452: 'After censoring, regression of nuisance variables, and temporal bandpass filtering,' - no need to repeat the steps as you mentioned them 3 sentences earlier.

We have removed it.

L458-459: sorry I find this description slightly confusing. What do you mean by 'modal'? Connectional -> connectivity profile. The whole thing could be simplified, if I understand correctly your vector of independent variables is a set of wm and microstructural 'connectivity' of the given node... if this is not the case, please make it clearer.

We have corrected it, please see line 488: “where 𝒔𝑖 is the 𝑖th SC profiles, 𝑛 is the number of SC profiles”.

L479: 'values and system-specific of 480 coupling'.

We have corrected it.

L500: 'regular' - regularisation.

We have changed it to “regularization”.

L567: Do you mean that in contrast to probabilistic with FSL you use deterministic methods within Camino? For L570, you introduce communication models through 'such as': did you fit all models like before? If not, it might be clearer to just list the ones you estimated rather than introduce through 'such as'.

We have changed the description to avoid ambiguity, please see lines 608-609: “We then calculated the communication properties of the WMC including communicability, mean first passage times of random walkers, and flow graphs (timescales=1).”

Citation [12], it is unusual to include competing interests in the citation, moreover, Dr. Bullmore mentioned is not in the authors' list - this is most likely an error with citation import, it would be good to double-check.

We have corrected it.

L590: Python scripts used to perform PLS regression can 591 be found at https://scikitlearn.org/. The link leads to general documentation for sklearn.

We have corrected it, please see lines 627-630: “Python scripts used to perform PLS regression can be found at https://scikit-learn.org/stable/modules/generated/sklearn.cross_decomposition.PLSRegression.html#sklearn.cro ss_decomposition.PLSRegression.”

P26 and 27 - there are two related sections: Data and code availability and Code availability - it might be worth merging into one section if possible.

We have corrected it, please see lines 623-633.

References

(1) Arnatkeviciute A, Fulcher BD, Fornito A. A practical guide to linking brain-wide gene expression and neuroimaging data. Neuroimage. 2019;189:353-67. Epub 2019/01/17. doi: 10.1016/j.neuroimage.2019.01.011. PubMed PMID: 30648605.

(2) Zhong S, He Y, Gong G. Convergence and divergence across construction methods for human brain white matter networks: an assessment based on individual differences. Hum Brain Mapp. 2015;36(5):1995-2013. Epub 2015/02/03. doi: 10.1002/hbm.22751. PubMed PMID: 25641208; PubMed Central PMCID: PMCPMC6869604.

(3) Waehnert MD, Dinse J, Weiss M, Streicher MN, Waehnert P, Geyer S, et al. Anatomically motivated modeling of cortical laminae. Neuroimage. 2014;93 Pt 2:210-20. Epub 2013/04/23. doi: 10.1016/j.neuroimage.2013.03.078. PubMed PMID: 23603284.

(4) Paquola C, Vos De Wael R, Wagstyl K, Bethlehem RAI, Hong SJ, Seidlitz J, et al. Microstructural and functional gradients are increasingly dissociated in transmodal cortices. PLoS Biol. 2019;17(5):e3000284. Epub 2019/05/21. doi: 10.1371/journal.pbio.3000284. PubMed PMID: 31107870.

(5) Haufe S, Meinecke F, Gorgen K, Dahne S, Haynes JD, Blankertz B, et al. On the interpretation of weight vectors of linear models in multivariate neuroimaging. Neuroimage. 2014;87:96-110. Epub 2013/11/19. doi: 10.1016/j.neuroimage.2013.10.067. PubMed PMID: 24239590.

(6) Demirtas M, Burt JB, Helmer M, Ji JL, Adkinson BD, Glasser MF, et al. Hierarchical Heterogeneity across Human Cortex Shapes Large-Scale Neural Dynamics. Neuron. 2019;101(6):1181-94 e13. Epub 2019/02/13. doi: 10.1016/j.neuron.2019.01.017. PubMed PMID: 30744986; PubMed Central PMCID: PMCPMC6447428.

(7) Deco G, Kringelbach ML, Arnatkeviciute A, Oldham S, Sabaroedin K, Rogasch NC, et al. Dynamical consequences of regional heterogeneity in the brain's transcriptional landscape. Sci Adv. 2021;7(29). Epub 2021/07/16. doi: 10.1126/sciadv.abf4752. PubMed PMID: 34261652; PubMed Central PMCID: PMCPMC8279501.

(8) Chen J, Tam A, Kebets V, Orban C, Ooi LQR, Asplund CL, et al. Shared and unique brain network features predict cognitive, personality, and mental health scores in the ABCD study. Nat Commun. 2022;13(1):2217. Epub 2022/04/27. doi: 10.1038/s41467-022-29766-8. PubMed PMID: 35468875; PubMed Central PMCID: PMCPMC9038754.

(9) Li J, Bzdok D, Chen J, Tam A, Ooi LQR, Holmes AJ, et al. Cross-ethnicity/race generalization failure of behavioral prediction from resting-state functional connectivity. Sci Adv. 2022;8(11):eabj1812. Epub 2022/03/17. doi: 10.1126/sciadv.abj1812. PubMed PMID: 35294251; PubMed Central PMCID: PMCPMC8926333.

(10) Thomas C, Ye FQ, Irfanoglu MO, Modi P, Saleem KS, Leopold DA, et al. Anatomical accuracy of brain connections derived from diffusion MRI tractography is inherently limited. Proc Natl Acad Sci U S A. 2014;111(46):16574-9. Epub 2014/11/05. doi: 10.1073/pnas.1405672111. PubMed PMID: 25368179; PubMed Central PMCID: PMCPMC4246325.

(11) Reveley C, Seth AK, Pierpaoli C, Silva AC, Yu D, Saunders RC, et al. Superficial white matter fiber systems impede detection of long-range cortical connections in diffusion MR tractography. Proc Natl Acad Sci U S A. 2015;112(21):E2820-8. Epub 2015/05/13. doi: 10.1073/pnas.1418198112. PubMed PMID: 25964365; PubMed Central PMCID: PMCPMC4450402.

(12) Gu Z, Jamison KW, Sabuncu MR, Kuceyeski A. Heritability and interindividual variability of regional structure-function coupling. Nat Commun. 2021;12(1):4894. Epub 2021/08/14. doi: 10.1038/s41467-021-25184-4. PubMed PMID: 34385454; PubMed Central PMCID: PMCPMC8361191.

(13) Liu ZQ, Vazquez-Rodriguez B, Spreng RN, Bernhardt BC, Betzel RF, Misic B. Time-resolved structure-function coupling in brain networks. Commun Biol. 2022;5(1):532. Epub 2022/06/03. doi: 10.1038/s42003-022-03466-x. PubMed PMID: 35654886; PubMed Central PMCID: PMCPMC9163085.

(14) Zamani Esfahlani F, Faskowitz J, Slack J, Misic B, Betzel RF. Local structure-function relationships in human brain networks across the lifespan. Nat Commun. 2022;13(1):2053. Epub 2022/04/21. doi: 10.1038/s41467-022-29770-y. PubMed PMID: 35440659; PubMed Central PMCID: PMCPMC9018911.

eLife assessment

This study addresses an important, understudied question using approaches that link molecular, circuit, and behavioral changes. The novel findings that Netrin-1 and UNC5c can guide dopaminergic innervation from the nucleus accumbens to the cortex during adolescence are solid. The data showing that the onset of Unc5 expression is sexually dimorphic in mice, and that in Siberian hamsters environmental effects on development are also sexually dimorphic are also solid. Reviewers identified some gaps in evidence for specificity of Netrin-1 expression, which, if filled, would strengthen the evidence for some of the claims. Future work would also benefit from Unc5C knockdown to corroborate the results and investigation of the cause-effect relationship. This paper will be of interest to those interested in neural development, sex differences, and/or dopamine function.

eLife assessment

The authors present a valuable computational platform, which aims to automate the workflow for coarse-grained simulations of biomolecules in the framework of the popular MARTINI model. The capability of the platform has been convincingly demonstrated by the application to a large number of proteins as well as macrocycles and polymers. On the other hand, because the developments have largely been based on the MARTINI model, some might argue that the general impact on the multi-scale simulation community is limited, leaving the support for the claimed significance incomplete.

Reviewer #1 (Public Review):

Summary:

In this study, the authors provide a new computational platform called Vermouth to automate topology generation, a crucial step that any biomolecular simulation starts with. Given a wide arrange of chemical structures that need to be simulated, varying qualities of structural models as inputs obtained from various sources, and diverse force fields and molecular dynamics engines employed for simulations, automation of this fundamental step is challenging, especially for complex systems and in case that there is a need to conduct high-throughput simulations in the application of computer-aided drug design (CADD). To overcome this challenge, the authors develop a programing library composed of components that carry out various types of fundamental functionalities that are commonly encountered in topological generation. These components are intended to be general for any type of molecules and not to depend on any specific force field and MD engines. To demonstrate the applicability of this library, the authors employ those components to re-assemble a pipeline called Martinize2 used in topology generation for simulations with a widely used coarse-grained model (CG) MARTINI. This pipeline can fully recapitulate the functionality of its original version Martinize but exhibit greatly enhanced generality, as confirmed by the ability of the pipeline to faithfully generate topologies for two high-complexity benchmarking sets of proteins.

Strengths:

The main strength of this work is the use of concepts and algorithms associated with induced subgraph in graph theory to automate several key but non-trivial steps of topology generation such as the identification of monomer residue units (MRU), the repair of input structures with missing atoms, the mapping of topologies between different resolutions, and the generation of parameters needed for describing interactions between MRUs. In addition, the documentation website provided by the authors is very informative, allowing users to get quickly started with Vermouth.

Weaknesses:

Although the Vermouth library is designed as a general tool for topology generation for molecular simulations, only its applications with MARTINI have been demonstrated in the current study. Thus, the claimed generality of Vermouth remains to be exmained. The authors may consider to point out this in their manuscript.

Reviewer #2 (Public Review):

This work introduces a Vermouth library framework to enhance software development within the Martini community. Specifically, it presents a Vermouth-powered program, Martinize2, for generating coarse-grained structures and topologies from atomistic structures. In addition to introducing the Vermouth library and the Martinize2 program, this paper illustrates how Martinize2 identifies atoms, maps them to the Martini model, generates topology files, and identifies protonation states or post-translational modifications. Compared with the prior version, the authors provide a new figure to show that Martinize2 can be applied to various molecules, such as proteins, cofactors, and lipids. To demonstrate the general application, Martinize2 was used for converting 73% of 87,084 protein structures from the template library, with failed cases primarily blamed on missing coordinates.

I was hoping to see some fundamental changes in the resubmitted version. To my disappointment, the manuscript remains largely unchanged (even the typo I pointed out previously was not fixed). I do not doubt that Martinize2 and Vermouth are useful to the Martini community, and this paper will have some impact. The manuscript is very technical and limited to the Martini community. The scientific insight for the general coarse-grained modeling community is unclear. The goal of the work is ambitious (such as high-throughput simulations and whole-cell modeling), but the results show just a validation of Martinize2. This version does not reverse my previous impression that it is incremental. As I pointed out in my previous review (and no response from the authors), all the issues associated with the Martini model are still there, e.g. the need for ENM. In this shape, I feel this manuscript is suitable for a specialized journal in computational biophysics or stays as part of the GitHub repository.

Reviewer #3 (Public Review):

The manuscript Kroon et al. described two algorithms, which when combined achieve high throughput automation of "martinizing" protein structures with selected protonation states and post-translational modifications. After the revisions provided by the authors, I recommend minor revision.

The authors have addressed most of my concerns provided previously. Specifically, showcasing the capability of coarse-graining other types of molecules (Figure 7) is a useful addition, especially for the booming field of therapeutic macrocycles.

My only additional concern is that to justify Martinize2 and Vermouth as a "high-throughput" method, the speed of these tools needs to be addressed in some form in the manuscript as a guideline to users.

Author response:

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

Reviewer #1 (Public Review):

Summary:

In this study, the authors provide a new computational platform called Vermouth to automate topology generation, a crucial step that any biomolecular simulation starts with. Given a wide arrange of chemical structures that need to be simulated, varying qualities of structural models as inputs obtained from various sources, and diverse force fields and molecular dynamics engines employed for simulations, automation of this fundamental step is challenging, especially for complex systems and in case that there is a need to conduct high-throughput simulations in the application of computer-aided drug design (CADD). To overcome this challenge, the authors develop a programming library composed of components that carry out various types of fundamental functionalities that are commonly encountered in topological generation. These components are intended to be general for any type of molecules and not to depend on any specific force field and MD engines. To demonstrate the applicability of this library, the authors employ those components to re-assemble a pipeline called Martinize2 used in topology generation for simulations with a widely used coarse-grained model (CG) MARTINI. This pipeline can fully recapitulate the functionality of its original version Martinize but exhibit greatly enhanced generality, as confirmed by the ability of the pipeline to faithfully generate topologies for two high-complexity benchmarking sets of proteins.

Strengths:

The main strength of this work is the use of concepts and algorithms associated with induced subgraph in graph theory to automate several key but non-trivial steps of topology generation such as the identification of monomer residue units (MRU), the repair of input structures with missing atoms, the mapping of topologies between different resolutions, and the generation of parameters needed for describing interactions between MRUs.

Weaknesses:

Although the Vermouth library appears promising as a general tool for topology generation, there is insufficient information in the current manuscript and a lack of documentation that may allow users to easily apply this library. More detailed explanation of various classes such as Processor, Molecule, Mapping, ForceField etc. that are mentioned is still needed, including inputs, output and associated operations of these classes. Some simple demonstration of application of these classes would be of great help to users. The formats of internal databases used to describe reference structures and force fields may also need to be clarified. This is particularly important when the Vermouth needs to be adapted for other AA/CG force fields and other MD engines.

We thank the reviewer for pointing out the strengths of the presented work and agree that one of the current limitations is the lack of documentation about the library. In the revision, we point more clearly to the documentation page of the Vermouth library, which contains more detailed information on the various processors. The format of the internal databases has also been added to the documentation page. Providing a simple demonstration of applications of these classes is a great suggestion, however, we believe that it is more convenient to provide those in the form of code examples in the documentation or for instance jupyter notebooks rather than in the paper itself.  

The successful automation of the Vermouth relies on the reference structures that need to be pre-determined. In case of the study of 43 small ligands, the reference structures and corresponding mapping to MARTINIcompatible representations for all these ligands have been already defined in the M3 force field and added into the Vermouth library. However, the authors need to comment on the scenario where significantly more ligands need to be considered and other force fields need to be used as CG representations with a lack of reference structures and mapping schemes.

We acknowledge that vermouth/martinize2 is not capable of automatically generating Martini mappings or parameters on the fly for unknown structures that are not part of the database. However, this capability is not the purpose of the program, which is rather to distribute and manage existing parameters. Unlike atomistic force fields, which frequently have automated topology builders, Martini parameters are usually obtained for a set of specific molecules at a time and benchmarked accordingly. As more parameters are obtained by researchers, they can be added to the vermouth library via the GitHub interface in a controlled manner. This process allows the database to grow and in our opinion will quickly grow beyond the currently implemented parameters. Furthermore, the API of Vermouth is set up in a way that it can easily interface with automated topology builders which are currently being developed. Hence this limitation in our view does not diminish the applicability of vermouth to high-throughput applications with many ligands. The framework is existing and works, now only more parameters have to be added.

Reviewer #2 (Public Review):

Summary:

This manuscript by Kroon, Grunewald, Marrink and coworkers present the development of Vermouth library for coarse grain assignment and parameterization and an updated version of python script, the Martinize2 program, to build Martini coarse grained (CG) models, primarily for protein systems.

Strengths:

In contrast to many mature and widely used tools to build all-atom (AA) models, there are few well-accepted programs for CG model constructions and parameterization. The research reported in this manuscript is among the ongoing efforts to build such tools for Martini CG modeling, with a clear goal of high-throughput simulations of complex biomolecular systems and, ultimately, whole-cell simulations. Thus, this manuscript targets a practical problem in computational biophysics. The authors see such an effort to unify operations like CG mapping, parameterization, etc. as a vital step from the software engineering perspective.

Weaknesses:

However, the manuscript in this shape is unclear in the scientific novelty and appears incremental upon existing methods and tools. The only "validation" (more like an example application) is to create Martini models with two protein structure sets (I-TASSER and AlphaFold). The success rate in building the models was only 73%, while the significant failure is due to incomplete AA coordinates. This suggests a dependence on the input AA models, which makes the results less attractive for high-throughput applications (for example, preparation/creation of the AA models can become the bottleneck). There seems to be an improvement in considering the protonation state and chemical modification, but convincing validation is still needed. Besides, limitations in the existing Martini models remain (like the restricted dynamics due to the elastic network, the electrostatic interactions or polarizability).

We thank the reviewer for pointing out the strengths of the presented work, but respectfully disagree with the criticism that the presented work is only incremental upon existing methods and tools. All MD simulations of structured proteins regardless of the force field or resolution rely on a decent initial structure to produce valid results. Therefore, failure upon detection of malformed protein input structures is an essential feature for any high-throughput pipeline working with proteins, especially considering the computational cost of MD simulations. We note that programs such as the first version of Martinize generate reasonable-looking input parameters that lead to unphysical simulations and wasted CPU hours.

The alpha-fold database for which we surveyed 200,000 structures only contained 7 problematic structures, which means that the success rate was 99% for this database. This example simply shows that users potentially have to add the step of fixing atomistic protein input structures, if they seek to run a high-throughput pipeline.

But at least they can be assured that martinize2 will make sure to check that no issues persist.

Furthermore, we note that the manuscript does not aim to validate or improve the existing Martini (protein) models. All example cases presented in the paper are subject to the limitations of the protein models for the reason that martinize2 is only the program to generate those parameters. Future improvements in the protein model, which are currently underway, will immediately be available through the program to the broader community.  

Reviewer #3 (Public Review):

Summary:

The manuscript Kroon et al. described two algorithms, which when combined achieve high throughput automation of "martinizing" protein structures with selected protonation states and post-translational modifications.

Strengths:

A large scale protein simulation was attempted, showing strong evidence that authors' algorithms work smoothly.

The authors described the algorithms in detail and shared the open-source code under Apache 2.0 license on GitHub. This allows both reproducibility of extended usefulness within the field. These algorithms are potentially impactful if the authors can address some of the issues listed below.

We thank the reviewer for pointing out the strengths.  

Weaknesses:

One major caveat of the manuscript is that the authors claim their algorithms aim to "process any type of molecule or polymer, be it linear, cyclic, branched, or dendrimeric, and mixtures thereof" and "enable researchers to prepare simulation input files for arbitrary (bio)polymers". However, the examples provided by the manuscript only support one type of biopolymer, i.e. proteins. Despite the authors' recommendation of using polyply along with martinize2/vermouth, no concrete evidence has been provided to support the authors' claim. Therefore, the manuscript must be modified to either remove these claims or include new evidence.

We acknowledge that the current manuscript is largely protein-centric. To some extent this results from the legacy of martinize version 1, which was also only used for proteins. However, to show that martinize2 also works for cyclic as well as branched molecules we implemented two additional test cases and updated formerly Figure 6 and now Figure 7. Crown ether is used as an example of a cyclic molecule whereas a small branched polyethylene molecule is a test case for branching. Needless to say both molecules are neither proteins nor biomolecules. 

Method descriptions on Martinize2 and graph algorithms in SI should be core content of the manuscript. I argue that Figure S1 and Figure S2 are more important than Figure 3 (protonation state). I recommend the authors can make a workflow chart combining Figure S1 and S2 to explain Martinize2 and graph algorithms in main text.

The reviewer's critique is fair. Given the already rather large manuscript, we tried to strike a balance between describing benchmark test cases, some practical usage information (e.g. the Histidine modification), and the algorithmic library side of the program. In particular, we chose to add the figure on protonation state, because how to deal with protonation states—in particular, Histidines—was amongst the top three raised issues by users on our GitHub page. Due to this large community interest, we consider the figure equally important. However, we moved Figure S1 from the Supporting Information into the manuscript and annotated the already mentioned text with the corresponding panels to more clearly illustrate the underlying procedure. 

In Figure 3 (protonation state), the figure itself and the captions are ambiguous about whether at the end the residue is simply renamed from HIS to HIP, or if hydrogen is removed from HIP to recover HIS.

Using either of the two routes yields the same parameters in the end, which are for the protonated Histidine. In the second route, the extra hydrogen on Histidine is detected as an additional atom and therefore a different logic flow is triggered. Atoms are never removed, but only compounded to a base block plus modification atoms. We adjusted the figure caption to point this out more clearly.  

In "Incorporating a Ligand small-molecule Database", the authors are calling for a community effort to build a small-molecule database. Some guidance on when the current database/algorithm combination does or does not work will help the community in contributing.

Any small molecule not part of the database will not work. However, martinize2 will quickly identify if there are missing components of the system and alert the users. At that point, the users can decide to make their files, guided by the new documentation pages. 

A speed comparison is needed to compare Martinize2 and Martinize.

We respectfully disagree that a speed comparison is needed. We already alerted in the manuscript discussion that martinize2 is slower, since it does more checks, is more general, and does not only implement a single protein model.

eLife assessment

This important study of artificial selection in microbial communities shows that the possibility of selecting a desired fraction of slow and fast-growing types is impacted by their initial fractions. The evidence, which relies on mathematical analysis and simulations of a stochastic model, is convincing. It highlights the tension between selection at the strain and the community level. This study should be of interest to researchers interested in ecology, both theoretical and experimental.

Reviewer #1 (Public Review):

Summary:

The authors demonstrate with a simple stochastic model that the initial composition of the community is important in achieving a target frequency during the artificial selection of a community.

Strengths:

To my knowledge, the intra-collective selection during artificial selection has not been seriously theoretically considered. However, in many cases, the species dynamics during the incubation of each selection cycle are important and relevant to the outcome of the artificial selection experiment. Stochasticity from birth and death (demographic stochasticity) plays a big role in these species' abundance dynamics. This work uses a simple framework to tackle this idea meticulously.

This work may or may not be related to hysteresis (path dependency). If this is true, maybe it would be nice to have a discussion paragraph talking about how this may be the case. Then, this work would even attract the interest of people studying dynamic systems.

Weaknesses:

(1) Connecting structure and function

In typical artificial selection literature, most of them select the community based on collective function. Here in this paper, the authors are selecting a target composition. Although there is a schematic cartoon illustrating the relationship between collective function (y-axis) and the community composition in the main Figure 1, there is no explicit explanation or justification of what may be the origin of this relationship. I think giving the readers a naïve idea about how this structure-function relationship arises in the introduction section would help. This is because the conclusion of this paper is that the intra-collective selection makes it hard to artificially select a community that has an intermediate frequency of f (or s). If there is really evidence or theoretical derivation from this framework that indeed the highest function comes from the intermediate frequency of f, then the impact of this paper would increase because the conclusions of this stochastic model could allude to the reasons for the prevalent failures of artificial selection in literature.

(2) Explain intra-collective and inter-collective selection better for readers.

The abstract, the introduction, and the result section use these terms or intra-collective and inter-collective selection without much explanation. A clear definition in the beginning would help the audience grasp the importance of this paper, because these concepts are at the core of this work.

(3) Achievable target frequency strongly depending on the degree of demographic stochasticity.

I would expect that the experimentalists would find these results interesting and would want to consider these results during their artificial selection experiments. The main Figure 4 indicates that the Newborn size N0 is a very important factor to consider during the artificial selection experiment. This would be equivalent to how much bottleneck is imposed on the artificial selection process in every iteration step (i.e., the ratio of serial dilution experiment). However, with a low population size, all target frequencies can be achieved, and therefore in these regimes, the initial frequency now does not matter much. It would be great for the authors to provide what the N0 parameter actually means during the artificial selection experiments. Maybe relative to some other parameter in the model. I know this could be very hard. But without this, the main result of this paper (initial frequency matters) cannot be taken advantage of by the experimentalists.

(4) Consideration of environmental stochasticity.

The success (gold area of Figure 2d) in this framework mainly depends on the size of the demographic stochasticity (birth-only model) during the intra-collective selection. However, during experiments, a lot of environmental stochasticity appears to be occurring during artificial selection. This may be out of the scope of this study. But it would definitely be exciting to see how much environmental stochasticity relative to the demographic stochasticity (variation in the Gaussian distribution of F and S) matters in succeeding in achieving the target composition from artificial selection.

(5) Assumption about mutation rates

If setting the mutation rates to zero does not change the result of the simulations and the conclusion, what is the purpose of having the mutation rates \mu? Also, is the unidirectional (S -> F -> FF) mutation realistic? I didn't quite understand how the mutations could fit into the story of this paper.

(6) Minor points

In Figure 3b, it is not clear to me how the frequency difference for the Intra-collective and the Inter-collective selection is computed.

In Figure 5b, the gold region (success) near the FF is not visible. Maybe increase the size of the figure or have an inset for zoom-in. Why is the region not as big as the bottom gold region?

Reviewer #2 (Public Review):

The authors provide an analytical framework to model the artificial selection of the composition of communities comprised of strains growing at different rates. Their approach takes into account the competition between the targeted selection at the level of the meta-community and the selection that automatically favors fast-growing cells within each replicate community. Their main finding is a tipping point or path-dependence effect, whereby compositions dominated by slow-growing types can only be reached by community-level selection if the community does not start and never crosses into a range of compositions dominated by fast growers during the dynamics.

These results seem to us both technically correct and interesting. We commend the authors on their efforts to make their work reproducible even when it comes to calculations via extensive appendices, though perhaps a table of contents and a short description of these appendices at the start of SI would help navigate them.

The main limitation in the current form of the article is that it could clarify how its assumptions and findings differ from and improve upon the rest of the literature:

- Many studies discuss the interplay between community-level evolution and species- or strain-level evolution. But "evolution" can be a mix of various forces, including selection, drift/randomness, and mutation/innovation.

- This work's specificity is that it focuses strictly on constant community-level selection versus constant strain-level selection, all other forces being negligible (neither stochasticity nor innovation/mutation matter at either level, as we try to clarify now).

- Regarding constant community-level selection, it is only briefly noted that "once a target frequency is achieved, inter-collective selection is always required to maintain that frequency due to the fitness difference between the two types" [pg. 3 {section sign}2]. In other words, action from the selector is required indefinitely to maintain the community in the desired state. This assumption is found in a fraction of the literature, but is still worth clarifying from the start as it can inform the practical applicability of the results.

- More importantly, strain-level evolution also boils down here to pure selection with a constant target, which is less usual in the relevant literature. Here, (1) drift from limited population sizes is very small, with no meaningful counterbalancing of selection, (2) pure exponential regime with constant fitness, no interactions, no density- or frequency-dependence, (3) there is no innovation in the sense that available types are unchanging through time (no evolution of traits such as growth rate or interactions) and (4) all the results presented seem unchanged when mutation rate mu = 0 (as noted in Appendix III), meaning that the conclusions are not "about" mutation in any meaningful way.

- Furthermore, the choice of mutation mechanism is peculiar, as it happens only from slow to fast grower: more commonly, one assumes random non-directional mutations, rather than purely directional ones from less fit to fitter (which is more of a "Lamarckian" idea). Given that mutation does not seem to matter here, this choice might create unnecessary opposition from some readers or could be considered as just one possibility among others.

It would be helpful to have all these points stated clearly so that it becomes easy to see where this article stands in an abundant literature and contributes to our understanding of multi-level evolution, and why it may have different conclusions or focus than others tackling very similar questions.

Finally, a microbial context is given to the study, but the assumptions and results are in no way truly tied to that context, so it should be clear that this is just for flavor.

Reviewer #3 (Public Review):

The authors address the process of community evolution under collective-level selection for a prescribed community composition. They mostly consider communities composed of two types that reproduce at different rates, and that can mutate one into the other. Due to such differences in 'fitness' and to the absence of density dependence, within-collective selection is expected to always favour the fastest grower, but the collective-level selection can oppose this tendency, to a certain extent at least. By approximating the stochastic within-generation dynamics and solving it analytically, the authors show that not only high frequencies of fast growers can be reproducibly achieved, aligned with their fitness advantage. Small target frequencies can also be maintained, provided that the initial proportion of fast growers is sufficiently small. In this regime, similar to the 'stochastic corrector' model, variation upon which selection acts is maintained by a combination of demographic stochasticity and of sampling at reproduction. These two regions of achievable target compositions are separated by a gap, encompassing intermediate frequencies that are only achievable when the bottleneck size is small enough or the number of communities is (disproportionately) larger.

A similar conclusion, that stochastic fluctuations can maintain the system over evolutionary time far from the prevalence of the faster-growing type, is then confirmed by analyzing a three-species community, suggesting that the qualitative conclusions of this study are generalizable to more complex communities.

I expect that these results will be of broad interest to the community of researchers who strive to improve community-level selection, but are often limited to numerical explorations, with prohibitive costs for a full characterization of the parameter space of such embedded populations. The realization that not all target collective functions can be as easily achieved and that they should be adapted to the initial conditions and the selection protocol is also a sobering message for designing concrete applications.

A major strength of this work is that the qualitative behaviour of the system is captured by an analytically solvable approximation so that the extent of the 'forbidden region' can be directly and generically related to the parameters of the selection protocol.

I however found the description of the results too succinct and I think that more could be done to unpack the mathematical results in a way that is understandable to a broader audience. Moreover, the phenomenon the authors characterize is of purely ecological nature. Here, mutations of the growth rate are, in my understanding, neither necessary (non-trivial equilibria can be maintained also when \mu =0) nor sufficient (community-level selection is necessary to keep the system far from the absorbing state) for the phenomenon described. Calling this dynamics community evolution reflects a widespread ambiguity, and is not ascribable just to this work. I find that here the authors have the opportunity to make their message clearer by focusing on the case where the 'mutation' rate \mu vanishes (Equations 39 & 40 of the SI) - which is more easily interpretable, at least in some limits - while they may leave the more general equations 3 & 4 in the SI. Combined with an analysis of the deterministic equations, that capture the possibility of maintaining high frequencies of fast growers, the authors could elucidate the dynamics that are induced by the presence of a second level of selection, and speculate on what would be the result of real open-ended evolution (not encompassed by the simple 'switch mutations' generally considered in evolutionary game theory), for instance discussing the invasibility (or not) of mutant types with slightly different growth rates.

The single most important model hypothesis that I would have liked to be discussed further is that the two types do not interact. Species interactions are not only essential to achieve inheritance of composition in the course of evolution but are generally expected to play a key role even on ecological time scales. I hope the authors plan to look at this in future work.

eLife assessment

This important study implicates Sempharon 4a in both mice and humans as a key suppressor of psoriatic inflammation. The data are in parts incomplete in defining the precise functionally relevant cellular source and mechanism. Nonetheless, this study brings new insight into psoriasis pathogenesis and a potential new therapeutic target.

Reviewer #1 (Public Review):

Summary:

In this study, Kume et al examined the role of the protein Semaphorin 4a in steady-state skin homeostasis and how this relates to skin changes seen in human psoriasis and imiquimod-induced psoriasis-like disease in mice. The authors found that human psoriatic skin has reduced expression of Sema4a in the epidermis. While Sema4a has been shown to drive inflammatory activation in different immune populations, this finding suggested Sema4a might be important for negatively regulating Th17 inflammation in the skin. The authors go on to show that Sema4a knockout mice have skin changes in key keratinocyte genes, increased gdT cells, and increased IL-17 similar to differences seen in non-lesional psoriatic skin, and that bone marrow chimera mice with WT immune cells and Sema4a KO stromal cells develop worse IMQ-induced psoriasis-like disease, further linking expression of Sema4a in the skin to maintaining skin homeostasis. The authors next studied downstream pathways that might mediate the homeostatic effects of Sema4a, focusing on mTOR given its known role in keratinocyte function. As with the immune phenotypes, Sema4a KO mice had increased mTOR activation in the epidermis in a similar pattern to mTOR activation noted in non-lesional psoriatic skin. The authors next targeted the mTOR pathway and showed rapamycin could reverse some of the psoriasis-like skin changes in Sema4a KO mice, confirming the role of increased mTOR in contributing to the observed skin phenotype.

Strengths:

The most interesting finding is the tissue-specific role for Sema4a, where it has previously been considered to play a mostly pro-inflammatory role in immune cells, this study shows that when expressed by keratinocytes, Sema4a plays a homeostatic role that when missing leads to the development of psoriasis-like skin changes. This has important implications in terms of targeting Sema4a pharmacologically. It also may yield a novel mouse model to study mechanisms of psoriasis development in mice separate from the commonly used IMQ model. The included experiments are well-controlled and executed rigorously.

Weaknesses:

A weakness of the study is the lack of tissue-specific Sema4a knockout mice (e.g. in keratinocytes only). The authors did use bone marrow chimeras, but only in one experiment. This work implies that psoriasis may represent a Sema4a-deficient state in the epidermal cells, while the same might not be true for immune cells. Indeed, in their analysis of non-lesional psoriasis skin, Sema4a was not significantly decreased compared to control skin, possibly due to compensatory increased Sema4a from other cell types. Unbiased RNA-seq of Sema4a KO mouse skin for comparison to non-lesional skin might identify other similarities besides mTOR signaling. Indeed, targeting mTOR with rapamycin reveres some of the skin changes in Sema4a KO mice, but not skin thickness, so other pathways impacted by Sema4a may be better targets if they could be identified. Utilizing WTKO chimeras in addition to global KO mice in the experiments in Figures 6-8 would more strongly implicate the separate role of Sema4a in skin vs immune cell populations and might more closely mimic non-lesional psoriasis skin.

Reviewer #2 (Public Review):

Summary:

Kume et al. found for the first time that Semaphorin 4A (Sema4A) was downregulated in both mRNA and protein levels in L and NL keratinocytes of psoriasis patients compared to control keratinocytes. In peripheral blood, they found that Sema4A is not only expressed in keratinocytes but is also upregulated in hematopoietic cells such as lymphocytes and monocytes in the blood of psoriasis patients. They investigated how the down-regulation of Sema4A expression in psoriatic epidermal cells affects the immunological inflammation of psoriasis by using a psoriasis mice model in which Sema4A KO mice were treated with IMQ. Kume et al. hypothesized that down-regulation of Sema4A expression in keratinocytes might be responsible for the augmentation of psoriasis inflammation. Using bone marrow chimeric mice, Kume et al. showed that KO of Sema4A in non-hematopoietic cells was responsible for the enhanced inflammation in psoriasis. The expression of CCL20, TNF, IL-17, and mTOR was upregulated in the Sema4AKO epidermis compared to the WT epidermis, and the infiltration of IL-17-producing T cells was also enhanced.

Strengths:

Decreased Sema4A expression may be involved in psoriasis exacerbation through epidermal proliferation and enhanced infiltration of Th17 cells, which helps understand psoriasis immunopathogenesis.

Weaknesses:

The mechanism by which decreased Sema4A expression may exacerbate psoriasis is unclear as yet.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Summary: 

This study investigated the role of CD47 and TSP1 in extramedullary erythropoiesis by utilization of both global CD47-/- mice and TSP1-/- mice. 

Strengths:  

Flow cytometry combined with spleen bulk and single-cell transcriptomics were employed. The authors found that stress-induced erythropoiesis markers were increased in CD47-/- spleen cells, particularly genes that are required for terminal erythroid differentiation. Moreover, CD47 dependent erythroid precursors population was identified by spleen scRNA sequencing. In contrast, the same cells were not detected in TSP1-/- spleen. These findings provide strong evidence to support the conclusion that the differential role of CD47 and TSP1 in extramedullary erythropoiesis in mouse spleen. 

Weaknesses: 

Methods and data analysis are appropriate. However, some clarifications are required. The discussion section needs to be expanded.  

(1) The sex of mice that were used in the study is unknown.  

(2) In the method of Single-cell RNA sequencing (page 10), it mentioned that single cell suspensions from mouse spleens were depleted of all mature hematopoietic cell lineages by passing through CD8a microbeads and CD8a+ T cell isolation Kit. As described, it is confusing what cell types are obtained for performing scRNAseq. More information is required for clarity.  

(3) The constitutive CD47 knockout mouse model is utilized in this study. The observed accumulation of erythroid precursors in the spleens of CD47-/- mice suggests a chronic effect of CD47 on spleen function. Can the current findings be extrapolated to acute scenarios involving CD47 knockdown or loss, as this may have more direct relevance to the potential side effects associated with an-CD47-mediated cancer therapy? Please expand on this topic in the discussion section.  

(1) The missing mouse gender information is incorporated into the revised manuscript. For flow cytometry, two male and two female mice of each genotype were used. For single cell RNA sequencing, two female and one male mouse of each genotype were used. For the bulk RNA sequencing four male cd47−/− mice and four male wildtype mice were used.

(2) We apologize for the confusing presentation, which has been corrected. The bulk RNA sequencing analysis identified elevated expression of erythropoietic genes in CD8+ spleen cells from cd47−/− versus wildtype mice that were obtained using magnetic bead depletion of all other lineages. Therefore, we used the same Miltenyi negative selection kit as the first step to prepare the cells for single cell RNA sequencing. These untouched cells were then depleted of most mature CD8 T cells using a Miltenyi CD8a(Ly2) antibody positive selection kit. An important consideration underlying this approach was recognizing that the commercial magnetic bead depletion kits used for preparing specific immune cell types are optimized to give relatively pure populations of the intended immune cells using wildtype mice. Our previous experience studying NK cell development in the cd47−/− mice taught us that NK precursors, which are rare in wildtype mouse spleens, accumulate in cd47−/− spleens and were not removed by the antibody cocktail optimized for wildtype spleen cells (Nath et al Front Immunol 2018). The present data indicate that erythroid precursors behave similarly.

(3) The Discussion was edited as recommended. Anemia is a prevalent side effect of several CD47 therapeutic antibodies being developed for cancer therapy. This anemia would be expected to induce erythropoiesis in bone marrow and possibly at extramedullary sites. Human spleen cells are not accessible to directly evaluate extramedullary erythropoiesis in cancer patients, but analysis of circulating erythroid precursors or liquid biopsy methods could be useful to detect induction of extramedullary erythropoiesis by these therapeutics. We are currently investigating the ability of CD47 antibodies to directly induce erythropoiesis using a human in vitro model.

Reviewer #2 (Public Review):

Summary: 

The authors used existing mouse models to compare the effects of ablating the CD47 receptor and its signaling ligand Thrombospondin. The CD47-KO model used in this study was generated by Kim et al, 2018, where hemolytic anemia and splenomegaly was reported. This study analyzes the cell composition of the spleens from CD47-KO and Thsp-KO, focusing on early hematopoietic and erythroid populations. The data broadly shows that splenomegaly in the CD47-KO is largely due to an increase in committed erythroid progenitors as seen by Flow Cytometry and single-cell sequencing, whereas the Thsp-KO shows a slight depletion of committed erythroid progenitors but is otherwise similar to WT in splenic cell composition.  

Strengths:

The techniques used are appropriate for the study and the data support the main conclusions of the study. This study provides novel insights into a putative role of Thsp-CD47 signaling in triggering definitive erythropoiesis in the mouse spleen in response to anemic stress and constitutes a good resource for researchers seeking to understand extramedullary erythropoiesis.  

Weaknesses:

The Flow cytometry data alone supports the authors' main conclusion and single-cell sequencing confirms them but does not add further information, other than those already observed in the Flow data. The single-cell sequencing analysis and presentation could be improved by using alternate clustering methods as well as separating the data by genotype and displaying them in order for readers to fully grasp the nuanced differences in marker expression between the genotypes. Further, it is not clear from the authors' description of their results whether the increased splenic erythropoiesis is a direct consequence of CD47-KO or a response to the anemic stress in this mouse model. The enrichment of cKit+ Ter119+ Sca1- cells in CD47-KO indicates that these are likely stress erythroid progenitors. Another CD47-KO mouse model (Lindberg et al 1996) has no reported erythroid defects and was also not examined in this study.  

(1) The reviewer asked, “whether the increased splenic erythropoiesis is a direct consequence of CD47-KO or a response to the anemic stress in this mouse model.” Our data supports both a direct role for CD47 and an indirect role resulting from the response to anemic stress. We cited our previous publications describing increased Sox2+ stem cells in spleens of Cd47 and Thbs1 knockout mice, but we neglected to emphasize another study where we found that bone marrow from cd47−/− mice subjected to the stress of ionizing radiation exhibited more colony forming units for erythroid (CFU-E) and burst-forming unit-erythroid (BFU-E) progenitors compared to bone marrow from irradiated wildtype mice (Maxhimer Sci Transl Med 2009). Taken together, our published data demonstrates that loss of CD47 results in an intrinsic protection of hematopoietic stem cells from genotoxic stress. This function of CD47 is thrombospondin-1-dependent and is consistent with the up-regulation of early erythroid precursors in the spleens of both knockout mice but cannot explain why the Thbs1−/−  mice have fewer committed erythroid precursors than wildtype. We cited studies that documented increased red cell turnover in cd47−/− mice but less red cell turnover in Thbs1−/−  mice compared to wildtype mice. Increased red cell clearance in cd47−/− mice is mediated by loss of the “don’t eat me” function of CD47 on red cells. In wildtype mice, clearance is augmented by thrombospondin-1 binding to the clustered CD47 on aging red cells (Wang, Aging Cell 2020). Thus, anemic stress in the mouse strains studied here decreases in the order cd47−/− > WT > Thbs−/−. This is consistent with the increased committed erythroid progenitors reported here in cd47−/− spleens and decreased committed progenitors in the Thbs1−/− spleens. 

(2) Based on the reviewer’s question regarding alternative mechanisms and the publication of Yang et al 2022 identifying a role for CD47 in stress erythropoiesis though transfer of mitochondria to erythroblasts, we asked whether cd47-/- erythroid precursors  would show decreased mRNA expression for mitochondrial chromosome genes (new Figure 4−figure supplement 3C). Some of these mRNAs were more abundant in cd47-/- and thbs1-/- erythroid cells, which is the opposite of what we expected based on Yang 2022 but consistent with our previous publications identifying thrombospondin-1 and CD47 as negative regulators of mitochondrial homeostasis in muscle cells and T cells.

(3) The cd47−/− mice used for the current study are the same strain as those reported by Lindberg et al in 1996, with additional backcrossing onto a C57BL/6 background.

Recommendations For The Authors:

Reviewer #2 (Recommendations For The Authors):

Suggestions for improved or additional experiments, data, or analyses.  

Significant efforts went into analyzing the type of erythroid progenitors by marker expression, but typical Flow cytometry strategies using Ter119 and CD44 combined with forward scatter can be used to stage the committed erythroid progenitors precisely.  

We appreciate this suggestion to extend the flow data. However, the upcoming retirement of the PI required closing our breeding colony, and the mice are no longer available.  

How can the difference between the erythroid phenotypes of the Lindberg et al 1996 CD47-KO (exon2 Neo knock-in) and Kim et al 2018 CD47-ko (exon1 26bp indel) be explained?  

We are not convinced that the erythroid phenotypes of the Lindberg and Kim CD47-KO mice differ at the age used in our studies. Kim et al. focused on progressive hemolytic anemia and changes in T cells in spleen that emerge at 26 weeks age, whereas the mice used here were younger. The Lindberg and Kim mice have similar spleen enlargement at the age we used.

Another manuscript under review from our lab suggests that cis-regulation of an adjacent colinear gene could contribute to some phenotypes observed when perturbing the Cd47 gene. The Lindberg mouse exhibits minimal perturbation of that adjacent gene, but we have no data regarding the Kim et al mouse. The reviewer’s question brought to our attention that we neglected to state in the Methods that the mice used here are the Lindberg mice, not the Kim mice. This omission is now corrected.

The authors used Lindberg mouse for 2018 study on NK cells and observed splenomegaly. Did they check for extramedullary erythropoiesis there?  

Retrospective examination of the RNAseq data for the spleen cells enriched in NK precursors used in our 2018 publication (Nath, 2018) reveals significantly elevated expression for a majority of the extramedullary erythroid markers listed in Table 1, but they were generally less abundant than observed for the lineage-depleted spleen cells used in the present manuscript.   

Author response table 1.

To clarify the stress erythropoiesis issue, it might be helpful to examine the sc-seq data for the expression of specific stress erythropoiesis markers in CD47-KO. Targets of BMP4 and Hedgehog signaling can also be examined. Further colony assays can help determine if stress BFU-Es are prevalent in the CD47-KO spleens and depleted in Thsp-KO  

As noted in Table 1, twelve of the genes we studied are established markers of stress-induced extramedullary erythropoiesis, and most of these were included in the scRNA seq data presented. Our previous publication demonstrated that bone marrow from cd47−/− mice subjected to the stress of ionizing radiation exhibited more colony forming units for erythroid (CFU-E) and burst-forming unit-erythroid (BFU-E) progenitors compared to bone marrow from irradiated wildtype mice (Maxhimer Sci Transl Med 2009). We have not performed colony formation assays using spleen.

To address the reviewer’s question regarding BMP4 and hedgehog signaling we performed gene set enrichment analysis for known BMP4 and hedgehog signaling signatures. Using GSE26351_UNSTIM_VS_BMP_PATHWAY_STIM_HEMATOPOIETIC_PROGENITORS, cd47-/- cells in cluster 12 or their CD34+ orCD34- subsets did not show significant enrichment for BMP4 targets compared to WT. Thbs1-/- cells in clusters 12 and 14 showed marginally significant depletion of the BMP4 signature (p=0.04 and p=0.023, respectively). Using the KEGG_HEDGEHOG_SIGNALING_PATHWAY, we did not find any significant enrichment. However, only a few genes in this pathway were detectable in the scRNAseq data. These data suggest that the BMP4 signaling may be regulated by thrombospondin-1, but properly testing this hypothesis would require achieving greater sequencing depth combined with a cell isolation method that better enriches the early hematopoietic progenitors that are known to utilize the BMP4 pathway.

In the reclustering of erythroid progenitors in Figure 5, inclusion of Gata1 as a selection marker may help capture more of the early erythroid progenitors from the dataset and provide a more complete picture of the erythroid populations. 

We thank the reviewer for suggesting inclusion of Gata1. We repeated the reclustering including Gata1 and found the selected cell count increased from 876 cells to 1007 cells. However, most of the increase was not in the erythroid cluster, which increased from 413 cells to 419 cells. Most of the increase represented Gata1+ T cells (548 cells including Gata1 versus 463 cells without). The revised manuscript presents genotype-dependent differential gene expression based on including Gata1 selection, but none of the specific conclusions were changed from the initial submission. The new Table 4 and Figure 7−figure supplement 1 enabled us to compare differential expression of erythropoietic genes obtained using supervised and unsupervised clustering and show that both methods yield comparable results.

Just out of curiosity, was there an attempt to make a CD47 Thsp double KO? . Is it viable?  

Cd47 KO mice are somewhat difficult breeders, and several previous attempts to cross with other transgenics have produced viable homozygous offspring that could not be propagated.

Recommendations for improving the wring and presentation.  

Perhaps readers would find it more intriguing if the paper led with the single-cell sequencing showing enrichment of erythroid populations in CD47-KO, and later confirmed with Flow Cytometry (even if this was not necessarily the order in which the experiments were done). 

We considered this suggestion but believe that some of the flow cytometry data is needed to understand why we focused on CD34+ and CD34- subsets and proliferation markers when analyzing the scRNAseq data

The single-cell sequencing data in Figure 3 might benefit from UMAP clustering as well. In addition, it would greatly help readers if the data points were separated by genotype and displayed after clustering. A similar analysis has been done in this paper: doi:10.1038/s41556-022-00898-9 by clustering different conditions together but displaying them separately by condition. 

We initially explored tSNE and UMAP clustering and obtained similar results. We have added violin plots separated by genotype in Figure 4-figure supplement 2. We also included improved clusters separated by genotype in the revised Figure 3 panels C and D and for the reclustering in Figure 6D. UMAP plots provided better presentation for the reclustering (revised Figure 7). All data have been updated to the latest pipeline as noted in the Methods.

Minor corrections to the text and figures.  

Figure 4: Labels and plot legends are illegible in general, please relabel manually and if possible, redo plots with bigger font size and legends (relatively easy using ggplot2) 

All figure panels were relabeled using larger fonts

Figure 5D: Individual plots are stacked randomly atop each other and in many cases, gene names are not visible. Please restack the layers and ensure that the gene names are visible 

Panel D was made a separate figure with enlarged labels (now Figure 7).

Supp Fig 2: Layout can be organized a little better. Consider splitting into two figures for better organization  

The figure was split as recommended. Now Figure 1-figure supplement 2 and Figure 2-figure supplement

1.

Abstract Line 10: "...mRNA expression of Kit, Ermap, and Tfrc, Induction of committed erythroid precursors is...". Replace comma after "Tfrc" with period   

Done.

Discussion Page 9 Line 8: "...WT spleens, s. mRNAs for some markers of committed erythroid cells including Nr3c1 mRNA...". Remove ", s" after spleens.   

Done.

eLife assessment

This study presents a valuable finding on the cell composition in mouse spleen depleted for the CD47 receptor and its signaling ligand Thrombospondin in hematopoietic differentiation. The supporting evidence is convincing with analytical improvements on the individual contributions of the signaling components and with functional studies. This work has implications for the role of CD47/Thsp in extramedullary erythropoiesis in mouse spleen and will be of interest to medical biologists working on cell signaling, transfusion medicine, and cell therapy.

Reviewer #1 (Public Review):

Summary:

This study investigated the role of CD47 and TSP1 in extramedullary erythropoiesis by utilization of both global CD47-/- mice and TSP1-/- mice.

Strengths:

Flow cytometry combined with spleen bulk and single cell transcriptomics were employed. The authors found that stress-induced erythropoiesis markers were increased in CD47-/- spleen cells, particularly genes that are required for terminal erythroid differentiation. Moreover, CD47 dependent erythroid precursors population was identified by spleen scRNA sequencing. In contrast, the same cells were not detected in TSP1-/- spleen. These findings provide strong evidence to support the conclusion that differential role of CD47 and TSP1 in extramedullary erythropoiesis in mouse spleen. Furthermore, the relevance of the current finding to the prevalent side effect (anemia) of anti-CD47 mediated cancer therapy has been discussed in the Discussion section.

Reviewer #3 (Public Review):

The authors used existing mouse models to compare the effects of ablating the CD47 receptor and its signaling ligand Thrombospondin. They analyze the cell composition of the spleens from CD47-KO and Thsp-KO using Flow Cytometry and single cell sequencing and focus mostly on early hematopoietic and erythroid populations. The data broadly shows that splenomegaly in the CD47-KO is largely due to an increase in committed erythroid progenitors, whereas the Thsp-KO shows a slight depletion of committed erythroid progenitors but is otherwise similar to WT in splenic cell composition. Thus, both their datasets supports the main conclusions of the study. One caveat of the single-cell dataset is that, insofar as the authors have explored and presented it, a clear picture of the mechanism driving extra medullary erythropoiesis in CD47-KO is lacking. This would be extremely valuable since one of the stated translational implications of this study is to assess and remedy the anemia caused by anti-CD47 therapy used in subtypes of AML. Nevertheless, this study provides novel insights into a putative role of Thsp-CD47 signaling in triggering definitive erythropoiesis in the mouse spleen in response to anemic stress and constitutes a good resource for researchers seeking to understand extramedullary erythropoiesis. This study also has generated data that will enable exploration of the possible adverse effects of using anti-CD47 therapies to treat AML.

eLife assessment

This valuable study describes a new type of NAD+ and Zn2+-independent protein lysine deacetylase in prokaryotes. These results extend the understanding of regulatory mechanisms related to bacterial lysine acetylation modifications however, the experimental evidence is incomplete and does not fully support the conclusions made. The work will be of interest to microbiologists studying metabolism and post-translational modifications.

Reviewer #1 (Public Review):

Summary:

This study by Wang et al. identifies a new type of deacetylase, CobQ, in Aeromonas hydrophila. Notably, the identification of this deacetylase reveals a lack of homology with eukaryotic counterparts, thus underscoring its unique evolutionary trajectory within the bacterial domain.

Strengths:

The manuscript convincingly illustrates CobQ's deacetylase activity through robust in vitro experiments, establishing its distinctiveness from known prokaryotic deacetylases. Additionally, the authors elucidate CobQ's potential cooperation with other deacetylases in vivo to regulate bacterial cellular processes. Furthermore, the study highlights CobQ's significance in the regulation of acetylation within prokaryotic cells.

Weaknesses:

While the manuscript is generally well-structured, some clarification and some minor corrections are needed.

Reviewer #2 (Public Review):

In recent years, lots of researchers have tried to explore the existence of new acetyltransferase and deacetylase by using specific antibody enrichment technologies and high-resolution mass spectrometry. This study adds to this effort. The authors studied a novel Zn2+- and NAD+-independent KDAC protein, AhCobQ, in Aeromonas hydrophila. They studied the biological function of AhCobQ by using a biochemistry method and used MS identification technology to confirm it. The results extend our understanding of the regulatory mechanism of bacterial lysine acetylation modifications. However, I find their conclusion to be a little speculative, and unfortunately, it also doesn't totally support the conclusion that the authors provided. In addition, regarding the figure arrangement, lots of the supplementary figures are not mentioned, and tables are not all placed in context.

Major concerns:

-In the opinion of this reviewer, is a little arbitrary to come to the title "Aeromonas hydrophila CobQ is a new type of NAD+- and Zn2+-independent protein lysine deacetylase in prokaryotes." This should be modified to delete the "in the prokaryotes", unless the authors get new or more evidence in the other prokaryotes for the existence of the AhCobQ.

-I was confused about the arrangement of the supplementary results. There are no citations for Figures S9-S19.

-No data are included for Tables S1-S6.

-The load control is not all integrated. All of the load controls with whole PAGE gel or whole membrane western blot results should be provided. Without these whole results, it is not convincing to come to the conclusion that the authors have.

-The materials & methods section should be thoroughly reviewed. It is unclear to me what exactly the authors are describing in the method. All the experimental designs and protocols should be described in detail, including growth conditions, assay conditions, purification conditions, etc.

-Relevant information should be included about the experiments performed in the figure legends, such as experimental conditions, replicates, etc. Often it is not clear what was done based on the figure legend description.

Reviewer #3 (Public Review):

Summary:

This study reports on a novel NAD+ and Zn2+-independent protein lysine deacetylase (KDAC) in Aeromonas hydrophila, termed AhCobQ (AHA_1389). This protein is annotated as a CobQ/CobB/MinD/ParA family protein and does not show similarity with known NAD+-dependent or Zn2+-dependent KDACs. The authors show that AhCobQ has NAD+ and Zn2+-independent deacetylase activity with acetylated BSA by western blot and MS analyses. They also provide evidence that the 195-245 aa region of AhCobQ is responsible for the deacetylase activity, which is conserved in some marine prokaryotes and has no similarity with eukaryotic proteins. They identified target proteins of AhCobQ deacetylase by proteomic analysis and verified the deacetylase activity using site-specific acetyllysine-incorporated target proteins. Finally, they show that AhCobQ activates isocitrate dehydrogenase by deacetylation at K388.

Strengths:

The finding of a new type of KDAC has a valuable impact on the field of protein acetylation. The characters (NAD+ and Zn2+-independent deacetylase activity in an unknown domain) shown in this study are very unexpected.

Weaknesses:

(1) As the characters of AhCobQ are very unexpected, to convince readers, MSMS data would be needed to exactly detect deacetylation at the target site in deacetylase activity assays. The authors show the MSMS data in assays with acetylated BSA, but other assays only rely on western blot.

(2) They prepared site-specific Kac proteins and used them in deacetylase activity assays. The incorporation of acetyllysine at the target site needs to be confirmed by MSMS and shown as supplementary data.

(3) The authors imply that the 195-245 aa region of AhCobQ may represent a new domain responsible for deacetylase activity. The feature of the region would be of interest but is not sufficiently described in Figure 5. The amino acid sequence alignments with representative proteins with conserved residues would be informative. It would be also informative if the modeled structure predicted by AlphaFold is shown and the structural similarity with known deacetylases is discussed.

eLife assessment

This paper reports a large drug repurposing screen based on an in vitro culture platform to identify compounds that can kill Plasmodium hypnozoites. This valuable work adds to the current repertoire of anti-hypnozoites agents and uncovers targetable epigenetic pathways to enhance our understanding of this mysterious stage of the Plasmodium life cycle. The data presented here are based on solid methodology and represent a starting point for further investigation of epigenetic inhibitors to treat P. vivax infection. This paper will be of interest to Plasmodium researchers and more broadly to readers in the fields of host-pathogen interactions and drug development.

Reviewer #1 (Public Review):

Summary:

Plasmodium vivax can persist in the liver of infected individuals in the form of dormant hypnozoites, which cause malaria relapses and are resistant to most current antimalarial drugs. This highlights the need to develop new drugs active against hypnozoites that could be used for radical cure. Here, the authors capitalize on an in vitro culture system based on primary human hepatocytes infected with P. vivax sporozoites to screen libraries of repurposed molecules and compounds acting on epigenetic pathways. They identified a number of hits, including hydrazinophthalazine analogs. They propose that some of these compounds may act on epigenetic pathways potentially involved in parasite quiescence. To provide some support to this hypothesis, they document DNA methylation of parasite DNA based on 5-methylcytosine immunostaining, mass spectrometry, and bisulfite sequencing.

Strengths:<br /> -The drug screen itself represents a huge amount of work and, given the complexity of the experimental model, is a tour de force.<br /> -The screening was performed in two different laboratories, with a third laboratory being involved in the confirmation of some of the hits, providing strong support that the results were reproducible.<br /> -The screening of repurposing libraries is highly relevant to accelerate the development of new radical cure strategies.

Weaknesses:

-The manuscript is composed of two main parts, the drug screening itself and the description of DNA methylation in Plasmodium pre-erythrocytic stages. Unfortunately, these two parts are loosely connected. First, there is no evidence that the identified hits kill hypnozoites via epigenetic mechanisms. The hit compounds almost all act on schizonts in addition to hypnozoites, therefore it is unlikely that they target quiescence-specific pathways. At least one compound, colforsin, seems to selectively act on hypnozoites, but this observation still requires confirmation. Second, while the description of DNA methylation is per se interesting, its role in quiescence is not directly addressed here. Again, this is clearly not a specific feature of hypnozoites as it is also observed in P. vivax and P. cynomolgi hepatic schizonts and in P. falciparum blood stages. Therefore, the link between DNA methylation and hypnozoite formation is unclear. In addition, DNA methylation in sporozoites may not reflect epigenetic regulation occurring in the subsequent liver stages.

-The mode of action of the hit compounds remains unknown. In particular, it is not clear whether the drugs act on the parasite or on the host cell. Merely counting host cell nuclei to evaluate the toxicity of the compounds is probably acceptable for the screen but may not be sufficient to rule out an effect on the host cell. A more thorough characterization of the toxicity of the selected hit compounds is required.

-There is no convincing explanation for the differences observed between P. vivax and P. cynomolgi. The authors question the relevance of the simian model but the discrepancy could also be due to the P. vivax in vitro platform they used.

-Many experiments were performed only once, not only during the screen (where most compounds were apparently tested in a single well) but also in other experiments. The quality of the data would be increased with more replication.

-While the extended assay (12 days versus 8 days) represents an improvement of the screen, the relevance of adding inhibitors of core cytochrome activity is less clear, as under these conditions the culture system deviates from physiological conditions.

Reviewer #2 (Public Review):

Summary:

In this manuscript, inhibitors of the P. vivax liver stages are identified from the Repurposing, Focused Rescue, and Accelerated Medchem (ReFRAME) library as well as a 773-member collection of epigenetic inhibitors. This study led to the discovery that epigenetics pathway inhibitors are selectively active against P. vivax and P. cynomolgi hypnozoites. Several inhibitors of histone post-translational modifications were found among the hits and genomic DNA methylation mapping revealed the modification on most genes. Experiments were completed to show that the level of methylation upstream of the gene (promoter or first exon) may impact gene expression. With the limited number of small molecules that act against hypnozoites, this work is critically important for future drug leads. Additionally, the authors gleaned biological insights from their molecules to advance the current understanding of essential molecular processes during this elusive parasite stage.

Strengths:<br /> -This is a tremendously impactful study that assesses molecules for the ability to inhibit Plasmodium hypnozoites. The comparison of various species is especially relevant for probing biological processes and advancing drug leads.

-The SI is wonderfully organized and includes relevant data/details. These results will inspire numerous studies beyond the current work.

Reviewer #3 (Public Review):

Although this work represents a massive screening effort to find new drugs targeting P. vivax hypnozoites, the authors should balance their statement that they identified targetable epigenetic pathways in hypnozoites.

• They should emphasize the potential role of the host cell in the presentation of the results and the discussion, as it is known that other pathogens modify the epigenome of the host cell (i.e. toxoplasma, HIV) to prevent cell division. Also, hydrazinophtalazines target multiple pathways (notably modulation of calcium flux) and have been shown to inhibit DNA-methyl transferase 1 which is lacking in Plasmodium.

• In a drug repurposing approach, the parasite target might also be different than the human target.

• The authors state that host-cell apoptotic pathways are downregulated in P. vivax infected cells (p. 5 line 162). Maybe the HDAC inhibitors and DNA-methyltransferase inhibitors are reactivating these pathways, leading to parasite death, rather than targeting parasites directly.

It would make the interpretation of the results easier if the authors used EC50 in µM rather than pEC50 in tables and main text. It is easy to calculate when it is a single-digit number but more complicated with multiple digits.

Authors mention hypnozoite-specific effects but in most cases, compounds are as potent on hypnozoite and schizonts. They should rather use "liver stage specific" to refer to increased activity against hypnozoites and schizonts compared to the host cell. The same comment applies to line 351 when referring to MMV019721. Following the same idea, it is a bit far-fetched to call MMV019721 "specific" when the highest concentration tested for cytotoxicity is less than twice the EC50 obtained against hypnozoites and schizonts.

Page 5 lines 187-189, the authors state "...hydrazinophtalazines were inactive when tested against P. berghei liver schizonts and P. falciparum asexual blood stages, suggesting that hypnozoite quiescence may be biologically distinct from developing schizonts". The data provided in Figure 1B show that these hydrazinophtalazines are as potent in P. vivax schizonts than in P. vivax hypnozoites, so the distinct activity seems to be Plasmodium species specific and/or host-cell specific (primary human hepatocytes rather than cell lines for P. berghei) rather than hypnozoite vs schizont specific.

Why choose to focus on cadralazine if abandoned due to side effects? Also, why test the pharmacokinetics in monkeys? As it was a marketed drug, were no data available in humans?

In the counterscreen mentioned on page 6, the authors should mention that the activity of poziotinib in P. berghei and P. cynomolgi is equivalent to cell toxicity, so likely not due to parasite specificity.

To improve the clarity and flow of the manuscript, could the authors make a recapitulative table/figure for all the data obtained for poziotinib and hydrazinophtalazines in the different assays (8-days vs 12-days) and laboratory settings rather than separate tables in main and supplementary figures. Maybe also reorder the results section notably moving the 12-day assay before the DNA methylation part.

The isobologram plot shows an additive effect rather than a synergistic effect between cadralazine and 5-azacytidine, please modify the paragraph title accordingly. Please put the same axis scale for both fractional EC50 in the isobologram graph (Figure 2A).

Concerning the immunofluorescence detection of 5mC and 5hmC, the authors should be careful with their conclusions. The Hoechst signal of the parasites is indistinguishable because of the high signal given by the hepatocyte nuclei. The signal obtained with the anti-5hmC in hepatocyte nuclei is higher than with the anti-5mC, thus if a low signal is obtained in hypnozoites and schizonts, it might be difficult to dissociate from the background. In blood stages (Figure S18), the best to obtain a good signal is to lyse the red blood cell using saponin, before fixation and HCl treatment.

To conclude that 5mC marks are the predominate DNA methylation mark in both P. falciparum and P. vivax, authors should also mention that they compare different stages of the life cycle, that might have different methylation levels.

Also, the authors conclude that "[...] 5mC is present at low level in P. vivax and P. cynomolgi sporozoites and could control liver stage development and hypnozoite quiescence". Based on the data shown here, nothing, except presence the of 5mC marks, supports that DNA methylation could be implicated in liver stage development or hypnozoite quiescence.

How many DNA-methyltransferase inhibitors were present in the epigenetic library? Out of those, none were identified as hits, maybe the hydrazinophtalazines effect is not linked to DNMT inhibition but another target pathway of these molecules like calcium transport?

The authors state (line 344): "These results corroborate our hypothesis that epigenetic pathways regulate hypnozoites". This conclusion should be changed to "[...] that epigenetic pathways are involved in P. vivax liver stage survival" because:<br /> • The epigenetic inhibitors described here are as active on hypnozoite than liver schizonts.<br /> • Again, we cannot rule out that the host cell plays a role in this effect and that the compound may not act directly on the parasite.

The same comment applies to the quote in lines 394 to 396. There is no proof in the results presented here that DNA methylation plays any role in the effect of hydrazinophtalazines in the anti-plasmodial activity obtained in the assay.

eLife assessment

The manuscript by Yang and coworkers presents valuable evidence that an in vitro brain blood barrier composed of endothelial cells, astrocytes, and neuroblastoma cells of human origin, would resemble better the in vivo condition. The presented results constitute solid evidence that GDNF induces the expression of VE-Cadherin and Claudin-5. Further, silencing of GDNF in the brain of mice altered brain blood barrier properties. This provides a new perspective on the interaction between neurons and endothelial cells and this model can be used to screen the permeability of the brain blood barrier to different drugs.

Reviewer #1 (Public Review):

Summary:

In this manuscript, the authors established an in vitro triple co-culture BBB model and demonstrated its advantages compared with the mono or double co-culture BBB model. Further, the authors used their established in vitro BBB model and combined it with other methodologies to investigate the specific mechanism that co-culture with astrocytes but also neurons enhanced the integrity of endothelial cells.

Strengths:

The results persuasively showed the established triple co-culture BBB model well mimicked several important characteristics of BBB compared with the mono-culture BBB model, including better barrier function and in vivo/in vitro correlation. The human-derived immortalized cells used made the model construction process faster and more efficient, and have a better in vivo correlation without species differences. This model is expected to be a useful high-throughput evaluation tool in the development of CNS drugs.

Based on the previous experimental results, detailed studies investigated how co-culture with neurons and astrocytes promoted claudin-5 and VE-cadherin in endothelial cells, and the specific signaling mechanisms were also studied. Interestingly, the authors found that neurons also released GDNF to promote barrier properties of brain endothelial cells, as most current research has focused on the promoting effect of astrocytes-derived GDNF on BBB. Meanwhile, the author also validated the functions of GDNF for BBB integrity in vivo by silencing GDNF in mouse brains. Overall, the experiments and data presented support their claim that, in addition to astrocytes, neurons also have a promoting effect on the barrier function of endothelial cells through GDNF secretion.

Weaknesses:

Although the authors demonstrated a highly usable for predicting the BBB permeability, recorded TEER measurements are still far from the human BBB in vivo reported measurements of TEER, and expression of transporters was not promoted by co-culture, which may lead to the model being unsuitable for studying drug transport mediated by transporters on BBB.

Reviewer #2 (Public Review):

Summary:

Yang and colleagues developed a new in vitro blood-brain barrier model that is relatively simple yet outperforms previous models. By incorporating a neuroblastoma cell line, they demonstrated increased electrical resistance and decreased permeability to small molecules.

Strengths:

The authors initially elucidated the soluble mediator responsible for enhancing endothelial functionality, namely GDNF. Subsequently, they elucidated the mechanisms by which GDNF upregulates the expression of VE-cadherin and Claudin-5. They further validated these findings in vivo, and demonstrated predictive value for molecular permeability as well. The study is meticulously conducted and easily comprehensible. The conclusions are firmly supported by the data, and the objectives are successfully achieved. This research is poised to advance future investigations in BBB permeability, leakage, dysfunction, disease modeling, and drug delivery, particularly in high-throughput experiments. I anticipate an enthusiastic reception from the community interested in this area. While other studies have produced similar results with tri-cultures (PMID: 25630899), this study notably enhances electrical resistance compared to previous attempts.

Weaknesses:

Considerable effort has been directed towards developing in vitro models that more closely resemble their in vivo counterparts, utilizing stem cell-derived NVU cells. Although these examples are currently rudimentary, they offer better BBB mimicry than Yang's study.

Additionally, some instances might benefit from more robust statistical tests; nonetheless, I do not think this would significantly alter the experimental conclusions.

Similar experiments with tri-cultures yielding analogous results have been reported by other authors (PMID: 25630899). TEER values are a bit higher than the aforementioned experiments; however, this study has values at least one order of magnitude lower than physiological levels.

Author response:

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

eLife assessment

This study presents a valuable finding on the mechanism to promote distant metastasis in breast cancer. The evidence supporting the claims of the authors is convincing. The work will be of interest to medical biologists working on breast cancer.

Public Reviews:

Reviewer #1 (Public Review):

Strengths

The paper has shown the expression of RGS10 is related to the molecular subtype, distant metastasis, and survival status of breast cancer. The study utilizes bioinformatic analyses, human tissue samples, and in vitro and in vivo experiments which strengthen the data. RGS10 was validated to inhibit EMT through a novel mechanism dependent on LCN2 and miR-539-5p, thereby reducing cancer cell proliferation, colony formation, invasion, and migration. The study elaborated the function of RGS10 in influencing the prognosis and biological behavior which could be considered as a potential drug target in breast cancer.

Weakness

The mechanism by which the miR-539-5p/RGS10/LCN2 axis may be related to the prognosis of cancer patients still needs to be elucidated. In addition, the sample size used is relatively limited. Especially, if further exploration of the related pathways and mechanisms of LCN2 can be carried out by using organoid models, as well as the potential of RGS10 as a biomarker for further clinical translation to verify its therapeutic target effect, which will make the data more convincing.

Answer: Thank you for your comments and suggestions. In future research, we will utilize large clinical cohorts and organoid models to further explore relevant research mechanisms.

Reviewer #2 (Public Review):

Liu et al., by focusing on the regulation of G protein-signaling 10 (RGS10), reported that RGS10 expression was significantly lower in patients with breast cancer, compared with normal adjacent tissue. Genetic inhibition of RGS10 caused epithelial-mesenchymal transition, and enhanced cell proliferation, migration, and invasion, respectively. These results suggest an inhibitory role of RGS10 in tumor metastasis. Furthermore, bioinformatic analyses determined signaling cascades for RGS10-mediated breast cancer distant metastasis. More importantly, both in vitro and in vivo studies evidenced that alteration of RGS10 expression by modulating its upstream regulator miR-539-5p affects breast cancer metastasis. Altogether, these findings provide insight into the pathogenesis of breast tumors and hence identify potential therapeutic targets in breast cancer.

The conclusions of this study are mostly well supported by data. However, there is a weakness in the study that needs to be clarified.

In Figure 2A, although some references supported that SKBR3 and MCF-7 possess poorly aggressive and less invasive abilities, examining only RGS10 expression in those cells, it could not be concluded that 'RGS10 acts as a tumor suppressor in breast cancer'. It would be better to introduce a horizontal comparison of the invasive ability of these 3 types of cells using an invasion assay.

Answer: Thank you for your comments and suggestions. MDA-MB-231, SKBR3, and MCF-7 originate from triple-negative breast cancer (high invasiveness), Her-2 receptor overexpression (relatively weak invasiveness), and luminal type breast cancer (relatively weak invasiveness) separately. Previous studies have demonstrated the invasive ability of these 3 types of cells. (PMID: 34390568)

Reviewer #3 (Public Review):

Distant metastasis is the major cause of death in patients with breast cancer. In this manuscript, Liu et al. show that RGS10 deficiency elicits distant metastasis via epithelial-mesenchymal transition in breast cancer. As a prognostic indicator of breast cancer, RGS10 regulates the progress of breast cancer and affects tumor phenotypes such as epithelial-mesenchymal transformation, invasion, and migration. The conclusions of this paper are mostly well supported by data, but some analyses need to be clarified.

(1) Because diverse biomarkers have been identified for EMT, it is recommended to declare the advantages of using RGS10 as an EMT marker.

Answer: Thank you for your comments. The dysregulation of RGS protein expression has been observed to be associated with various types of cancer. (PMID: 26293348). Previous studies have shown that RGS10 knocking down can lead to chemotherapy resistance of ovarian cancer cells to paclitaxel, cisplatin, and vincristine. In colorectal tumors, the transcription of RGS10 is regulated by DNA methylation and histone deacetylation. As a key regulatory factor in the G protein signaling pathway, RGS 10 is involved in tumor development including survival, polarization, adhesion, chemotaxis, and differentiation, these hints suggest RGS10 might be a marker for EMT in breast cancer.

(2) The authors utilized databases to study the upstream regulatory mechanisms of RSG10. It is recommended to clarify why the authors focused on miRNAs rather than other epigenetic modifications.

Answer: Thank you for your comments. miRNAs are short-chain non-coding RNA molecules that bind to the target mRNA's 3 'untranslated region (3'UTR) to cause mRNA degradation or translation inhibition, thus regulating gene expression in cells. These small molecules play a crucial role in regulating the expression of cancer-related genes and can act as tumor promoters or tumor suppressors. To further improve the molecular mechanism of malignant biological behavior of breast cancer cells with RGS10, we verified that miR-539-5p might be the upstream regulation target of RGS10 through bioinformatics prediction and in-vitro experiments.

(3) The role of miR-539-5p in breast cancer has been described in previous studies. Hence, it is recommended to provide detailed elaboration on how miR-539-5p regulates the expression of RSG10.

Answer: Thank you for your comments. To verify the effect of miRNA-539-5p regulating the expression of RSG10, we transfected miR-539-5p mimic, miR-539-5p mimic NC, miR-539-5p inhibitor, miR-539-5p inhibitor NC in SKBR3 cells and MDA-MB-231 cells respectively, and verified the expression of RGS10 through RT-qPCR and Western blot experiments. The results showed that compared with the transfected miR-539-5p mimic NC or wild-type SKBR3 cells, RGS10 m RNA and protein levels were significantly reduced. On the contrary, after MDA-MB-231 cells were transfected with miR-539-5p inhibitor to inhibit the expression of miR-539-5p, RGS10 mRNA and protein levels in MDA-MB-231 cells were significantly increased (Fig. 3.4A-C, Fig. 3.5A-C). This indicates that miR-539-5p can target and regulate RGS10.

(4) To enhance the clarity and interpretability of the Western blot results, it would be advisable to mark the specific kilodalton (kDa) values of the proteins.

Answer: Thank you for your comments and suggestions. We have corrected to mark the specific kilodalton (kDa) values of the proteins in WB.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

The function of RGS10 in breast cancer was identified in the paper. However, some major issues in this paper need to be specified:

(1) From reading the introduction section and its references, RGS proteins participate in multiple essential cellular processes and may be tumor initiators or suppressors (Li et al., 2023). This article focuses on the significance of RGS10 in breast cancer, it is recommended to show how the function of RGS10 exhibits therapeutic significance in other types of cancer.

Answer: Thanks for your comments and suggestions on our findings. The dysregulation of RGS protein expression has been observed to be associated with various types of cancer. Especially in ovarian cancer cells. (PMID: 26293348). It has been found that the RGS10 expression is lower than that of normal ovarian cells. (PMID: 21044322). In addition, it has been found that knocking down RGS10 can enhance the vitality of ovarian cancer cells and promote chemoresistance by activating the Rheb GTP/mTOR signaling pathway. (PMID: 26319900). A study suggests that RGS10 mediates inflammation signaling regulation in SKOV-3 ovarian cancer cells with high expression of TNF and COX-2 after RGS10 knockdown. In colorectal tumors, RGS10 transcription is regulated by DNA methylation and histone deacetylation. (PMID: 35810565). RGS10 expression also are associated with poor prognosis in laryngeal cancer, hepatocellular carcinoma, and pediatric acute myeloid leukemia. (PMID: 32776811, PMID: 26516143, PMID: 30538250)

(2) The authors characterize RGS10 protein expression in the breast cancer cell lines MDA-MB-231, MCF7, and SKBR3 in vitro Figure 2A. However, more information would strengthen the data - e.g. information on the expression of RGS10 protein and the survival in public databases, as well as the correlation between RGS10 and Her-2 expression.

Answer: Thanks for your comments. we have checked the correlation of RGS10 expression and survival rate of Her-2 positive breast cancer patients in a public database. Although there is no significant difference in the “p” value, however, RGS10 high-expression patients have a favorable prognosis tendency than RGS10 low-expression patients after the 100th month.

Author response image 1.

(3) Regarding the current situation of clinical trials in the RGS family, the potential to develop RGS 10 for clinic translation is a driving factor for EMT.

Answer: Thank you for your comments. The RGS (G protein signal transduction regulator) gene family provides an important "braking" function for the cell receptor family of G-protein coupled receptors (GPCR). GPCR controls hundreds of important functions in systemic cells and is the largest class of drug targets, with over one-third of FDA approved drugs treating diseases by binding to GPCR and altering its activity. When GPCRs are activated by hormones or neurotransmitters, they initiate signaling cascades within host cells through signal-carrying proteins called G proteins. The function of the RGS protein is to inactivate the G protein, thereby shutting down this signaling cascade reaction, which limits G protein signal transduction and allows cells to reset and receive new incoming signals. If it were not for it, the signals triggered by GPCR would inappropriately remain on, and the signal transduction would experience dysfunction (PMID: 33007266). The potential to develop RGS10 as a driving factor of EMT is meaningful for clinic translation.

(4) In Figure 3A, the paper showed that differential gene expression revealed 70 genes were significantly upregulated in RGS10-depleted SKBR3 cells, The authors didn't show any data on the expression of other EMT-related proteins in pathway analysis.

Answer: Thank you for your comments. The enrichment analysis of RNA sequencing in RGS10-depleted SKBR3 cells suggests that high correlation factors that are associated with EMT, such as TAGLN, TNFSF10, NDUFA4L2, CCN5, PHGDH, ST3GAL5, ANG, and LCN2.

(5) In Figure 3B, the paper focuses on LCN2 in pathway analysis, however, the author did not elaborate on the significance of LCN2-related pathways in EMT.

Answer: Thank you for your comments. Some studies have the significance of LCN2-related pathways in EMT. It was confirmed that LCN2 upregulation triggered by PTEN insufficiency induces EMT to promote migration and invasion in MCF7 cells (PMID: 27466505). The activation of STAT3 contributes to an increase in LCN2 expression, which activates ERK pathway-dependent EMT, thus promoting lung metastasis in MDA-MB-231 cells in breast cancer (PMID: 33473115). The silencing of LCN2 reduced the ability of migration and invasion of SUM149 cells and the proportion of tumor stem cells, suggesting that LCN2 may mediate the invasion and metastasis of cancer cells by regulating the stemness of breast cancer cells. The biological effects of LCN2 small molecule inhibitors ZINC00640089 and ZINC00784494 targeting IBC cells have been confirmed. The siRNA-mediated silencing of LCN2 in IBC cells significantly reduces cell proliferation, viability, migration, and invasion. (PMID: 34445288).

(6) Minor: the author did not conduct a semi-quantitative analysis of the immunohistochemical results of RGS10.

Answer: Thank you for your suggestion. We would like to demonstrate the qualitative analysis of RGS10 immunohistochemistry. The semi-quantitative analysis is not required in the paper.

Reviewer #2 (Recommendations For The Authors):

The role of RGS10 was well-characterized in this study, However, some minor points need to be modified.

(1) Page 15 line 296, description of cell proliferation was missing, please modify.

Answer: Thank you for your comments. We have corrected the description of cell proliferation on Page 15 highlighted in red.

(2) In Figure 2C, the title of the Y-axis was missing.

Answer: Thank you for your comments. We have corrected the description of the Y-axis title in Figure 2C.

(3) Describe the transfection reagent that was used in this study, and incorporated into the methods section.

Answer: Thank you for your comments. We have added the description of the transfection reagent to the methods section.

(4) The manuscript needs proofreading.

Answer: Thank you for your comments. We have proofread the manuscript.

Reviewer #2 (Public Review):

Liu et al., by focusing on the regulation of G protein-signaling 10 (RGS10), reported that RGS10 expression was significantly lower in patients with breast cancer, compared with normal adjacent tissue. Genetic inhibition of RGS10 caused epithelial-mesenchymal transition, and enhanced cell proliferation, migration, and invasion, respectively. These results suggest an inhibitory role of RGS10 in tumor metastasis. Furthermore, bioinformatic analyses determined signaling cascades for RGS10-mediated breast cancer distant metastasis. More importantly, both in vitro and in vivo studies evidenced that alteration of RGS10 expression by modulating its upstream regulator miR-539-5p affects breast cancer metastasis. Altogether, these findings provide insight into the pathogenesis of breast tumors and hence identify potential therapeutic targets in breast cancer.

The conclusions of this study are mostly well supported by data.

Reviewer #3 (Public Review):

Distant metastasis is the major cause of death in patients with breast cancer. In this manuscript, Liu et al. show that RGS10 deficiency elicits distant metastasis via epithelial-mesenchymal transition in breast cancer. As a prognostic indicator of breast cancer, RGS10 regulates the progress of breast cancer and affects tumor phenotypes such as epithelial-mesenchymal transformation, invasion, and migration. The conclusions of this paper are mostly well supported by data.

eLife assessment

This valuable paper first demonstrated that RGS10 was identified as a biomarker to evaluate the prognosis of breast cancer. To prevent the loss of RGS10 theoretically provide a new strategy for the treatment of breast cancer. The evidence supporting the claims of the authors is solid, although inclusion of a larger number of patient samples and an animal model would have strengthened the study. The work will be of interest to clinicians working on breast cancer.

Reviewer #1 (Public Review):

The paper has shown the expression of RGS10 is related to the molecular subtype, distant metastasis, and survival status of breast cancer. The study utilizes bioinformatic analyses, human tissue samples, and in vitro and in vivo experiments which strengthen the data. RGS10 was validated to inhibit EMT through a novel mechanism dependent on LCN2 and miR-539-5p, thereby reducing cancer cell proliferation, colony formation, invasion, and migration. The study elaborated on the function of RGS10 in influencing the prognosis and biological behavior which could be considered as a potential drug target in breast cancer.

eLife assessment

This study investigates the role of the Cadherin Flamingo (Fmi) in cell competition in developing tissues in Drosophila melanogaster. The findings are valuable in that they show that Fmi is required in winning cells in several competitive contexts. The evidence supporting the conclusions is solid, as the authors identify Fmi as a potential new regulator of cell competition, however, they don't delve into a mechanistic understanding of how this occurs.

Reviewer #1 (Public Review):

Summary:

This paper is focused on the role of Cadherin Flamingo (Fmi) - also called Starry night (stan) - in cell competition in developing Drosophila tissues. A primary genetic tool is monitoring tissue overgrowths caused by making clones in the eye disc that express activated Ras (RasV12) and that are depleted for the polarity gene scribble (scrib). The main system that they use is ey-flp, which makes continuous clones in the developing eye-antennal disc beginning at the earliest stages of disc development. It should be noted that RasV12, scrib-i (or lgl-i) clones only lead to tumors/overgrowths when generated by continuous clones, which presumably creates a privileged environment that insulates them from competition. Discrete (hs-flp) RasV12, lgl-i clones are in fact out-competed (PMID: 20679206), which is something to bear in mind.

The authors show that clonal loss of Fmi by an allele or by RNAi in the RasV12, scrib-i tumors suppresses their growth in both the eye disc (continuous clones) and wing disc (discrete clones). The authors attributed this result to less killing of WT neighbors when Myc over-expressing clones lacking Fmi, but another interpretation (that Fmi regulates clonal growth) is equally as plausible with the current results. Next, the authors show that scrib-RNAi clones that are normally out-competed by WT cells prior to adult stages are present in higher numbers when WT cells are depleted for Fmi. They then examine death in RasV12, scrib-i ey-FLP clones, or in discrete hs-FLP UAS-Myc clones. They state that they see death in WT cells neighboring RasV12, scrib-i clones in the eye disc (Figures 4A-C). Next, they write that RasV12, scrib-I cells become losers (i.e., have apoptosis markers) when Fmi is removed. Neither of these results are quantified and thus are not compelling. They state that a similar result is observed for Myc over-expression clones that lack Fmi, but the image was not compelling, the results are not quantified and the controls are missing (Myc over-expressing clones alone and Fmi clones alone). They then want to test whether Myc over-expressing clones have more proliferation. They show an image of a wing disc that has many small Myc overexpressing clones with and without Fmi. The pHH3 results support their conclusion that Myc overexpressing clones have more pHH3, but I have reservations about the many clones in these panels (Figures 5L-N). They show that the cell competition roles of Fmi are not shared by another PCP component and are not due to the Cadherin domain of Fmi. The authors appear to interpret their results as Fmi is required for winner status. Overall, some of these results are potentially interesting and at least partially supported by the data, but others are not supported by the data.

Strengths:

Fmi has been studied for its role in planar cell polarity, and its potential role in competition is interesting.

Weaknesses:

(1) In the Myc over-expression experiments, the increased size of the Myc clones could be because they divide faster (but don't outcompete WT neighbors). If the authors want to conclude that the bigger size of the Myc clones is due to out-competition of WT neighbors, they should measure cell death across many discs of with these clones. They should also assess if reducing apoptosis (like using one copy of the H99 deficiency that removes hid, rpr, and grim) suppresses winner clone size. If cell death is not addressed experimentally and quantified rigorously, then their results could be explained by faster division of Myc over-expressing clones (and not death of neighbors). This could also apply to the RasV12, scrib-i results.

(2) This same comment about Fmi affecting clone growth should be considered in the scrib RNAi clones in Figure 3.

(3) I don't understand why the quantifications of clone areas in Figures 2D, 2H, 6D are log values. The simple ratio of GFP/RFP should be shown. Additionally, in some of the samples (e.g., fmiE59 >> Myc, only 5 discs and fmiE59 vs >Myc only 4 discs are quantified but other samples have more than 10 discs). I suggest that the authors increase the number of discs that they count in each genotype to at least 20 and then standardize this number.

(4) There is a typo when referring to Figures 3C-D. It should be Figure 2C-D.

(5) Figure 4 - shows examples of cell death. Cas3 is written on the figure but Dcp-1 is written in the results. Which antibody was used? The authors need to quantify these results. They also need to show that the death of cells is part of the phenotype, like an H99 deficiency, etc (see above).

(6) It is well established that clones overexpressing Myc have increased cell death. The authors should consider this when interpreting their results.

(7) A better characterization of discrete Fmi clones would also be helpful. I suggest inducing hs-flp clones in the eye or wing disc and then determining clone size vs twin spot size and also examining cell death etc. If such experiments have already been done and published, the authors should include a description of such work in the preprint.

(8) We need more information about the expression pattern of Fmi. Is it expressed in all cells in imaginal discs? Are there any patterns of expression during larval and pupal development?

(9) Overall, the paper is written for specialists who work in cell competition and is fairly difficult to follow, and I suggest re-writing the results to make it accessible to a broader audience.

Reviewer #2 (Public Review):

Summary:

In this manuscript, Bosch et al. reveal Flamingo (Fmi), a planar cell polarity (PCP) protein, is essential for maintaining 'winner' cells in cell competition, using Drosophila imaginal epithelia as a model. They argue that tumor growth induced by scrib-RNAi and RasV12 competition is slowed by Fmi depletion. This effect is unique to Fmi, not seen with other PCP proteins. Additional cell competition models are applied to further confirm Fmi's role in 'winner' cells. The authors also show that Fmi's role in cell competition is separate from its function in PCP formation.

Strengths:

(1) The identification of Fmi as a potential regulator of cell competition under various conditions is interesting.

(2) The authors demonstrate that the involvement of Fmi in cell competition is distinct from its role in planar cell polarity (PCP) development.

Weaknesses:

(1) The authors provide a superficial description of the related phenotypes, lacking a comprehensive mechanistic understanding. Induction of apoptosis and JNK activation are general outcomes, but it is important to determine how they are specifically induced in Fmi-depleted clones. The authors should take advantage of the power of fly genetics and conduct a series of genetic epistasis analyses.

(2) The depletion of Fmi may not have had a significant impact on cell competition; instead, it is more likely to have solely facilitated the induction of apoptosis.

(3) To make a solid conclusion for Figure 1, the authors should investigate whether complete removal of Fmi by a mutant allele affects tumor growth induced by expressing RasV12 and scrib RNAi throughout the eye.

(4) The authors should test whether the expression level of Fmi (both mRNA and protein) changes during tumorigenesis and cell competition.

Reviewer #3 (Public Review):

Summary:

In this manuscript, Bosch and colleagues describe an unexpected function of Flamingo, a core component of the planar cell polarity pathway, in cell competition in the Drosophila wing and eye disc. While Flamingo depletion has no impact on tumour growth (upon induction of Ras and depletion of Scribble throughout the eye disc), and no impact when depleted in WT cells, it specifically tunes down winner clone expansion in various genetic contexts, including the overexpression of Myc, the combination of Scribble depletion with activation of Ras in clones or the early clonal depletion of Scribble in eye disc. Flamingo depletion reduces the proliferation rate and increases the rate of apoptosis in the winner clones, hence reducing their competitiveness up to forcing their full elimination (hence becoming now "loser"). This function of Flamingo in cell competition is specific to Flamingo as it cannot be recapitulated with other components of the PCP pathway, and does not rely on the interaction of Flamingo in trans, nor on the presence of its cadherin domain. Thus, this function is likely to rely on a non-canonical function of Flamingo which may rely on downstream GPCR signaling.

This unexpected function of Flamingo is by itself very interesting. In the framework of cell competition, these results are also important as they describe, to my knowledge, one of the only genetic conditions that specifically affect the winner cells without any impact when depleted in the loser cells. Moreover, Flamingo does not just suppress the competitive advantage of winner clones, but even turns them into putative losers. This specificity, while not clearly understood at this stage, opens a lot of exciting mechanistic questions, but also a very interesting long-term avenue for therapeutic purposes as targeting Flamingo should then affect very specifically the putative winner/oncogenic clones without any impact in WT cells.

The data and the demonstration are very clean and compelling, with all the appropriate controls, proper quantification, and backed-up by observations in various tissues and genetic backgrounds. I don't see any weakness in the demonstration and all the points raised and claimed by the authors are all very well substantiated by the data. As such, I don't have any suggestions to reinforce the demonstration.

While not necessary for the demonstration, documenting the subcellular localisation and levels of Flamingo in these different competition scenarios may have been relevant and provided some hints on the putative mechanism (specifically by comparing its localisation in winner and loser cells).

Also, on a more interpretative note, the absence of the impact of Flamingo depletion on JNK activation does not exclude some interesting genetic interactions. JNK output can be very contextual (for instance depending on Hippo pathway status), and it would be interesting in the future to check if Flamingo depletion could somehow alter the effect of JNK in the winner cells and promote downstream activation of apoptosis (which might normally be suppressed). It would be interesting to check if Flamingo depletion could have an impact in other contexts involving JNK activation or upon mild activation of JNK in clones.

Strengths:

- A clean and compelling demonstration of the function of Flamingo in winner cells during cell competition.

- One of the rare genetic conditions that affects very specifically winner cells without any impact on losers, and then can completely switch the outcome of competition (which opens an interesting therapeutic perspective in the long term)

Weaknesses:

- The mechanistic understanding obviously remains quite limited at this stage especially since the signaling does not go through the PCP pathway.

Author response:

We would like to thank the reviewers for their constructive feedback. We have thoroughly considered their concerns and comments and we aim to include some additional results in an updated version of this manuscript. In addition, we would like to address some of the comments, with which we respectfully disagree. Below is our point-by-point reply.

Reviewer 1:

Summary:

This paper is focused on the role of Cadherin Flamingo (Fmi) - also called Starry night (stan) - in cell competition in developing Drosophila tissues. A primary genetic tool is monitoring tissue overgrowths caused by making clones in the eye disc that express activated Ras (RasV12) and that are depleted for the polarity gene scribble (scrib). The main system that they use is ey-flp, which makes continuous clones in the developing eye-antennal disc beginning at the earliest stages of disc development. It should be noted that RasV12, scrib-i (or lgl-i) clones only lead to tumors/overgrowths when generated by continuous clones, which presumably creates a privileged environment that insulates them from competition. Discrete (hs-flp) RasV12, lgl-i clones are in fact out-competed (PMID: 20679206), which is something to bear in mind. 

We think it is unlikely that the outcome of RasV12, scrib (or lgl) competition depends on discrete vs. continuous clones or on creation of a privileged environment. As shown in the same reference mentioned by the reviewer, the outcome of RasV12, scrib (or lgl) tumors greatly depends on the clone being able to grow to a certain size. The authors show instances of discrete clones where larger RasV12, lgl clones outcompete the surrounding tissue and eliminate WT cells by apoptosis, whereas smaller clones behave more like losers. It is not clear what aspect of the environment determines the ability of some clones to grow larger than others, but in neither case are the clones prevented from competition. Other studies show that in mammalian cells, RasV12, scrib clones are capable of outcompeting the surrounding tissue, such as in Kohashi et al (2021), where cells carrying both mutations actively eliminate their neighbors.

The authors show that clonal loss of Fmi by an allele or by RNAi in the RasV12, scrib-i tumors suppresses their growth in both the eye disc (continuous clones) and wing disc (discrete clones). The authors attributed this result to less killing of WT neighbors when Myc over-expressing clones lacking Fmi, but another interpretation (that Fmi regulates clonal growth) is equally as plausible with the current results.

See point (1) for a discussion on this.

Next, the authors show that scrib-RNAi clones that are normally out-competed by WT cells prior to adult stages are present in higher numbers when WT cells are depleted for Fmi. They then examine death in RasV12, scrib-i ey-FLP clones, or in discrete hs-FLP UAS-Myc clones. They state that they see death in WT cells neighboring RasV12, scrib-i clones in the eye disc (Figures 4A-C). Next, they write that RasV12, scrib-I cells become losers (i.e., have apoptosis markers) when Fmi is removed. Neither of these results are quantified and thus are not compelling. They state that a similar result is observed for Myc over-expression clones that lack Fmi, but the image was not compelling, the results are not quantified and the controls are missing (Myc over-expressing clones alone and Fmi clones alone).

We assayed apoptosis in UAS-Myc clones in eye discs but neglected to include the results in Figure 4. We will include them in the updated manuscript. Regarding Fmi clones alone, we direct the reviewer’s attention to Fig. 2 Supplement 1 where we showed that fminull clones cause no competition. Dcp-1 staining showed low levels of apoptosis unrelated to the fminull clones or twin-spots, and we will comment on this in the revised manuscript.

Regarding the quantification of apoptosis, we did not provide a quantification, in part because we observe a very clear visual difference between groups (Fig. 4A-K), and in part because it is challenging to come up with a rigorous quantification method. For example, how far from a winner clone can an apoptotic cell be and still be considered responsive to the clone? For UAS-Myc winner clones, we observe a modest amount of cell death both inside and outside the clones, consistent with prior observations. For fminull UAS-Myc clones, we observe vastly more cell death within the fminull UAS-Myc clones and modest death in nearby wildtype cells, and consequently a much higher ratio of cell death inside vs outside the clone. Because of the somewhat arbitrary nature of quantification, and the dramatic difference, we initially chose not to provide a quantification. However, given the request, we chose an arbitrary distance from the clone boundary in which to consider dying cells and counted the numbers for each condition. We view this as a very soft quantification, but will report it in a way that captures the phenomenon in the revised manuscript.

They then want to test whether Myc over-expressing clones have more proliferation. They show an image of a wing disc that has many small Myc overexpressing clones with and without Fmi. The pHH3 results support their conclusion that Myc overexpressing clones have more pHH3, but I have reservations about the many clones in these panels (Figures 5L-N).

As the reviewer’s reservations are not specified, we have no specific response.

They show that the cell competition roles of Fmi are not shared by another PCP component and are not due to the Cadherin domain of Fmi. The authors appear to interpret their results as Fmi is required for winner status. Overall, some of these results are potentially interesting and at least partially supported by the data, but others are not supported by the data.

Strengths: 

Fmi has been studied for its role in planar cell polarity, and its potential role in competition is interesting.

Weaknesses:

(1) In the Myc over-expression experiments, the increased size of the Myc clones could be because they divide faster (but don't outcompete WT neighbors). If the authors want to conclude that the bigger size of the Myc clones is due to out-competition of WT neighbors, they should measure cell death across many discs of with these clones. They should also assess if reducing apoptosis (like using one copy of the H99 deficiency that removes hid, rpr, and grim) suppresses winner clone size. If cell death is not addressed experimentally and quantified rigorously, then their results could be explained by faster division of Myc over-expressing clones (and not death of neighbors). This could also apply to the RasV12, scrib-i results.

Indeed, Myc clones have been shown to divide faster than WT neighbors, but that is not the only reason clones are bigger. As shown in (de la Cova et al, 2004), Myc-overexpressing cells induce apoptosis in WT neighbors, and blocking this apoptosis results in larger wings due to increased presence of WT cells. Also, (Moreno and Basler, 2004) showed that Myc-overexpressing clones cause a reduction in WT clone size, as WT twin spots adjacent to 4xMyc clones are significantly smaller than WT twin spots adjacent to WT clones. In the same work, they show complete elimination of WT clones generated in a tub-Myc background. Since then, multiple papers have shown these same results. It is well established then that increased cell proliferation transforms Myc clones into supercompetitors and that in the absence of cell competition, Myc-overexpressing discs produce instead wings larger than usual.

In (de la Cova et al, 2004) the authors already showed that blocking apoptosis with H99 hinders competition and causes wings with Myc clones to be larger than those where apoptosis wasn’t blocked. As these results are well established from prior literature, there is no need to repeat them here.

(2) This same comment about Fmi affecting clone growth should be considered in the scrib RNAi clones in Figure 3.

In later stages, scrib RNAi clones in the eye are eliminated by WT cells. While scrib RNAi clones are not substantially smaller in third instar when competing against fmi cells (Fig 3M), by adulthood we see that WT clones lacking Fmi have failed to remove scrib clones, unlike WT clones that have completely eliminated the scrib RNAi clones by this time. We therefore disagree that the only effect of Fmi could be related to rate of cell division.

(3) I don't understand why the quantifications of clone areas in Figures 2D, 2H, 6D are log values. The simple ratio of GFP/RFP should be shown. Additionally, in some of the samples (e.g., fmiE59 >> Myc, only 5 discs and fmiE59 vs >Myc only 4 discs are quantified but other samples have more than 10 discs). I suggest that the authors increase the number of discs that they count in each genotype to at least 20 and then standardize this number.

Log(ratio) values are easier to interpret than a linear scale. If represented linearly, 1 means equal ratios of A and B, while 2A/B is 2 and A/2B is 0.5. And the higher the ratio difference between A and B, the starker this effect becomes, making a linear scale deceiving to the eye, especially when decreased ratios are shown. Using log(ratios), a value of 0 means equal ratios, and increased and decreased ratios deviate equally from 0.

Statistically, either analyzing a standardized number of discs for all conditions or a variable number not determined beforehand has no effect on the p-value, as long as the variable n number is not manipulated by p-hacking techniques, such as increasing the n of samples until a significant p-value has been obtained. While some of our groups have lower numbers, all statistical analyses were performed after all samples were collected. For all results obtained by cell counts, all samples had a minimum of 10 discs due to the inherent though modest variability of our automated cell counts, and we analyzed all the discs that we obtained from a given experiment, never “cherry-picking” examples. For the sake of transparency, all our graphs show individual values in addition to the distributions so that the reader knows the n values at a glance.

(5) Figure 4 - shows examples of cell death. Cas3 is written on the figure but Dcp-1 is written in the results. Which antibody was used? The authors need to quantify these results. They also need to show that the death of cells is part of the phenotype, like an H99 deficiency, etc (see above).

Thank you for flagging this error. We used cleaved Dcp-1 staining to detect cell death, not Cas3 (Drice in Drosophila). We will update all panels replacing Cas3 by Dcp-1.

As described above, cell death is a well established consequence of myc overexpression induced cell death and we feel there is no need to repeat that result. To what extent loss of Fmi induces excess cell death or reduces proliferation in “would-be” winners, and to what extent it reduces “would-be” winners’ ability to eliminate competitors are interesting mechanistic questions that are beyond the scope of the current manuscript.

(6) It is well established that clones overexpressing Myc have increased cell death. The authors should consider this when interpreting their results.

We are aware that Myc-overexpressing clones have increased cell death, but it has also been demonstrated that despite that fact, they behave as winners and eliminate WT neighboring cells. And as mentioned in comment (1), WT clones generated in a 3x and 4x Myc background are eliminated and removed from the tissue, and blocking cell death increases the size of WT “losers” clones adjacent to Myc overexpressing clones.

(7) A better characterization of discrete Fmi clones would also be helpful. I suggest inducing hs-flp clones in the eye or wing disc and then determining clone size vs twin spot size and also examining cell death etc. If such experiments have already been done and published, the authors should include a description of such work in the preprint.

We have already analyzed the size of discrete Fmi clones and showed that they did not cause any competition, with fmi-null clones having the same size as WT clones in both eye and wing discs. We direct the reviewer’s attention to Figure 2 Supplement 1.

(8) We need more information about the expression pattern of Fmi. Is it expressed in all cells in imaginal discs? Are there any patterns of expression during larval and pupal development?

Fmi is equally expressed by all cells in all imaginal discs in Drosophila larva and pupa. We will include this information in the updated manuscript.

(9) Overall, the paper is written for specialists who work in cell competition and is fairly difficult to follow, and I suggest re-writing the results to make it accessible to a broader audience.

We have endeavored to both provide an accessible narrative and also describe in sufficient detail the data from multiple models of competition and complex genetic systems. We hope that most readers will be able, at a minimum, to follow our interpretations and the key takeaways, while those wishing to examine the nuts and bolts of the argument will find what they need presented as simply as possible.

Reviewer 2:

Summary:

In this manuscript, Bosch et al. reveal Flamingo (Fmi), a planar cell polarity (PCP) protein, is essential for maintaining 'winner' cells in cell competition, using Drosophila imaginal epithelia as a model. They argue that tumor growth induced by scrib-RNAi and RasV12 competition is slowed by Fmi depletion. This effect is unique to Fmi, not seen with other PCP proteins. Additional cell competition models are applied to further confirm Fmi's role in 'winner' cells. The authors also show that Fmi's role in cell competition is separate from its function in PCP formation.

We would like to thank the reviewer for their thoughtful and positive review.

Strengths:

(1) The identification of Fmi as a potential regulator of cell competition under various conditions is interesting.

(2) The authors demonstrate that the involvement of Fmi in cell competition is distinct from its role in planar cell polarity (PCP) development.

Weaknesses:

(1) The authors provide a superficial description of the related phenotypes, lacking a comprehensive mechanistic understanding. Induction of apoptosis and JNK activation are general outcomes, but it is important to determine how they are specifically induced in Fmi-depleted clones. The authors should take advantage of the power of fly genetics and conduct a series of genetic epistasis analyses.

We appreciate that this manuscript does not address the mechanism by which Fmi participates in cell competition. Our intent here is to demonstrate that Fmi is a key contributor to competition. We indeed aim to delve into mechanism, are currently directing our efforts to exploring how Fmi regulates competition, but the size of the project and required experiments are outside of the scope of this manuscript. We feel that our current findings are sufficiently valuable to merit sharing while we continue to investigate the mechanism linking Fmi to competition.

(2) The depletion of Fmi may not have had a significant impact on cell competition; instead, it is more likely to have solely facilitated the induction of apoptosis.

We respectfully disagree for several reasons. First, loss of Fmi is specific to winners; loss of Fmi has no effect on its own or in losers when confronting winners in competition. And in the Ras V12 tumor model, loss of Fmi did not perturb whole eye tumors – it only impaired tumor growth when tumors were confronted with competitors. We agree that induction of apoptosis is affected, but so too is proliferation, and only when in winners in competition.

(3) To make a solid conclusion for Figure 1, the authors should investigate whether complete removal of Fmi by a mutant allele affects tumor growth induced by expressing RasV12 and scrib RNAi throughout the eye.

We agree with the reviewer that this is a worthwhile experiment, given that RNAi has its limitations. However, as fmi is homozygous lethal at the embryo stage, one cannot create whole disc tumors mutant for fmi. As an approximation to this condition, we have introduced the GMR-Hid, cell-lethal combination to eliminate non-tumor tissue in the eye disc. Following elimination of non-tumor cells, there remains essentially a whole disc harboring fminull tumor. Indeed, this shows that whole fminull tumors overgrow similar to control tumors, confirming that the lack of Fmi only affects clonal tumors. We will provide those results in the updated manuscript.

(4) The authors should test whether the expression level of Fmi (both mRNA and protein) changes during tumorigenesis and cell competition.

This is an intriguing point that we would like to validate. We are currently performing immunostaining for Fmi in clones to confirm whether its levels change during competition. We will provide these results in the updated manuscript.

Reviewer 3:

Summary: <br /> In this manuscript, Bosch and colleagues describe an unexpected function of Flamingo, a core component of the planar cell polarity pathway, in cell competition in the Drosophila wing and eye disc. While Flamingo depletion has no impact on tumour growth (upon induction of Ras and depletion of Scribble throughout the eye disc), and no impact when depleted in WT cells, it specifically tunes down winner clone expansion in various genetic contexts, including the overexpression of Myc, the combination of Scribble depletion with activation of Ras in clones or the early clonal depletion of Scribble in eye disc. Flamingo depletion reduces the proliferation rate and increases the rate of apoptosis in the winner clones, hence reducing their competitiveness up to forcing their full elimination (hence becoming now "loser"). This function of Flamingo in cell competition is specific to Flamingo as it cannot be recapitulated with other components of the PCP pathway, and does not rely on the interaction of Flamingo in trans, nor on the presence of its cadherin domain. Thus, this function is likely to rely on a non-canonical function of Flamingo which may rely on downstream GPCR signaling.

This unexpected function of Flamingo is by itself very interesting. In the framework of cell competition, these results are also important as they describe, to my knowledge, one of the only genetic conditions that specifically affect the winner cells without any impact when depleted in the loser cells. Moreover, Flamingo does not just suppress the competitive advantage of winner clones, but even turns them into putative losers. This specificity, while not clearly understood at this stage, opens a lot of exciting mechanistic questions, but also a very interesting long-term avenue for therapeutic purposes as targeting Flamingo should then affect very specifically the putative winner/oncogenic clones without any impact in WT cells.

The data and the demonstration are very clean and compelling, with all the appropriate controls, proper quantification, and backed-up by observations in various tissues and genetic backgrounds. I don't see any weakness in the demonstration and all the points raised and claimed by the authors are all very well substantiated by the data. As such, I don't have any suggestions to reinforce the demonstration.

While not necessary for the demonstration, documenting the subcellular localisation and levels of Flamingo in these different competition scenarios may have been relevant and provided some hints on the putative mechanism (specifically by comparing its localisation in winner and loser cells). 

Also, on a more interpretative note, the absence of the impact of Flamingo depletion on JNK activation does not exclude some interesting genetic interactions. JNK output can be very contextual (for instance depending on Hippo pathway status), and it would be interesting in the future to check if Flamingo depletion could somehow alter the effect of JNK in the winner cells and promote downstream activation of apoptosis (which might normally be suppressed). It would be interesting to check if Flamingo depletion could have an impact in other contexts involving JNK activation or upon mild activation of JNK in clones.

We would like to thank the reviewer for their thorough and positive review.

Strengths: 

- A clean and compelling demonstration of the function of Flamingo in winner cells during cell competition.

- One of the rare genetic conditions that affects very specifically winner cells without any impact on losers, and then can completely switch the outcome of competition (which opens an interesting therapeutic perspective in the long term)

Weaknesses: 

- The mechanistic understanding obviously remains quite limited at this stage especially since the signaling does not go through the PCP pathway.

Reviewer 2 made the same comment in their weakness (1), and we refer to that response. In future work, we are excited to better understand the pathways linking Fmi and competition.

eLife assessment

This manuscript reports a valuable new mechanism of regulation of the glutamine synthetase in the archaeon Methanosarcina mazei and clarifies the direct activation of glutamine synthetase activity by 2-oxoglutarate, thus introducing a novel understanding of how 2-oxoglutarate serves as a central indicator of carbon and nitrogen sensing. The authors provide solid evidence using mass photometry, specific activity measurements, and single particle cryo-EM data. This study is of interest to biologists working on the regulation of metabolism.

Reviewer #1 (Public Review):

Summary:

This study shows a new mechanism of GS regulation in the archaean Methanosarcina maze and clarifies the direct activation of GS activity by 2-oxoglutarate, thus featuring another way in which 2-oxoglutarate acts as a central status reporter of C/N sensing.

Mass photometry and single particle cryoEM structure analysis convincingly show the direct regulation of GS activity by 2-OG promoted formation of the dodecameric structure of GS. The previously recognized small proteins GlnK1 and Sp26 seem to play a subordinate role in GS regulation, which is in good agreement with previous data. Although these data are quite clear now, there remains one major open question: how does 2-OG further increase GS activity once the full dodecameric state is achieved (at 5 mM)? This point needs to be reconsidered.

Strengths:

Mass photometry reveals a dynamic mode of the effect of 2-OG on the oligomerization state of GS. Single particle Cryo-EM reveals the mechanism of 2-OG mediated dodecamer formation.

Weaknesses:

It is not entirely clear, how very high 2-OG concentrations activate GS beyond dodecamer formation.

The data presented in this work are in stark contrast to the previously reported structure of M. mazei GS by the Schumacher lab. This is very confusing for the scientific community and requires clarification. The discussion should consider possible reasons for the contradictory results.

Importantly, it is puzzling how Schumacher could achieve an apo-structire of dodecameeric GS? If 2-OG is necessary for dodecameric formation, this should be discussed. If GlnK1 doesn't form a complex with the dodecameric GS, how could such a complex be resolved there?

In addition, the text is in principle clear but could be improved by professional editing. Most obviously there is insufficient comma placement.

Reviewer #2 (Public Review):

Summary:

Herdering et al. introduced research on an archaeal glutamine synthetase (GS) from Methanosarcina mazei, which exhibits sensitivity to the environmental presence of 2-oxoglutarate (2-OG). While previous studies have indicated 2-OG's ability to enhance GS activity, the precise underlying mechanism remains unclear. Initially, the authors utilized biophysical characterization, primarily employing a nanomolar-scale detection method called mass photometry, to explore the molecular assembly of Methanosarcina mazei GS (M. mazei GS) in the absence or presence of 2-OG. Similar to other GS enzymes, the target M. mazei GS forms a stable dodecamer, with two hexameric rings stacked in tail-to-tail interactions. Despite approximately 40% of M. mazei GS existing as monomeric or dimeric entities in the detectable solution, the majority spontaneously assemble into a dodecameric state. Upon mixing 2-OG with M. mazei GS, the population of the dodecameric form increases proportionally with the concentration of 2-OG, indicating that 2-OG either promotes or stabilizes the assembly process. The cryo-electron microscopy (cryo-EM) structure reveals that 2-OG is positioned near the interface of two hexameric rings. At a resolution of 2.39 Å, the cryo-EM map vividly illustrates 2-OG forming hydrogen bonds with two individual GS subunits as well as with solvent water molecules. Moreover, local side-chain reorientation and conformational changes of loops in response to 2-OG further delineate the 2-OG-stabilized assembly of M. mazei GS.

Strengths & Weaknesses:

The investigation studies the impact of 2-oxoglutarate (2-OG) on the assembly of Methanosarcina mazei glutamine synthetase (M mazei GS). Utilizing cutting-edge mass photometry, the authors scrutinized the population dynamics of GS assembly in response to varying concentrations of 2-OG. Notably, the findings demonstrate a promising and straightforward correlation, revealing that dodecamer formation can be stimulated by 2-OG concentrations of up to 10 mM, although GS assembly never reaches 100% dodecamerization in this study. Furthermore, catalytic activities showed a remarkable enhancement, escalating from 0.0 U/mg to 7.8 U/mg with increasing concentrations of 2-OG, peaking at 12.5 mM. However, an intriguing gap arises between the incomplete dodecameric formation observed at 10 mM 2-OG, as revealed by mass photometry, and the continued increase in activity from 5 mM to 10 mM 2-OG for M mazei GS. This prompts questions regarding the inability of M mazei GS to achieve complete dodecamer formation and the underlying factors that further enhance GS activity within this concentration range of 2-OG.

Moreover, the cryo-electron microscopy (cryo-EM) analysis provides additional support for the biophysical and biochemical characterization, elucidating the precise localization of 2-OG at the interface of two GS subunits within two hexameric rings. The observed correlation between GS assembly facilitated by 2-OG and its catalytic activity is substantiated by structural reorientations at the GS-GS interface, confirming the previously reported phenomenon of "funnel activation" in GS. However, the authors did not present the cryo-EM structure of M. mazei GS in complex with ATP and glutamate in the presence of 2-OG, which could have shed light on the differences in glutamine biosynthesis between previously reported GS enzymes and the 2-OG-bound M. mazei GS.

Furthermore, besides revealing the cryo-EM structure of 2-OG-bound GS, the study also observed the filamentous form of GS, suggesting that filament formation may be a universal stacking mechanism across archaeal and bacterial species. However, efforts to enhance resolution to investigate whether the stacked polymer is induced by 2-OG or other factors such as ions or metabolites were not undertaken by the authors, leaving room for further exploration into the mechanisms underlying filament formation in GS.

Reviewer #3 (Public Review):

Summary:

The current manuscript investigates the effect of 2-oxoglutarate and the Glk1 protein as modulators of the enzymatic reactivity of glutamine synthetase. To do this, the authors rely on mass photometry, specific activity measurements, and single-particle cryo-EM data.

From the results obtained, the authors convey that glutamine synthetase from Methanosarcina mazei exists in a non-active monomeric/dimeric form under low concentrations of 2-oxoglutarate, and its oligomerization into a dodecameric complex is triggered by higher concentration of 2-oxoglutarate, also resulting in the enhancement of the enzyme activity.

Strengths:

Glutamine synthetase is a crucial enzyme in all domains of life. The dodecameric fold of GS is recurrent amongst prokaryotic and archaea organisms, while the enzyme activity can be regulated in distinct ways. This is a very interesting work combining protein biochemistry with structural biology.

The role of 2-OG is here highlighted as a crucial effector for enzyme oligomerization and full reactivity.

Weaknesses:

Various opportunities to enhance the current state-of-the-art were missed. In particular, omissions of the ligand-bound state of GnK1 leave unexplained the lack of its interaction with GS (in contradiction with previous results from the authors). A finer dissection of the effect and role of 2-oxoglurate are missing and important questions remain unanswered (e.g. are dimers relevant during early stages of the interaction or why previous GS dodecameric structures do not show 2-oxoglutarate).

Author response:

Reviewer #1 (Public Review):

We thank Reviewer #1 for the professional evaluation and raising important points. We will address those comments in the updated manuscript and especially improve the discussion in respect to the two points of concern.

(1) How can GlnA1 activity further be stimulated with further increasing 2-OG after the dodecamer is already fully assembled at 5 mM 2-OG.

We assume a two-step requirement for 2-OG, the dodecameric assembly and the priming of the active sites. The assembly step is based on cooperative effects of 2-OG and does not require the presence of 2-OG in all 2-OG-binding pockets: 2-OG-binding to one binding pocket also causes a domino effect of conformational changes in the adjacent 2-OG-unbound subunit, as also described for Methanothermococcus thermolithotrophicus GS in Müller et al. 2023. Due to the introduction of these conformational changes, the dodecameric form becomes more favourable even without all 2-OG binding sites being occupied. With higher 2-OG concentrations present (> 5mM), the activity increased further until finally all 2-OG-binding pockets were occupied, resulting in the priming of all active sites (all subunits) and thereby reaching the maximal activity.

(2) The contradictory results with previously published data on the structure of M. mazei by Schumacher et al. 2023.

We certainly agree that it is confusing that Schumacher et al. 2023 obtained a dodecameric structure without the addition of 2-OG, which we claim to be essential for the dodecameric form. 2-OG is a cellular metabolite that is naturally present in E. coli, the heterologous expression host both groups used. Since our main question focused on analysing the 2-OG effect on GS, we have performed thorough dialysis of the purified protein to remove all 2-OG before performing MP experiments. In the absence of 2-OG we never observed significant enzyme activity and always detected a fast disassembly after incubation on ice. We thus assume that a dodecamer without 2-OG in Schuhmacher et al. 2023 is an inactive oligomer of a once 2-OG-bound form, stabilized e.g. by the presence of 5 mM MgCl2.

The GlnA1-GlnK1-structure (crystallography) by Schumacher et al. 2023 is in stark contrast to our findings that GlnK1 and GlnA1 do not interact as shown by mass photometry with purified proteins. A possible reason for this discrepancy might be that at the high protein concentrations used in the crystallization assay, complexes are formed based on hydrophobic or ionic protein interactions, which would not form under physiological concentrations.

Reviewer #2 (Public Review):

We thank Reviewer #2 for the detailed assessment and valuable input. We will address those comments in the updated manuscript and clarify the message.

(1) The discrepancy of the dodecamer formation (max. at 5 mM 2-OG) and the enzyme activity (max. at 12.5 mM 2-OG).

We assume that there are two effects caused by 2-OG: 1. cooperativity of binding (less 2-OG needed to facilitate dodecamer formation) and 2. priming of each active site. See also Reviewer #1 R.1). We assume this is the reason why the activity of dodecameric GlnA1 can be further enhanced by increased 2-OG concentration until all catalytic sites are primed.

(2) The lack of the structure of a 2-OG and ATP-bound GlnA1.

Although we strongly agree that this would be a highly interesting structure, it seems out of the scope of a typical revision to request new cryo-EM structures. We evaluate the findings of our present study concerning the 2-OG effects as important insights into the strongly discussed field of glutamine synthetase regulation, even without the requested additional structures.

(3) The observed GlnA1-filaments are an interesting finding.

We certainly agree with the referee on that point, that the stacked polymers are potentially induced by 2-OG or ions. However, it is out of the main focus of this manuscript to further explore those filaments. Nevertheless, this observation could serve as an interesting starting point for future experiments.

Reviewer #3 (Public Review):

We thank Reviewer #3 for the expert evaluation and inspiring criticism.

(1) Encouragement to examine ligand-bound states of GlnK1.

We agree and plan to perform the suggested experiments exploring the conditions under which GlnA1 and GlnK1 might interact. We will perform the MP experiments in the presence of ATP. In GlnA1 activity test assays when evaluating the presence/effects of GlnK1 on GlnA1 activity, however, ATP was always present in high concentrations and still we did not observe a significant effect of GlnK1 on the GlnA1 activity.

(2) The exact role of 2-OG could have been dissected much better.

We agree on that point and will improve the clarity of the manuscript. See also Reviewer #1 R.1.

(3) The lack of studies on dimers.

This is actually an interesting point, which we did not consider during writing the manuscript. Now, re-analysing all our MP data in this respect, GlnA1 is likely a dimer as smallest species. Consequently, we will add more supplementary data which supports this observation and change the text accordingly.

(4) Previous studies und structures did not show the 2-OG.

We assume that for other structures, no additional 2-OG was added, and the groups did not specifically analyse for this metabolite either. All methanoarchaea perform methanogenesis and contain the oxidative part of the TCA cycle exclusively for the generation of glutamate (anabolism) but not a closed TCA cycle enabling them to use internal 2-OG concentration as internal signal for nitrogen availability. In the case of bacterial GS from organisms with a closed TCA cycle used for energy metabolism (oxidation of acetyl CoA) like e.g. E. coli, the formation of an active dodecameric GS form underlies another mechanism independent of 2-OG. In case of the recent M. mazei GS structures published by Schumacher et al. 2023, the dodecameric structure is probably a result from the heterologous expression and purification from E. coli. (See also Reviewer #1 R.2). One example of methanoarchaeal glutamine synthetases that do in fact contain the 2-OG in the structure, is Müller et al. 2023.

eLife assessment

This landmark work by Lewis and Hegde represents the most significant breakthrough in membrane and secretory biogenesis in recent years. Their work reveals with outstanding clarity how nascent transmembrane segments can pass through the gate of Sec61 into the ER membrane through the coordinated motions of a conformationally and compositionally dynamic machine. Among many other insights, the authors discovered how a new factor, RAMP4, contributes to the formation and function of the lateral gate for certain substrates. The technical quality of the work is exceptional, setting the bar appropriately high.

Reviewer #1 (Public Review):

The paper meticulously explores various conformations and states of the ribosome-translocon complex. Employing advanced techniques such as cryoEM structural determination and AlphaFold modeling, the study delves into the dynamic nature of the ribosome-translocon complex. The findings from these analyses unveil crucial insights, significantly advancing our understanding of the co-translational translocation process in cellular mechanisms.

To begin with, the authors employed a construct comprising the first two transmembrane domains of rhodopsin as a model for studying protein translocation. They conducted in vitro translation, followed by the purification of the ribosome-translocon complex, and determined its cryoEM structures. An in-depth analysis of their ribosome-translocon complex structure revealed that the nascent chain can pass through the lateral gate of translocon Sec61, akin to the behavior of a Signaling Peptide. Additionally, Sec61 was found to interact with 28S rRNA helix 24 and the ribosomal protein uL24. In summary, their structural model aligns with the through-pore model of insertion, contradicting the sliding model.

Secondly, the authors successfully identified RAMP4 in their ribosome-translocon complex structure. Notably, the transmembrane domain of RAMP4 mimics the binding of a Signaling Peptide at the lateral gate of Sec61, albeit without unplugging. Intriguingly, RAMP4 is exclusively present in the non-multipass translocon ribosome-translocon complex, not in those containing multipass translocon. This observation suggests that co-translational translocation specifically occurs in the Sec61 channel that includes bound RAMP4. Additionally, the authors discovered an interaction between the C-tail of ribosomal proteins uL22 and the translocon Sec61, providing valuable insights into the nascent chain's behavior.

Moving on to the third point, the focused classification unveiled TRAP complex interactions with various components. The authors propose that the extra density observed in their novel ribosome-translocon complex can be attributed to calnexin, a major binder of TRAP according to previous studies. Furthermore, the new structure reveals a TRAP-OSTA interaction. This newly identified TRAP-OSTA interaction offers a potential explanation for why patients with TRAP delta defects exhibit congenital disorders of glycosylation.

In conclusion, this paper presents a robust contribution to the field with its thorough structural and modeling analyses. The significance of the findings is evident, providing valuable insights into the intricate mechanisms of protein co-translational translocation. The well-crafted writing, meticulous analyses, and clear figures collectively contribute to the overall strength of the paper.

Reviewer #2 (Public Review):

Summary:

In the manuscript Lewis and Hegde present a structural study of the ribosome-bound multipass translocon (MPT) based on re-analysis of cryo-EM single particle data of ribosome-MPTs processing the multipass transmembrane substrate RhoTM2 from a previous publication (Smalinskaité et al, Nature 2022) and AlphaFold2 multimer modeling. Detailed analysis of the laterally open Sec61 is obtained from PAT-less particles.

The following major claims are made:

- TMs can bind similarly to the Sec61 lateral gate as signal peptides.

- Ribosomal H59 is in immediate proximity to basic residues of TMs and signal peptides, suggesting it may contribute to the positive-inside rule.

- RAMP4/SERP1 binds to the Sec61 lateral gate and the ribosome near 28S rRNA's helices 47, 57, and 59 as well as eL19, eL22, and eL31.

- uL22 C-terminal tail binds H24/47 blocking a potential escape route for nascent peptides to the cytosol.

- TRAP and BOS compete for binding to Sec61 hinge.

- Calnexin TM binds to TRAPg.

- NOMO wedges between TRAP and MPT.

Strengths:

The manuscript contains numerous novel new structural analyses and their potential functional implications. While all findings are exciting, the highlight is the discovery of RAMP4/SERP1 near the Sec61 lateral gate. Overall, the strength is the thorough and extensive structural analysis of the different high-resolution RTC classes as well as the expert bioinformatic evolutionary analysis.

Author response:

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

Reviewer #1 (public review and recommendations for the authors):

Major points:

(1) The identification of RAMP4 is a pivotal discovery in this paper. The sophisticated AlphaFold prediction, de novo model building of RAMP4's RBD domain, and sequence analyses provide strong evidence supporting the inclusion of RAMP4 in the ribosome-translocon complex structure.

However, it is crucial to ensure the presence of RAMP4 in the purified sample. Particularly, a validation step such as western blotting for RAMP4 in the purified samples would strengthen the assertion that the ribosome-translocon complex indeed contains RAMP4. This is especially important given the purification steps involving stringent membrane solubilization and affinity column pull-down.

As suggested, we have added Western blots showing that RAMP4 is retained at secretory translocons (and not multipass translocons) after solubilisation, affinity purification, and recovery of ribosome-translocon complexes (Fig. 3F). This data supports both our assignment of RAMP4 in ribosome-translocon complexes, and also the structure-based proposition that its occupancy is mutually exclusive with the multipass translocon (in particular, the PAT complex).  

(2) Despite the comprehensive analyses conducted by the authors, it is challenging to accept the assertion that the extra density observed in TRAP class 1 corresponds to calnexin. The additional density in TRAP class 1 appears to be less well-resolved, and the evidence for assigning it as calnexin is insufficient. The extra density there can be any proteins that bind to TRAP. It is recommended that the authors examine the density on the ER lumen side. An investigation into whether calnexin's N-globular domain and P-domain are present in the ER lumen in TRAP class 1 would provide a clearer understanding.

We agree that the Calnexin assignment is less confident than the other assignments in this manuscript, and that further support would be ideal. We have exhaustively searched our maps for any unexplained density connected with the putative Calnexin TMD, and have found none. This is consistent with Calnexin's lumenal domain being flexibly linked to its TMD, and thus would not be resolved in a ribosome-aligned reconstruction.

Our assignment of this TMD to Calnexin was based on existing biochemical data (referenced in the paper) favouring this as the best working hypothesis by far: Calnexin is TRAP’s only abundant co-purifying factor, and their interaction is sensitive to point mutations in the Calnexin TMD. Recognising that this is not conclusive, we have ensured that the text and figures consistently describe this assignment as provisional or putative.

(3) In the section titled 'TRAP competes and cooperates with different translocon subunits,' the authors present a compelling explanation for why TRAP delta defects can lead to congenital disorders of glycosylation. To enhance this explanation, it would be valuable if the authors could provide additional analyses based on mutations mentioned in the references. Specifically, examining whether these mutations align with the TRAP delta-OSTA structure models would strengthen the link between TRAP delta defects and the observed congenital disorders of glycosylation.

We agree that mapping disease-causing point mutants to the TRAP delta structure could be potentially informative. Unfortunately, the referenced TRAP delta disease mutants act by simply impairing TRAP delta expression, and thus admit no such fine-grained analyses. However, sequence conservation is our next best guide to mutant function. We note in the text that the contact site charges on TRAP delta and RPN2 are conserved, and that the closest-juxtaposed interaction pair (K117 on TRAPδ and D386 on RPN2) is also the most conserved.

Here are some minor points:

(1) In the introduction, when the EMC, PAT, and BOS complexes were initially mentioned, it would be beneficial for the authors to provide more context or cite relevant references. This additional information will aid readers in better understanding these complexes, ensuring a smoother comprehension of their significance in the context of the study.

The Introduction has been edited to provide more context with relevant references. 

(2) In Figure 7, it would be valuable for the authors to include details on how they sampled the sequence alignments. 

To clarify this methodological point, we have revised the Figure 7 caption to include these sentences: “The logo plots in panels A and D represent an HMM generated by jackHMMER upon convergence after querying UniProtKB’s metazoan sequences with the human TRAPα sequence. Only signal above background is shown, as rendered by Skylign.org.”

Reviewer #2 (public review and recommendations for the authors):

Strengths:

The manuscript contains numerous novel new structural analyses and their potential functional implications. While all findings are exciting, the highlight is the discovery of RAMP4/SERP1 near the Sec61 lateral gate. Overall, the strength is the thorough and extensive structural analysis of the different high-resolution RTC classes as well as the expert bioinformatic evolutionary analysis.

Weaknesses:

A minor downside of the manuscript is the sheer volume of analyses and mechanistic hypotheses, which makes it sometimes difficult to follow. The authors might consider offloading some analyses based on weaker evidence to the supplement to maximize impact.

We agree that the manuscript is long, but we have retained what we feel are the most important findings in the main text because the supplement is often undiscoverable via literature searches. Indeed, we chose eLife for its flexibility regarding article length and suitability for extended and detailed analyses. 

Major:

- Figure S1 does not capture the fact that a PAT-free subset of particles is analyzed. The PAT classification step should be added.

We apologise for having caused some confusion on this point: we do not show a PAT classification step because there was none. Instead we reanalysed the whole dataset with a focus on Sec61 and TRAP. The very little PAT present (9% of particles, per Smalinskaitė et al. 2022) appeared as a very weak density in some of the closed-Sec and weak-TRAP classes.

- The assignment of calnexin appears highly speculative. As the authors acknowledge the EM density is clearly of insufficient resolution for identification, and also AF2 does not render orthogonal support for the interpretation. The binding to TRAPg also does not explain complex formation in lower eukaryotes that do not have TRAPg. The authors may consider moving the calnexin assignment and interpretation to the supplement as it appears highly speculative. In any case, it should not be referred to as a hypothesis and not a structure.

We agree that the Calnexin assignment is less confident than the other assignments in this manuscript, and that further support would be ideal. Our assignment of this TMD to Calnexin was based on existing biochemical data (referenced in the paper) favouring this as the best working hypothesis by far: Calnexin is TRAP’s only abundant co-purifying factor, and their interaction is sensitive to point mutations in the Calnexin TMD. Recognising that this is not conclusive, we have ensured that the text and figures consistently describe this assignment as provisional or putative.

- P. 8: "This extensive competition explains why prior studies found TRAP in only 40% of MPT complexes, but at high occupancy at all other RTCs29". The interpretation is at odds with a recent re-analysis of the same dataset (preprint: Gemmer et al 2023, https://doi.org/10.1101/2023.11.28.569136), which finds TRAP occupancy to negatively correlate with PAT, not BOS.

The reviewer is correct that the Gemmer study demonstrates a negative correlation between PAT and TRAP occupancy, but it does not, as the reviewer claims, argue against a negative correlation between BOS and TRAP. In fact it agrees that Sec61•BOS•PAT complex would clash with TRAP, and that therefore “BOS could trigger release of TRAP from the multipass translocon.” Thus, there is no conflict between the two studies. The revised text in this passage now cites the Gemmer et al. preprint and clarifies that TRAP is partially displaced by competition with BOS, but retained at the translocon via its ribosome-binding domain.  

- P. 7/8: the authors suggest that TRAPd may be important for OSTA recruitment and hence TRAPd deletion may cause glycosylation defects in patients by failure to recruit OSTA. However, cryo-ET studies (Pfeffer et al, Nat. Comms 2017) showed that OSTA still binds in patient-derived microsomes (and the OSTA-TRAPd interaction). The author should discuss their model in the light of these data.

As explained in the text, our hypothesis predicts that TRAPδ is more important for OSTA’s recruitment to the RTC than for its RTC affinity: “OSTA’s attraction to TRAPδ is weak compared to its binding to the ribosome, but TRAPδ may nonetheless help recruit OSTA, since TRAPδ would attract OSTA from most possible angles of approach, whereas OSTA’s ribosome contacts are stereospecific.” Therefore the fact that Pfeffer et al. 2017 found OSTA at some TRAPδ-negative RTCs is not surprising. For confirmation we would look for TRAPδ-dependent glycosylation sites in fast-folding domains or otherwise kinetically sensitive loci, and indeed TRAP-dependence screens return complex profiles that could be consistent with such a mechanism (Phoomak et al. 2021).

- Some confidence measure for the assignment of SERP1/RAMP4 should be provided adding support for the claim "The resolution of the RBD density was sufficient for de novo modelling". Indeed, the N-terminal ribosome-bound segment appears well resolved and programs like Modelangelo or FindMySequence should provide a confidence measure for the assignment of the density to SERP1. The TM part appears less well resolved, but the connectivity to the Nterminus may justify the assignment, which should be elaborated on.

Although we appreciate the value of tools like Modelangelo or FindMySequence, and would have used them if we were resting our assignment of RAMP4 on its RBD alone, we feel that such analyses would be superfluous here. They would quantify only the buildability of RAMP4’s

RBD, whereas the real question of RAMP4’s assignability is independently supported by AlphaFold’s confirmation of RAMP4’s TMD as the Sec61-binding density, and further biochemical data provided or cited in the paper.

- P. 3: "Because PAT complex recruitment and MPT assembly are just beginning, ..." the implicit kinetic model seems to be that the MPT subcomplexes assemble on ribosome and Sec61. What is the evidence for this model and later recruitment of PAT (as opposed to GEL, BOS, and PAT binding pre-assembled)?

The work of Sundaram et al. (PMID 36261522) established that PAT, GEL and BOS do not coassociate appreciably in the absence of the ribosome-Sec61 complex. This is consistent with the structural data in Smalinskaite et al. (PMID 36261528), which shows that PAT, GEL, and BOS each contact the ribosome (and Sec61 in the case of PAT and BOS), but have few if any specific contacts among themselves. Finally, data in both of these studies show that recruitment of each complex to the RNC is not lost when any of them is missing, arguing that each is capable of independent recruitment to ribosome-Sec61 complexes. 

- p. 4: the meaning of the sentence "Stabilising interactions with this widely conserved motif may help Sec61 respond to its diverse substrates with a consistent open state." is not entirely clear. Published single-particle cryo-EM structures of RTC appear to have resulted in various degrees of openness.

Here we were referring not to RTC structures in general, but to substrate-engaged RTCs in particular.  The two substrate-engaged RTC structures under discussion in this paragraph are nearly identical (Figure 2c) despite large differences in substrate sequence (RhoTM2 vs preprolactin’s SP). We were surprised to find that this engaged structure creates noncovalent bonds between the Sec61 N-half and the ribosome. This bonding would tend to stabilise this particular engaged structure, and this stabilisation helps explain why the newly observed TMengaged structure is so similar to the previously observed SP-engaged structure. Without this stabilising N-half interaction, one might instead expect to see more variability, such as the reviewer suggests.

- A recent analysis of heimdallarchaea already hypothesized TRAP in these organisms and should be cited: Eme et al, Nature 618:992-999 (2023). The novel findings of this manuscript compared to Eme et al should be discussed.

We thank the reviewer for bringing this relevant contemporaneous work to our attention. Reviewing the putative TRAP homologs identified by Eme et al, we find that most do not in fact appear to be TRAP homologs at all, judged by the measures used in our work (reciprocal HHpred queries against the human proteome and predicted structural similarity). This is not surprising since Eme et al. relied on low-threshold sequence similarity searches rather than structural measures. To acknowledge this work, we have added a sentence as follows (italics): “To test whether these candidates are also similar to TRAPαβγ in sequence, we used them to perform reciprocal HHpred queries of the human proteome, and in each case the corresponding human TRAP protein was the top hit (E = 0.031 for TRAPα, 9.4×10-14 for TRAP β, and 110 for

TRAPγ). A contemporaneous study has also claimed to find TRAP homologs in

Heimdallarchaeota (Eme et al. 2023), although some caution is warranted in these assignments because they do not seem to share predicted structural similarity to TRAP subunits and do not find human homologs in reciprocal HHpred queries.”

- Given that the authors expand the evolutionary analysis of TRAP to archaea it would be helpful if sampling for RAMP4 were consistent (i.e., is TRAP present in the early eukaryotes that do not feature RAMP4? Is RAMP4 absent from heimdallarchaea?).

As stated in the text, RAMP4’s absence from early-branching eukaryotic taxa indicates that it was also absent from their archaeal ancestors. We did of course run such queries for completeness and indeed find no archaeal RAMP4. TRAP, for its part, is generally present in early-branching eukaryotic taxa, as stated in the text, and this necessarily includes those from which RAMP4 is absent.

- The authors may consider discussing (Gemmer et al 2023, https://doi.org/10.1101/2023.11.28.569136), which comes to similar conclusions for NEMO integration into the MPT.

We thank the reviewer for bringing this relevant work to our attention. We have added the following sentence to the section on NOMO: “Contemporaneous work has arrived at a similar model for PLD10-12 but did not model PLD1 (Gemmer et al. 2023).”

- The abundance approximation of RAMP4 in the native translocon by OccuPy should probably be taken with a grain of salt. The '80%' mentioned in the conclusion may stick around and could eventually turn out to be closer to 100%.

It is certainly possible that the occupancy of RAMP4 is higher than OccuPy estimates.

Unfortunately no available method can provide occupancy estimates with confidence intervals. The Western blots we have added to the revised manuscript are consistent with high occupancy, but cannot discriminate between 80 or 100%.

Minor

- p. 5: The following sentence is incomplete: "Together, these factors explain why RAMP4's occupancy in prior cryo-EM maps was low enough to be overlooked, although in hindsight seems to be visible in several7,68,69"

Thank you for catching this typo. We have revised the sentence as follows: “Together, these factors explain why RAMP4's occupancy in prior cryo-EM maps was low enough to be overlooked, although in hindsight it is visible in several of them.”

eLife assessment

The manuscript describes a valuable method to boost WNT signaling in a tissue-specific manner. The work extends previous data from the authors based on fusing an RSPO2 mutant protein to an antibody that binds ASGR1/2. In the current manuscript, two new antibodies with similar effects are described, that expand this solid approach and provide alternatives for potential future clinical applications. This manuscript will be of interest to all scientists studying protein engineering and cellular targeting.

Reviewer #1 (Public Review):

Summary:

The authors have previously described a way to boost WNT/CTNNB1 signaling in a tissue-specific manner, by directing an RSPO2 mutant protein (RSPO2RA) to a liver-specific receptor (ASGR1/2). This is done by fusing the RSPO2RA to an antibody that binds ASGR1/2.

Here the authors describe two new antibodies, 8M24 and 8G8, with similar effects. 8M24 shows specificity for ASGR1, while 8G8 has broader affinity for mouse/human ASGR1/2.<br /> The authors resolve and describe the crystal structure of the hASGR1CRD:8M24 complex and the hASGR2CRD:8G8 complex in great detail, which help explain the specificities of the 8M24 and 8G8 antibodies. Their epitopes are non-overlapping.<br /> Upon fusion of the antibodies to an RSPO2RA (an RSPO mutant), these antibodies are able to enhance WNT signaling by promoting the ASGR1-mediated clearance of ZNRF3/RNF43, thereby increasing cell surface expression of FZD. This has previously also been shown to be the case for RSPO2RA fused to an anti-ASGR1 antibody 4F3 - and the paper also tests how the antibodies compare to the 4F3 fusion.

Strengths:

(1) One challenge in treating diseases, is the fact that one would like therapeutics to be highly specific - not just in terms of their target (e.g. aimed at a specific protein of interest) but also in terms of tissue specificity (i.e. affecting only tissue X but leaving all others unaffected). This study broadens the collection of antibodies that can be used for this purpose and thus expands a potential future clinical toolbox.

(2) The authors have addressed questions raised after a first round of review, e.g. by showing that ASGR1 is itself indeed ubiquitinated.

Weaknesses:

(1) Some questions remain as to how 8M24 and 8G8 compare to 4F3.

(2) Some questions remain as to the specificity of the approach: the initial goal was not to also downregulate ASGR1 per se, so this targeting to a specific receptor/membrane protein is not trivial and/or neutral.

Author response:

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

Public Reviews:

Reviewer #1 (Public Review):

Weaknesses:

The authors demonstrate that ASGR1 is degraded in response to RSPO2RA-antibody treatment through both the proteasomal and the lysosomal pathway, suggesting that this is due to the RSPO2RA-mediated recruitment of ZNRF3/RNF43, which have E3 ubiquitin ligase activity. The paper doesn't show, however, if ASGR1 is indeed ubiquitinated.

We thank the reviewer for this comment. We have now conducted ASGR1 ubiquitination assays by immunoprecipitation (IP) of ubiquitin in the membrane protein extract, and immunoblotting (IB) ASGR1 after treating HepG2 cells with our SWEETS molecules or controls. The new data demonstrated ubiquitination of ASGR1 with SWEETS treatment (new Fig. S3A and S3B). Additionally, we blocked the potential ubiquitination of ASGR1 by mutating the two lysine residues in the cytoplasmic domain and compared the ASGR1 degradation after SWEETS treatment. The new data show that removing the potential ubiquitylation Lys sites prevented ASGR1 degradation post SWEETS treatment (new Fig. S3C). These new results provide direct evidence that ASGR1 is ubiquitinated to undergo lysosome or proteasome degradation.

The authors conclude that the RSPO2A-Ab fusions can act as a targeted protein degredation platform, because they can degrade ASGR. While I agree with this statement, I would argue that the goal of these Abs would not be to degrade ASGR per se. The argumentation is a bit confusing here. This holds for both the results and the discussion section: The authors focus on the dual role of their agents, i.e. on promoting both WNT signaling AND on degrading ASGR1. They might want to reconsider how they present their data (e.g. it may be interesting to target ASGR1, but one would presumably then like to do this without also increasing WNT responsiveness?).

We thank the reviewer for this comment. As the reviewer states, the initial goal of the RSPO2RA-ab fusions was to generate tissue-specific RSPO mimetics that focus on elimination of E3. As an unintended consequence, we observed enhanced elimination of ASGR as well. While this was unintended, the results did provide POC that when an E3 ligase is brought into proximity of another protein, ubiquitination and degradation of this protein may occur. Additionally, our results highlight that one needs to be careful in fully assessing the impact of bispecific molecules on the intended target as well as unintended targets to understand the potential side effects of such bispecific molecules. We have revised the manuscript to make this more clear, both in the Results and Discussion sections.

Lines 326-331: The authors use a lot of abbreviations for all of the different protein targeting technologies, but since they are hinting at specific mechanisms, it would be better to actually describe the biological activity of LYTAC versus AbTAC/PROTAB/REULR so non-experts can follow.

We thank the reviewer for this suggestion. We have added more details in the Discussion to highlight the different mechanisms of the various systems described.

Can the authors comment on how 8M24 and 8G8 compare to 4F3? The latter seems a bit more specific (ie. lower background activity in the absence of ASGR1 in 5C)? Are there any differences/advances between 8M24 and 8G8 over 4F3? This remains unclear.

These three antibodies bind different regions/epitopes on ASGR. 8M24 and 8G8 bind non-overlapping epitopes on the carbohydrate recognition domain (CRD), while 4F3 binds the stalk region outside of the CRD. This information is in the Results section of the manuscript. We do not believe that the difference in the ASGR binding epitopes contributes to the slight differences in the background activity. The slight differences may be due to differences in the conformation of the antibodies resulting from the differences in their primary sequences, and these differences may not be significant. We have now repeated the experiments in Fig. 5C and 5D to address the reviewer’s next comment on the axis. These new data (new Fig. 5C and 5D) show less background differences between the molecules.

Can the authors ensure that the axes are labelled/numbered similarly for Fig 5B-D? This will make it easier to compare 5C and 5D.

We thank the reviewer for this suggestion. The y-axes in Fig. 5B–D now have the same scale and number format. For Figs. 5C and 5D, we focus on the potency increases of the SWEETS molecules post ASGR1 overexpression.

Reviewer #2 (Public Review):

Weaknesses:

The authors show crystal structures for binding of these antibodies to ASGR1/2, and hypothesize about why specificity is mediated through specific residues. They do not test these hypotheses.

We thank the reviewer for this comment. We did not further test the residue contributions to binding and specificity as this is not the main focus of the current manuscript. We have revised the section and tuned down the claims for specificity.

The authors demonstrate in hepatocyte cell lines that these function as mimetics, and that they do not function in HEK cells, which do not express ASGR1. They do not perform an exhaustive screen of all non-hepatocyte cells, nor do they test these molecules in vivo.

We agree with the reviewer. For the 4F3-based SWEETS molecule, additional in vitro and in vivo specificity characterized were performed and described in Zhang et al., Sci Rep, 2020. Since 8M24 is human specific and 8G8 only weakly interacts with mouse receptors, in vivo experiments in mouse were not performed. While we did not extensively test the 8M24- and 8G8-based SWEETS on additional cell lines or in vivo, we do believe the data presented strongly support the hepatocyte-specific effects of these molecules.

Surprisingly, these molecules also induced loss of ASGR1, which the authors hypothesize is due to ubiquitination and degradation, initiated by the E3 ligases recruited to ASGR1. They demonstrate that inhibition of either the proteasome or lysosome abrogates this effect and that it is dependent on E1 ubiquitin ligases. They do not demonstrate direct ubiquitination of ASGR1 by ZNRF3/RNF43.

We thank the reviewer for this comment. We have now conducted ASGR1 ubiquitination assays by immunoprecipitation (IP) of ubiquitin in the membrane protein extract, and immunoblotting (IB) ASGR1 after treating HepG2 cells with our SWEETS molecules or controls. The new data demonstrate ubiquitination of ASGR1 with SWEETS treatment (new Figs. S3A and S3B). Additionally, we blocked the potential ubiquitination of ASGR1 by mutating the two lysine residues in the cytoplasmic domain and compared the ASGR1 degradation after SWEETS treatment. The new data show that removing the potential ubiquitylation Lys sites prevented ASGR1 degradation post SWEETS treatment (new Fig. S3C). These new results provide direct evidence that ASGR1 is ubiquitinated to undergo lysosome or proteasome degradation.

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

There are multiple instances where articles (i.e. the use of "the") are missing.

We thank the reviewer for this comment. Following the suggestion, the manuscript has gone through a detailed review by an editorial service, and these and other grammatical errors have been corrected.

Reviewer #2 (Recommendations For The Authors):

The best I can think of is to inject these into Wnt reporter mice (or maybe humanized mice) and see if the liver lights up while other tissues do not.

We thank the reviewer for this suggestion. The liver specificity was demonstrated in vivo in our earlier publication (SciRep, 10:13951, 2020) with the 4F3-RSPO2RA molecule. Unfortunately, as the results in this manuscript show, the new ASGR binders 8M24 and 8G8 either do not bind or only weakly interact with mouse receptors. Therefore, the in vivo experiments were not performed here.

You could also consider addressing some of the statements in the manuscript that are currently hypothetical experimentally.

We thank the reviewer for this comment. We did not further test the residues’ contribution to binding and specificity as this is not the main focus of the current manuscript. We have revised the section and tuned down the claims for specificity.

It would be easier to compare the graphs in 5B-D if all Y-axes were the same scale, with the same scientific notation.

We thank the reviewer for this suggestion. The y-axes in Fig. 5B-D now have the same scale and number format. For Figs. 5C and 5D, we focus on the potency increases of the SWEETS molecules post ASGR1 overexpression.

Some of the western blots in Figure 6 do not have antibody/target labels, making them harder to interpret.

All the Western blots antibody/target labels are on the right side of the blots for each panel, we have now made the text bold and thus easier to identify.

Figure 6 and Supplementary Figure 2 are the same I think.

Figure 6 and Supplementary Figure 2 show the same experimental set-up performed on two different cell lines, Fig. 6 is on Huh7 cells and Supplementary Fig. 2 is on HepG2 cells. The results from these two cell lines are quite consistent, making their appearance very similar.

eLife assessment

This important research article provides a novel approach to measure imaginal disc growth and uses this approach to explore the roles of Fat and Dachsous, two conserved protocadherins, in late larval development. The authors have addressed all referee concerns and the evidence supporting the authors' findings overall are compelling.

Reviewer #1 (Public Review):

The manuscript presents novel results on the regulation of Drosophila wing growth by the protocadherins Ds and Fat. The manuscript performs a more careful analysis of disc volume, larval size, and the relationship between the two, in normal and mutant larvae, and after localized knockdown or overexpression of Fat and Ds. Not all of the results are equally surprising given the previous work on Fat, Ds, and their regulation of disc growth, pupariation, and the Hippo pathway, but the presentation and detail of the presented data is new. The most novel results concern the scaling of gradients of Fat and Ds protein during development, a largely unstudied gradient of Fat protein, and using overexpression of Ds to argue that changes in the Ds gradient do not underlie the slowing and halting of cell divisions during development.

Reviewer #2 (Public Review):

This manuscript from Liu et al. examines the role of Fat and Dachsous, two transmembrane proto-cadherins that function both in planar cell polarity and in tissue growth control mediated by the Hippo pathway. The authors developed a new method for measuring growth of the wing imaginal disc during late larval development and then used this approach to examine the effects of disruption of Fat/Dachsous function on disc growth. The authors show that during mid to late third instar the wing imaginal disc normally grows in a linear rather than exponential fashion and that this occurs due to slowing of the mitotic cell cycle as the disc grows during this period. Consistent with their known role in regulating Hippo pathway activity, this slowing of growth is disrupted by loss of Fat/Dachsous function. The authors also observed a previously unreported gradient of Fat protein across the wing blade. However, graded expression of Fat or Dachsous is not necessary for proper growth regulation in the late third instar because ectopic Dachsous expression, which affects gradients of both Dachsous and Fat, has no growth phenotype.

Author response:

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

Response to Reviews

All reviewers were positive about the rigor and impact of our work and offered a number of very helpful suggestions. We have done a number of suggested experiments, whose results have been added to the revision. We have also used their suggestions to improve the clarity and precision with which we describe and interpret our results.

Reviewer 1 found the paper to be clearly written, with novel results, and the conclusions relevant and solid. This review offered many insights and thoughtful suggestions, which we have adopted to greatly improve the manuscript. The referee’s points are listed below with our responses.

The study chooses to examine growth only in the prospective wing blade (the "pouch") rather than the wing disc as a whole. This can create biases, as fat and ds manipulations often cause stronger effects on growth, and on Hippo signaling targets, in the adjacent hinge regions of the disc. So I am curious about this choice. 

Actually, several experiments described in the manuscript measured growth in regions of the wing disc that did not include the pouch (Fig 1 supplement 4). We found that in the second phase of allometric growth, growth of the pouch was greater than growth of the hinge-notum (Fig.1G and Fig 1 supplement 4).  We also looked at the effect of Ds and Fat on growth of the hinge-notum (Fig 4 supplement 1 and Fig 5 supplement 2). Loss of Ds or Fat also affected allometric growth of the pouch differently from their effects on allometric growth of the hinge-notum. We therefore treated analysis of each region independently. Greater focus was given to wing pouch growth because it was in this region that we detected the interesting gradient properties in Fat and Ds expression.

The limitation to the wing region also creates some problems for the measurements themselves. The division between wing and pouch is not a strict lineage boundary, and thus cells can join or leave this region, creating two different reasons for changes in wing pouch size; growth of cells already in the region, or recruitment of cells into or out of the region. The authors do not discuss the second mechanism.

We agree with this assessment that pouch growth can occur via lineage-restricted growth or by recruitment of cells into the region. This has now been clarified in the Introduction and the Discussion with discussion of the second mechanism.

It is not at all clear that the markers for the pouch used by the authors are stable during development. One of these is Vg expression, or the Vg quadrant enhancer. But the Vgexpressing region is thought to increase by recruitment over late second and third instar through a feed-forward mechanism by which Vg-expressing cells induce Vg expression in adjacent cells. In fact, this process is thought to be driven in part by Fat and Ds (Zecca et al 2010). So when the authors manipulate Fat and Ds are they increasing growth or simply increasing Vg recruitment? I would prefer that this limitation be addressed. 

There is the possibility that the feedforward recruitment of disc cells to express Vg leads to some expansion of the measured pouch domain. However, we argue that the recruitment mechanism may not be contributing significantly to the phenomena we measured in this study. 1) We limited our analysis of pouch growth to the third instar stage. In Fig.2, Zecca and Struhl (2007 doi 10.1242/dev.006411) found that recruitment was much stronger in clones induced at first instar rather than third instar, and so they limited their clonal analysis throughout the paper to first instar induced clones. Thus, it is unclear how much the feedforward recruitment mechanism contributes to pouch growth in the mid-to-late third instar. 2) We detected an effect of Ds and Fat on how rapidly the cell cycle slows down over time in pouch cells. The effect is entirely consistent with it having a causal effect on wing pouch growth. For example, nub>Ds(RNAi) causes the average third instar pouch cell to divide ~25% more rapidly than normal, when comparing the slopes in Figure 6. Note that at the beginning of the third instar, the average pouch cell has a similar doubling time whether lacking Ds or not (Figure 6). When we measured the final size of the wing pouch at the end of the third instar, nub>Ds(RNAi) caused the pouch to be ~30% larger than normal (Figure 5). This effect is quite comparable to the effect of Ds RNAi on cell doubling.

To provide more rigorous evidence that the effect of Fat and Ds on cell cycle dynamics is primarily responsible for their effects on wing growth that we measured, we have adapted the simple growth modeling framework from Wartlick et al (2011) and fit our cell cycle measurements made for different genotypes. These fits give us estimates for instantaneous cell growth rates over time, and using these estimates, we simulated the theoretical growth trajectory of the entire wing pouch for wildtype and ds / fat RNAi animals. When we compare these model predictions of wing growth to our pouch volume measurements over time, they agree very well with one another. These

analyses and results are now discussed in the Results and presented in Fig. 6 supplement 2. Overall, it supports a model that Fat and Ds regulate cell cycle dynamics in the wing pouch during third instar and this effect is primarily responsible for Fat and Ds’s effect on overall wing pouch growth in that timeframe. It does not rule out that Fat and Ds might also affect Vg recruitment at third instar, but such effects must be small relative to the primary effect on the cell cycle. It is feasible that Fat and Ds work via the feedforward mechanism at earlier larval stages. We have now discussed all this in detail in the Discussion considering the limitation of recruitment. 

The second pouch marker the authors use is epithelial folding, but this also has problems, as Fat and Ds manipulations change folding. Even in wild type, the folding patterns are complex. For instance, to make folding fit the Vg-QE pattern at late third the authors appear to be jumping in the dorsal pouch between two different sets of folds (Fig 1S2A). The authors also do not show how they use folding patterns in younger, less folded discs, nor provide evidence that the location of the folds are the same and do not shift relative to the cells. They also do not explain how they use folds and measure at later wpp and bpp stages, as the discs unfold and evert, exposing cells that were previously hidden in the folds.

The primary marker we used for the pouch boundary were the folds. We agree with the reviewer that our original description of how we defined the pouch boundary using the folds was inadequate. We now have substantially expanded the Methods section describing how we defined the boundary at all stages using the folds, including a supplementary figure (Fig 1 supplement 2). Importantly, in our measurements, we did not exclude the pouch regions within the folds but included them (see also the next point). Our microscopy detected fluorescence in the folds, and surface rendering allowed us to visualize fold structure and its contents. In younger discs with less folding, we defined the boundary by the location of the Wg inner ring. The folds were more prominent in older L3 larval discs and in the WPP and later stages since the wings had not fully everted yet. Therefore, we used accepted morphological definitions of the pouch boundary from the literature to define the boundaries. We were able to do so even though, as the reviewer notes, the fold architecture evolves as the larvae age. We agree with the reviewer that defining a boundary based on morphology could be error prone, especially prone to systematic error based on age. It is the main reason we directly compared the morphologically defined boundaries to boundaries defined by the Vg quadrant expression domain for many wing discs across all ages. As seen in Fig 1 supplement 3C, the two methods are in strong agreement with one another for discs of all ages. There is a slight overestimate of the pouch boundary using the morphological method, but the error is small (2.5%) and independent of disc size.  

Finally, the authors limit their measurements to cells with exposed apical faces and thus a measurable area but apparently ignore the cells inside the folds. At late third, however, a substantial amount of the prospective wing blade is found within the folds, especially where they are deepest near the A/P compartment boundary. Using the third vein sensory organ precursors as markers, the L3-2 sensillum is found just distal to the fold, the L3-1 and the ACV sensilla are within the fold, and the GSR of the distal hinge is found just proximal to the fold. That puts the proximal half of the central wing blade in the fold, and apparently uncounted in their assays. These cells will however be exposed at wpp and especially bpp stages. How are the authors adjusting for this? 

We apologize for not describing the methods of measurement thoroughly in the original submission. In fact, we did make measurements of cells located within the folds of the wing pouch at all stages. Z stacks of optical sections were collected that transversed the disc, including the folds. Using surface detection algorithms, we could make spatial measurements (xyz distances and areas) of the material within the folds enveloping the apical pouch. Therefore, we could measure the surface area and volume of the wing pouch that included the folds. This was indeed what we did and reported in the original submission. A much more complete description of the process has now been added to the Methods.

On the other hand, we could not reliably measure Fat-GFP or Ds-GFP fluorescence intensity in cells deep in the folds due to light scattering. Therefore, we did not assay the entire gradient across the pouch. Of the cells we did measure, we know their relative distance to the center of the pouch, defined as the intersection of the AP and DV boundaries. Therefore, fluorescence intensities could be directly compared across stages since they were calibrated by the centerpoint of the pouch. We have added text to the Methods to clarify this.

Stabilizing and destabilizing interactions between Fat and Ds- The authors describe a distal accumulation of Fat protein in the wing, and show that this is unlikely to be through Fat transcription. They further try to test whether the distal accumulation depends on destabilization of proximal Fat by proximal Ds by looking at Fat in ds mutant discs. However, the authors do not describe how they take into account the stabilizing effects of heterophilic binding between the extracellular domains (ECDs) of Fat and Ds; without one, the junctional levels and stability of the other is reduced (Ma et al., 2003; Hale et al. 2015). So when they show that the A-P gradient of Fat is reduced in a ds mutant, is this because of the loss of a destabilizing effect of Ds on Fat, as they assume, or is it because all junctional Fat has been destabilized by loss of extracelluarlar binding to Ds? The description of the Fat gradient in Ds mutants is also confusing (see note 6 below), making this section difficult for the reader to follow. 

We did not intend to imply that Ds actively inhibits Fat. We now describe the implications of the result more clearly in the Results and Discussion with reference to the prior Hale and Ma study of heterophilic stabilization. It is worth noting that Ma et al 2003 saw elevated junctional Fat in ds mutant cells if they were surrounded by other ds mutant cells. This is consistent with our results. We also apologize for the confusion in describing the Fat gradient and have reworded the section in the Results to make it more clear.

The authors do not propose or test a mechanism for the proposed destabilization. Fat and Ds bind not only through their ECDs, but binding has now also been demonstrated through their ICDs (Fulford et al. 2023)

We now discuss possible mechanisms in the Discussion and include the Fulford reference in the Results.

Ds gradient scales by volume, rather than cell number - This is an intriguing result, but the authors do not discuss possible mechanisms.

We have now added discussion of possible mechanisms in the Discussion.

Fat and Ds are already known to have autonomous effects on growth and Hippo signaling from clonal analyses and localized knockdowns. One novelty here is showing that localized knockdown does not delay pupariation in the way that whole animal knockdown does, although the mechanism is not investigated. Another novelty is that the authors find stronger wing pouch overgrowth after localized ds RNAi or whole disc loss of fat than after localized fat RNAi, the latter being only 11% larger. The fat RNAi result would have been strengthened by testing different fat RNAi stocks, which vary in their strength and are commonly weaker than null mutations, or stronger drivers such as the ap-gal4 they used for some of their ds-RNAi experiments or use of UAS-dcr2. Another reason for caution is that Garoia (2005) found much stronger overgrowth in fat mutant clones, which were about 75% larger than control clones.

We thank the reviewer for this suggestion. Indeed, the weak effect of Fat RNAi had been due to the specific RNAi driver. We followed the reviewer’s suggestion and tested other RNAi stocks. We had in hand an RNAi driver against GFP that we had found in unrelated studies to be a very potent repressor of GFP expression. Since we had been using a knock-in allele of GFP inserted in frame to Fat throughout this study, we applied nub>Gal4 UAS-GFP RNAi to knock down homozygous Fat-GFP. The effect of the knockdown was very strong, as measured by residual 488nm fluorescence above background autofluorescence after knockdown. Correcting for background autofluorescence, we estimate that only 4.5% of Fat-GFP remained under RNAi conditions (Figure 5 - figure supplement 3). 

Using the more potent RNAi reagent, we repeated the various experiments related to

Fat. We observed a 42% increase in wing pouch growth, which is similar to that of Ds RNAi. We also observed an effect of Fat RNAi on the average cell cycle time of wing pouch cells. There was still a linear coupling between the cell cycle duration and wing pouch size, but the slope of the coupling was smaller with Fat RNAi. This was very similar to what Ds RNAi does to the cell cycle. Therefore, we have replaced the data from the original Fat RNAi experiments with the new data and modified the text throughout the manuscript to describe the new results.

Flattening of Ds gradient does not slow growth. One model suggests that the flattening of the Ds gradient, and thus polarized Ds-Fat binding, account for slowed growth in older discs. The difficulty in the past has been that two ways of flattening the Ds gradient, either removing Ds or overexpressing Ds uniformly, give opposite results; the first increases growth, while the latter slows it. Both experiments have the problem of not just flattening the gradient, but also altering overall levels of Ds-Fat binding, which will likely alter growth independent of the gradients. Here, the authors instead use overexpression to create a strong Ds gradient (albeit a reversely oriented one) that does not flatten, and show that this does not prevent growth from slowing and arresting.

To make sure that this is not some effect caused by using a reverse gradient, one might instead induce a more permanent normally oriented Ds gradient and see if this also does not alter growth; there is a ds Trojan gal4 line available that might work for this, and several other proximal drivers.

Again, we thank the reviewer for this suggestion. We followed the reviewer’s suggestion and generated Trojan-Gal4 mediated overexpression of Ds. The Ds protein gradient was strongly amplified by Trojan-Gal4 but remained normally oriented. However, it only caused a modest (12%) increase in wing pouch volume. It did not significantly alter Fat expression dynamics nor the dynamics of cell cycle duration. This new data has been added to the Results (Fig. 7 and Fig 7 supplement 2) and discussed at length in the text.

Another possible problem is that, unlike previous studies, the authors have not blocked the Four-jointed gradient; Fj alters Fat-Ds binding and might regulate polarity independently of Ds expression. A definitive test would be to perform the tests above in four-joined mutant discs.

We examined a fj null mutant (fjp1/d1) and found that it did not alter final wing pouch size (Fig. 2 - figure supplement 3E). Moreover, neither Fat nor Ds expression were altered in the fj mutant (Fig. 2- figure supplement 3C,D). 

The Discussion of these data should be improved. The authors state in the Discussion "The significance of these dynamics is unclear, but the flattening of the Fat gradient is not a trigger for growth cessation." While the Discussion mentions the effects of Ds on Fat distribution in some detail, this is the only phrase that discusses growth, which is surprising given how often the gradient model of growth control is mentioned elsewhere. The reader would be helped if details are given about what experiment supports this conclusion, the effect on not only growth cessation but cell cycle time, and why the result differs from those of Rogjula 2008 and Willecke 2008 using Ds and Fj overexpression.

We have rewritten the Discussion to better reflect the results and incorporate the reviewer’s criticisms.

The authors spend much of the discussion speculating on the possibility that Fat and Ds control growth by changing the wing's sensitivity to the BMP Dpp. As the manuscript contains no new data on Dpp, this is somewhat surprising. The discussion also ignores Schwank (2011), who argues that Fat and Dpp are relatively independent. There have also been studies showing genetic interactions between Fat and signaling pathways such as Wg (Cho and Irvine 2004) and EGF (Garoia 2005).

We have modified the discussion to be more inclusive of mechanisms connecting Fat and other signaling pathways, and we deleted some of the speculation about Dpp. However, since Dpp is the only known growth factor whose local concentration linearly scales with average cell doubling time (the process we found Ds/Fat regulates), there is a logical connection that readers deserve to know about. Therefore, we have retained some discussion of the hypothesis that the two might be linked through cell cycle duration. It is for future studies to test that hypothesis as it is beyond the scope of this paper.

That said, there are studies that discount the work of Wartlick’s Dpp model, eg. Schwank et al 2012, arguing that Dpp regulates growth permissively by limiting an antigrowth factor, Brinker. We have added this reference and the others in the Discussion to discuss alternative models where Fat/Ds act in parallel to Dpp. 

Wpp and Bpp- First, the charts treat wpp as if it is a fixed number of hours after 5 day larvae, but this will not be true in fat and ds mutants with extended larval life. This should be mentioned.

We have clarified this distinction in the figure legends.

How are the authors limiting bpp to 1 hr from wpp? Prepupa are brown and lack air bubbles, but that spans 5 hours of disc changes from barely everted to fully wing-like.

We deliberately chose 1 hour post WPP because we wanted to measure final wing volume with minimal eversion. We agree with the reviewer’s concerns with calling this BPP and we now call it WPP+1  

"However, growth of the wing pouch ceased at the larva-pupa molt and its size remained constant".

The transition from late third to wpp shown in the figure is not the pupal molt. Unlike in most insects, in Drosophila the larval cuticle is not molted away, it is remodeled during pupariation into the prepupal case. The pupal cuticle is not formed until 6 hr APF, which is why the initial stages are termed pre-pupal. Also, there is at least one more set of cell divisions that occur in later pupal stages (for instance, see recent work from the Buttitta lab).

We have changed the reference of pupal molt to larva-prepupal transition throughout the manuscript.

"In contrast, the notum-hinge exhibited simpler linear-like positive allometric growth (Fig. 1 - figure supplement 3C) 

This oversimplifies, as there is still a strong inflection after the third time point, albeit not as large as with the wing because there is less notal growth.

We have reworded the text as suggested. 

"whereas at the WPP stage, dividing cells were only found in a narrow zone where sensory organ precursor cells undergo two divisions to generate future sensory organs (Fig. 1 - figure supplement 4C-E)."

While there are more dividing cells at the anterior D/V, which will form sensory bristles, there are also dividing cells elsewhere, including in the posterior and scattered through the pouch, where there are no sensory precursors. Sensory organs are limited to the wing margin and the very few campaniform sensilla found on the prospective third vein. The Sens-GFP shown here, meant to identify sensory precursors, does not look much like the Sens expression in Nolo et al 2000. Anterior is on the left in 1S4A-D, but on the right in E.

We thank the reviewer for this observation. Indeed, the Sens-GFP signal in the figure is too broad. This was owing to bleed-through of the PHH3 signal. Since the pattern of dividing cells at the WPP stage has been so well characterized in the literature, as has the pattern of Sens+ cells at that stage (ie, Nolo et al 2000), we have removed these panels and now simply cite the relevant literature.  

"The gradient was asymmetric along the AP axis, being lower at the A margin than the P margin."

The use of "margin" here is a bit confusing, as the term is usually used to describe the wing margin; that is, the D/V compartment boundary in the disc that forms the edge of the wing. Can the authors use a different term? It would also be helpful to point out that the A and P extremes are also, because of the geometry of the disc, the prospective proximal portions of the wing margin, and the hinge, especially since the authors are including the regions proximal to the most distal fold.

We have reworded it as suggested.   

The graphed loss of the Fat A-P gradient between day 5 third and wpp is dramatic. Given that the changes in folding at wpp might alter which cells are being graphed, can the authors show a photo?

We have now included a photo of Fat-GFP at WPP in Fig 2 - figure supplement 2E.

"Since Ds levels are highest and most steep near the margins, perhaps Ds inhibits Fat expression in a dose- or gradient-dependent manner. We also followed Fat-GFP dynamics in the ds mutant. We did not observe the progressive flattening of the FatGFP profile to the WPP wing (Fig. 2 - figure supplement 3A). Instead, the Fat-GFP profile was graded at the WPP stage and flattened somewhat more by the BPP stage (Fig. 2 - figure supplement 3B)."

This description does not tell the reader if there is any less grading of Fat in the ds mutant compared with wild type; instead, it sounds like it is more graded, as gradation continues at wpp. This would then contradict the hypothesis that proximal Ds is required to create the distal Fat gradient.

The Fat signals for the two genotypes are directly comparable as the samples were imaged together with the same microscope settings.  Fig 2M shows that the Fat gradient is less graded compared to the wildtype. We have reworded the text to make this more clear. But this graded expression persists longer into WPP, not the level of gradation. The reason for this is not understood.

The figure, on the other hand, looks like Fat is less graded, although as noted above this could instead be caused by loss of the stable Ds-bound Fat normally found at junctions. 

Fig 2M shows an increase in Fat levels at the proximal regions of the ds mutant pouch, where Ds is normally most concentrated. This makes the overall profile look less graded. 

Confusingly, in the Discussion the authors state: "Loss of Ds affects the Fat gradient such that distribution of Fat is uniformly upregulated to peak levels." There is no mention of "peak levels" in the Results, and no mention of "graded" expression in the Discussion. I am unclear on how the absolute levels are being determined and would be surprised if there were peak levels after loss of Ds-bound Fat from junctions.

The absolute levels between the genotypes were determined by carefully calibrated fluorescence of Fat-GFP from samples imaged at the same time with the same settings. We used the word peak to refer to the highest level of Fat-GFP within a given gradient profile. Clearly, the description is confusing and so we have deleted the word and modified the text to clarify the meaning.

"Interestingly, the reversed Ds gradient caused a change in the Fat gradient (Fig. 7E). Its peak also became skewed to the anterior and did not normally flatten at the WPP stage."

This result contradicts the author's earlier model that proximal Ds destabilizes Fat. Instead, the result fits the stabilization of Fat caused by binding to endogenous or overexpressed Ds or Ds ECD (Ma et al. 2003; Matakatsu and Blair, 2004; 2006; Hale et al. 2015).

We agree that the reversed Ds affects Fat differently than the loss-of-function ds phenotype. We were not intending to propose a model based on the ds mutant, but a simple interpretation of the result. The reversed Ds experiment generates on its own a simple interpretation that is not consistent with the other. This speaks to the complexity of the system. We have changed the text in the Results to make this less confusing.

Reviewer 2 found the paper to provide insights into normal growth of the wing and useful tools for measurement of growth features. This review offered many insights and thoughtful suggestions, which we have adopted to greatly improve the manuscript. The referee’s points are listed below with our responses.

Although the approach used to measure volume is new to this study, the basic finding that imaginal disc growth slows at the mid-third instar stage has been known for some time from studies that counted disc cell number during larval development (Fain and Stevens, 1982; Graves and Schubiger, 1982). Although these studies did not directly measure disc volume, because cell size in the disc is not known to change during larval development, cell number is an accurate measure of tissue volume. However, it is worth noting that the approach used here does potentially allow for differential growth of different regions of the disc.

We had cited the older literature in reference to our results. We have now noted the approach’s usefulness in measuring different disc regions such as the pouch.

Related to point 1, a main conclusion of this study, that cell cycle length scales with growth of the wing, is based on a developmentally limited analysis that is restricted to the mid-third instar larval stage and later (early third instar begins at 72 hr - the authors' analysis started at 84 hr). The previous studies cited above made measurements from the beginning of the 3rd instar and combined them with previous histological analyses of cell numbers starting at the beginning of the 2nd instar. Interestingly, both studies found that cell number increases exponentially from the start of the 2nd instar until mid-third instar, and only after that point does the cell cycle slow resulting in the linear growth reported here. The current study states that growth is linear due to scaling of cell cycle with disc size as though this is a general principle, but from the earlier studies, this is not the case earlier in disc development and instead applies only to the last day of larval life.

We apologize for not making this distinction clearer in the original manuscript. Indeed, growth is initially exponential and shifts to a more linear-like regime in the mid third instar. Our focus in the manuscript is primarily this latter phase. We have now rewritten the text in the Introduction, Results and Discussion to make this very clear. 

While cell number and pouch volume increase exponentially from the start of the 2nd instar, the cell cycle already begins to slow down during the 2nd instar, as found with mitotic index measurements done by Wartlick et al 2011. Using their data to model cell cycle duration as a function of pouch area, we find that during the 2nd instar, cell cycle duration also increases as the size of the wing pouch increases. This is shown in the figure (panel C) below. Note that this relationship appears nonlinear and is quantitatively distinct from the relationship for third instar wing growth.

Author response image 1.

The analysis of the roles of Fat and Dachsous presented here has weaknesses that should be addressed. It is very curious that the authors found that depletion of Fat by RNAi in the wing blade had essentially no effect on growth while depletion of Dachsous did, given that the loss of function overgrowth phenotype of null mutations in fat is more severe than that of null mutations in dachsous (Matakatsu and Blair, 2006). An obvious possibility is that the Fat RNAi transgene employed in these experiments is not very efficient. The authors tried to address this by doubling the dose of the transgene, but it is not clear to me that this approach is known to be effective. The authors should test other RNAi transgenes and additionally include an analysis of growth of discs from animals homozygous for null alleles, which as they note survive to the late larval stages.

We thank the reviewer for this suggestion. Indeed, the weak effect of Fat RNAi had been due to the specific RNAi driver. We followed the reviewer’s suggestion and tested other RNAi stocks. We had in hand an RNAi driver against GFP that we had found in unrelated studies to be a very potent repressor of GFP expression. Since we had been using a knock-in allele of GFP inserted in frame to Fat throughout this study, we applied nub>Gal4 UAS-GFP RNAi to knock down homozygous Fat-GFP. The effect of the knockdown was very strong, as measured by remaining 488nm fluorescence above background fluorescence after knockdown. Correcting for background fluorescence, we estimated that only 4.5% of Fat-GFP remained under RNAi conditions (Figure 5 - figure supplement 3). 

Using the more potent RNAi reagent, we repeated the various experiments related to Fat. We observed a 42% increase in wing pouch growth, which is similar to that of Ds RNAi. We also observed an effect of Fat RNAi on the average cell cycle time of wing pouch cells. There was still a linear coupling between the cell cycle duration and wing pouch size, but the slope of the coupling was smaller with Fat RNAi. This was very similar to what Ds RNAi does to the cell cycle. Therefore, we have replaced the data from the original Fat RNAi experiments with the new data and modified the text throughout the manuscript to describe the new results.

It is surprising that the authors detect a gradient of Fat expression that has not been seen previously given that this protein has been extensively studied. It is also surprising that they find that expression of Nubbin Gal4 is graded across the wing blade given that previous studies indicate that it is uniform (ie. Martín et al. 2004). These two surprising findings raise the possibility that the quantification of fluorescence could be inaccurate. The curvature of the wing blade makes it a challenging tissue to image, particularly for quantitative measurements.

Fat protein expression not being uniform has been observed before but not carefully quantified (see Mao et al., 2009, Strutt and Strutt 2002).  Martin et al. 2004 (doi 10.1242/dev.013) claimed that Nub-Gal4 is uniform without actually measuring it. Please consult Fig 1A and 2A in their paper, which clearly shows stronger expression in the center/distal region of the pouch. 

Regarding systematic errors in quantification, we took great pains to minimize them. We carefully divided the complex folded disc’s z stack into an apical region of interest (ROI) that included the distal domain of the wing pouch and a basal ROI that included the folds encompassing the pouch. We then used a published and widely used surface detection algorithm (ImSAnE) that captures a 3D region of interest (ROI) that can be curved and complex in shape (in z space) because the user creates a surface spline of the ROI. The resulting output treats the ROI as a virtual 2D object. This obviates the need to perform max projections of confocal stacks, which often create artifacts that the reviewer speaks of. Instead, ImSAnE eliminates such artifacts, and it is the gold standard for image processing of ROIs with 3D curvature. 

Moreover, our pipeline does detect uniform expression if it is there. We used a da-Gal4 driver in Fig. 2K,L - this driver is widely acknowledged to be uniformly expressed in tissues of the fly. When it drives a control fluorescent marker (Bazooka-mCherry), our analysis pipeline detects a uniform expression pattern across the wing pouch (Fig. 2L). When the same Gal4 transgene drives Fat-HA in the same tissue, our pipeline detects a graded expression pattern of Fat-HA (Fig. 2L). In fact, this experiment co-expressed both Fat-HA and the control marker in the same disc. Thus, we feel confident that our analysis is not inaccurate.

Author response:

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

eLife assessment

This is a valuable study that develops a new model of the way muscle responds to perturbations, synthesizing models of how it responds to small and large perturbations, both of which are used to predict how muscles function for stability but also how they can be injured, and which tend to be predicted poorly by classic Hill-type models. The evidence presented to support the model is solid, since it outperforms Hill-type models in a variety of conditions. Although the combination of phenomenological and mechanistic aspects of the model may sometimes make it challenging to interpret the output, the work will be of interest to those developing realistic models of the stability and control of movement in humans or other animals.

Reviewer #1 (Public Review):

Muscle models are important tools in the fields of biomechanics and physiology. Muscle models serve a wide variety of functions, including validating existing theories, testing new hypotheses, and predicting forces produced by humans and animals in health and disease. This paper attempts to provide an alternative to Hill-type muscle models that includes contributions of titin to force enhancement over multiple time scales. Due to the significant limitations of Hill-type models, alternative models are needed and therefore the work is important and timely.

The effort to include a role for titin in muscle models is a major strength of the methods and results. The results clearly demonstrate the weaknesses of Hill models and the advantages of incorporating titin into theoretical treatments of muscle mechanics. Another strength is to address muscle mechanics over a large range of time scales.

The authors succeed in demonstrating the need to incorporate titin in muscle models, and further show that the model accurately predicts in situ force of cat soleus (Kirsch et al. 1994; Herzog & Leonard, 2002) and rabbit posts myofibrils (Leonard et al. 2010). However, it remains unclear whether the model will be practical for use with data from different muscles or preparations. Several ad hoc modifications were described in the paper, and the degree to which the model requires parameter optimization for different muscles, preparations and experiment types remains unclear.

I think the authors should state how many parameters require fitting to the data vs the total number of model parameters. It would also be interesting for the authors to discuss challenges associated with modeling ex vivo and in vivo data sets, due to differences in means of stimulation vs. model inputs.

(1) I think the authors should state how many parameters require fitting to the data vs the total number of model parameters.

The total number of model parameters are listed in Table 1. Each parameter has, in addition, references listed for the source of data (if one exists) along with how the data were used (’C’ calculate, ’F’ fit, ’E’ estimated, or ’S’ for scaled) for the specific simulations that appear in this paper. While this is a daunting number of parameters, only a few of these parameters must be updated when modeling a new musculotendon.

Similar to a Hill-type muscle model, at least 5 parameters are needed to fit the VEXAT model to a specific musculotendon: maximum isometric force (fiso), optimal contractile element (CE) length, pennation angle, maximum shortening velocity, and tendon slack length. However, similar to a Hill model, it is only possible to use this minimal set of parameters by making use of default values for the remaining set of parameters. The defaults we have used have been extracted from mammalian muscle (see Table 1) and may not be appropriate for modeling muscle tissue that differs widely in terms of the ratio of fast/slow twitch fibers, titin isoform, temperature, and scale.

Even when these defaults are appropriate, variation is the rule for biological data rather than the exception. It will always be the case that the best fit can only be obtained by fitting more of the model’s parameters to additional data. Standard measurements of the active force-length relation, passive forcelength relation, and force-velocity relations are quite helpful to improve the accuracy of the model to a specific muscle. It is challenging to improve the fit of the model’s cross-bridge (XE) and titin models because the data required are so rare. The experiments of Kirsch et al., Prado et al, and Trombitas et´ al. are unique to our knowledge. However, if more data become available, it is relatively straight forward to update the model’s parameters using the methods described in Appendix B or the code that appears online (https://github.com/mjhmilla/Millard2023VexatMuscle).

We have modified the manuscript to make it clear that, in some circumstances, the burden of parameter identification for the VEXAT model can be as low as a Hill model:

- Section 3: last two sentences of the 2nd paragraph, found at: Page 10, column 2, lines 1-12 of MillardFranklinHerzog v3.pdf and 05 MillardFranklinHerzog v2 v3 diff.pdf

- Table 1: last two sentences of the caption, found at: Page 11 of MillardFranklinHerzog v3.pdf and 05 MillardFranklinHerzog v2 v3 diff.pdf

(2) It would also be interesting for the authors to discuss challenges associated with modeling ex vivo and in vivo data sets, due to differences in means of stimulation vs. model inputs.

All of the experiments simulated in this work are in-situ or ex-vivo. So far the main challenges of simulating any experiment have been quite consistent across both in-situ and ex-vivo datasets: there are insufficient data to fit most model parameters to a specific specimen and, instead, defaults from the literature must be used. In an ideal case, a specimen would have roughly ten extra trials collected so that the maximum isometric force, optimal fiber length, active force-length relation, passive force-length relation (upto ≈ 0_._6_f_oM), and the force-velocity relations could be identified from measurements rather than relying on literature values. Since most lab specimens are viable for a small number of trials (with the exception of cat soleus), we don’t expect this situation to change in future.

However, if data are available the fitting process is pretty straight forward for either in-situ or ex-vivo data: use a standard numerical method (for example non-linear least squares, or the bisection method) to adjust the model parameters to reduce the errors between simulation and experiment. The main difficulty, as described in the previous paragraph, is the availability of data to fit as many parameters as possible for a specific specimen. As such, the fitting process really varies from experiment to experiment and depends mainly on the richness of measurements taken from a specific specimen, and from the literature in general.

Working from in-vivo data presents an entirely different set of challenges. When working with human data, for example, it’s just not possible to directly measure muscle force with tendon buckles, and so it is never completely clear how force is distributed across the many muscles that typically actuate a joint. Further, there is also uncertainty in the boundary condition of the muscle because optical motion capture markers will move with respect to the skeleton. Video fluoroscopy offers a method of improving the accuracy of measured boundary conditions, though only for a few labs due to its great expense. A final boundary condition remains impossible to measure in any case: the geometry and forces that act at the boundaries as muscle wraps over other muscles and bones. Fitting to in-vivo data are very difficult.

While this is an interesting topic, it is tangent to our already lengthy manuscript. Since these reviews are public, we’ll leave it to the motivated reader to find this text here.

Reviewer #2 (Public Review):

This model of skeletal muscle includes springs and dampers which aim to capture the effect of crossbridge and titin stiffness during the stretch of active muscle. While both crossbridge and titin stiffness have previously been incorporated, in some form, into models, this model is the first to simultaneously include both. The authors suggest that this will allow for the prediction of muscle force in response to short-, mid- and long-range stretches. All these types of stretch are likely to be experienced by muscle during in vivo perturbations, and are known to elicit different muscle responses. Hence, it is valuable to have a single model which can predict muscle force under all these physiologically relevant conditions. In addition, this model dramatically simplifies sarcomere structure to enable this muscle model to be used in multi-muscle simulations of whole-body movement.

In order to test this model, its force predictions are compared to 3 sets of experimental data which focus on short-, mid- and long-range perturbations, and to the predictions of a Hill-type muscle model. The choice of data sets is excellent and provide a robust test of the model’s ability to predict forces over a range of length perturbations. However, I find the comparison to a Hill-type muscle model to be somewhat limiting. It is well established that Hill-type models do not have any mechanism by which they can predict the effect of active muscle stretch. Hence, that the model proposed here represents an improvement over such a model is not a surprise. Many other models, some of which are also simple enough to be incorporated into whole-body simulations, have incorporated mechanistic elements which allow for the prediction of force responses to muscle stretch. And it is not clear from the results presented here that this model would outperform such models.

The paper begins by outlining the phenomenological vs mechanistic approaches taken to muscle modelling, historically. It appears, although is not directly specified, that this model combines these approaches. A somewhat mechanistic model of the response of the crossbridges and titin to active stretch is combined with a phenomenological implementation of force-length and force-velocity relationships. This combination of approaches may be useful improving the accuracy of predictions of muscle models and whole-body simulations, which is certainly a worthy goal. However, it also may limit the insight that can be gained. For example, it does not seem that this model could reflect any effect of active titin properties on muscle shortening. In addition, it is not clear to me, either physiologically or in the model, what drives the shift from the high stiffness in short-range perturbations to the somewhat lower stiffness in mid-range perturbations.

(1) It is well established that Hill-type models do not have any mechanism by which they can predict the effect of active muscle stretch.

While many muscle physiologists are aware of the limitations of the Hill model, these limitations are not so well known among computational biomechanists. There are at least two reasons for this gap: there are few comprehensive evaluations of Hill models against several experiments, and some of the differences are quite nuanced. For example, active lengthening experiments can be replicated reasonably well using a Hill model if the lengthening is done on the ascending limb of the force length curve. Clearly the story is quite different on the descending limb as shown in Figure 9. Similarly, as Figure 8 shows, by choosing the right combination of tendon model and perturbation bandwidth it is possible to get reasonably accurate responses from the Hill model to stochastic length changes. Yet when a wide variety of perturbation bandwidths, magnitudes, and tendon models are tested it is clear that the Hill model cannot, in general, replicate the response of muscle to stochastic perturbations. For these reasons we think many of the Hill model’s drawbacks have not been clearly understood by computational biomechanists for many years now.

(2) Many other models, some of which are also simple enough to be incorporated into whole-body simulations, have incorporated mechanistic elements which allow for the prediction of force responses to muscle stretch. And it is not clear from the results presented here that this model would outperform such models.

We agree that it will be valuable to benchmark other models in the literature using the same set of experiments. Hopefully we, or perhaps others, will have the good fortune to secure research funding to continue this benchmarking work. This will, however, be quite challenging: few muscle models are accompanied by a professional-quality open-source implementation. Without such an implementation it is often impossible to reproduce published results let alone provide a fair and objective evaluation of a model.

(3) For example, it does not seem that this model could reflect any effect of active titin properties on muscle shortening.

The titin model described in the paper will provide an enhancement of force during a stretch-shortening cycle. This certainly would be an interesting next experiment to simulate in a future paper.

(4) In addition, it is not clear to me, either physiologically or in the model, what drives the shift from the high stiffness in short-range perturbations to the somewhat lower stiffness in mid-range perturbations.

We can only respond to what drives the frequency dependent stiffness in the model, though we’re quite interested in what happens physiologically. Hopefully that there are some new experiments done to examine this phenomena in the future. In the case of the model, the reasons are pretty straight forward: the formulation of Eqn. 16 is responsible for this shift.

Equation 16 has been formulated so that the acceleration of the attachment point of the XE is driven by the force difference between the XE and a reference Hill model (numerator of the first term in Eqn. 16) which is then low pass filtered (denominator of the first term in Eqn. 16). Due to this formulation the attachment point moves less when the numerator is small, or when the differences in the numerator change rapidly and effectively become filtered out. When the attachment point moves less, more of the CE’s force output is determined by variations in the length of the XE and its stiffness.

On the other hand, the attachment point will move when the numerator of the first term in Eqn. 16 is large, or when those differences are not short lived. When the attachment point moves to reduce the strain in the XE, the force produced by the XE’s spring-damper is reduced. As a result, the CE’s force output is less influenced by variations of the length of the XE and its stiffness.

Reviewer #2 (Recommendations for the Authors):

I find the clarity of the manuscript to be much improved following revision. While I still find the combination of phenomenological and mechanistic approaches to be a little limiting with regards to our understanding of muscle contraction, the revised description of small length changes makes the interpretation much less confusing.

Similarly, while I agree that Hill-type models are widely used their limitations have been addressed extensively and are very well established. Hence, moving forward I think it would be much more valuable to start to compare these newer models to one another rather than just showing an improvement over a Hill model under (very biologically important) conditions which that model has no capacity to predict forces.

(1) While I still find the combination of phenomenological and mechanistic approaches to be a little limiting with regards to our understanding of muscle contraction ...

We have had to abstract some of the details of reality to have a model that can be used to simulate hundreds of muscles. In contrast, FiberSim produced by Kenneth Campbell’s group uses much less abstraction and might be of greater interest to you. FiberSim’s models include individual cross-bridges, titin molecules, and an explicit representation of the spatial geometry of a sarcomere. While this model is a great tool for testing muscle physiology questions through simulation, it is computationally expensive to use this model to simulate hundreds of muscles simultaneously.

Kosta S, Colli D, Ye Q, Campbell KS. FiberSim: A flexible open-source model of myofilament-level contraction. Biophysical journal. 2022 Jan 18;121(2):175-82.https://campbell-muscle-lab.github.io/FiberSim/

(2) Similarly, while I agree that Hill-type models are widely used their limitations have been addressed extensively and are very well established.

Please see our response 1 to Reviewer # 1.

(3) Hence, moving forward I think it would be much more valuable to start to compare these newer models to one another rather than just showing an improvement over a Hill model under (very biologically important) conditions which that model has no capacity to predict forces.

Please see our response to 2 to Reviewer #1.

eLife assessment

This is a valuable study that develops a new model of the way muscle responds to perturbations, synthesizing models of how it responds to small and large perturbations, both of which are used to predict how muscles function for stability but also how they can be injured, and which tend to be predicted poorly by classic Hill-type models. The evidence presented to support the model is solid, since it outperforms Hill-type models in a variety of conditions. Although the combination of phenomenological and mechanistic aspects of the model may sometimes make it challenging to interpret the output, the work will be of interest to those developing realistic models of the stability and control of movement in humans or other animals.

Reviewer #1 (Public Review):

Muscle models are important tools in the fields of biomechanics and physiology. Muscle models serve a wide variety of functions, including validating existing theories, testing new hypotheses, and predicting forces produced by humans and animals in health and disease. This paper attempts to provide an alternative to Hill-type muscle models that includes contributions of titin to force enhancement over multiple time scales. Due to the significant limitations of Hill-type models, alternative models are needed and therefore the work is important and timely.

The effort to include a role for titin in muscle models is a major strength of the methods and results. The results clearly demonstrate the weaknesses of Hill models and the advantages of incorporating titin into theoretical treatments of muscle mechanics. Another strength is to address muscle mechanics over a large range of time scales.

The authors succeed in demonstrating the need to incorporate titin in muscle models, and further show that the model accurately predicts in situ force of cat soleus (Kirsch et al. 1994; Herzog & Leonard, 2002) and rabbit posts myofibrils (Leonard et al. 2010). However, it remains unclear whether the model will be practical for use with data from different muscles or preparations. Several ad hoc modifications were described in the paper, and the degree to which the model requires parameter optimization for different muscles, preparations and experiment types remains unclear.

Reviewer #2 (Public Review):

This model of skeletal muscle includes springs and dampers which aim to capture the effect of crossbridge and titin stiffness during the stretch of active muscle. While both crossbridge and titin stiffness have previously been incorporated, in some form, into models, this model is the first to simultaneously include both. The authors suggest that this will allow for the prediction of muscle force in response to short-, mid- and long-range stretches. All these types of stretch are likely to be experienced by muscle during in vivo perturbations, and are known to elicit different muscle responses. Hence, it is valuable to have a single model which can predict muscle force under all these physiologically relevant conditions. In addition, this model dramatically simplifies sarcomere structure to enable this muscle model to be used in multi-muscle simulations of whole-body movement.

In order to test this model, its force predictions are compared to 3 sets of experimental data which focus on short-, mid- and long-range perturbations, and to the predictions of a Hill-type muscle model. The choice of data sets is excellent and provide a robust test of the model's ability to predict forces over a range of length perturbations. However, I find the comparison to a Hill-type muscle model to be somewhat limiting. It is well established that Hill-type models do not have any mechanism by which they can predict the effect of active muscle stretch. Hence, that the model proposed here represents an improvement over such a model is not a surprise. Many other models, some of which are also simple enough to be incorporated into whole-body simulations, have incorporated mechanistic elements which allow for the prediction of force responses to muscle stretch. And it is not clear from the results presented here that this model would outperform such models.

The paper begins by outlining the phenomenological vs mechanistic approaches taken to muscle modelling, historically. It appears, although is not directly specified, that this model combines these approaches. A somewhat mechanistic model of the response of the crossbridges and titin to active stretch is combined with a phenomenological implementation of force-length and force-velocity relationships. This combination of approaches may be useful improving the accuracy of predictions of muscle models and whole-body simulations, which is certainly a worthy goal. However, it also may limit the insight that can be gained. For example, it does not seem that this model could reflect any effect of active titin properties on muscle shortening. In addition, it is not clear to me, either physiologically or in the model, what drives the shift from the high stiffness in short-range perturbations to the somewhat lower stiffness in mid-range perturbations.

Reviewer #1 (Public Review):

Summary:<br /> In the paper by Choi et al., the authors aimed to develop base editing strategies to convert CAG repeats to CAA repeats in the huntingtin gene (HTT), which causes Huntington's disease (HD). They hypothesized that this conversion would delay disease onset by shortening the uninterrupted CAG repeat. Using HEK-293T cells as a model, the researchers employed cytosine base editors and guide RNAs (gRNAs) to efficiently convert CAG to CAA at various sites within the CAG repeat. No significant indels, off-target edits, transcriptome alterations, or changes in HTT protein levels were detected. Interestingly, somatic CAG repeat expansion was completely abolished in HD knock-in mice carrying CAA-interrupted repeats.

Strengths:<br /> This study represents the first proof-of-concept exploration of the cytosine base editing technique as a potential treatment for HD and other repeat expansion disorders with similar mechanisms.

Weaknesses:<br /> Given that HD is a neurodegenerative disorder, it is crucial to determine the efficiency of the base editing strategies tested in this manuscript and their feasibility in relevant cells affected by HD and the brain, which needed to be improved in this manuscript.

Reviewer #2 (Public Review):

Summary:<br /> In a proof-of-concept study with the aspiration of developing a treatment to delay HD onset, Choi et al. design and test an A>G DNA base editing strategy to exploit the recently established inverse relationship between the number of uninterrupted CAG repeats in polyglutamine repeat expansions and the age-of-onset of Huntington's Disease (HD). Most of the study is devoted to optimizing a base editing strategy typified by BE4max and gRNA2. The base editing is performed in human HEK293 cells engineered with a 51 CAG canonical repeat and in HD knock-in mice harboring 105+ CAG repeats.

Weaknesses:<br /> Genotypic data on DNA editing are not portrayed in a clear manner consistent with the study's goal, namely reducing the number of uninterrupted CAG repeats by a clinically relevant amount according to the authors' least square approximated mean age-at-onset. No phenotypic data are presented to show that editing performed in either model would lead to reduced hallmarks of HD onset.

More evidence is needed to support the central claims and therapeutic potential needs to be more adequate.

Reviewer #3 (Public Review):

Summary:<br /> In human patients with Huntington's disease (HD), caused by a CAG repeat expansion mutation, the number of uninterrupted CAG repeats at the genomic level influences age-at-onset of clinical signs independent of the number of polyglutamine repeats at the protein level. In most patients, the CAG repeat terminates with a CAA-CAG doublet. However, CAG repeat variants exist that either do not have that doublet or have two doublets. These variants consequently differ in their number of uninterrupted CAG repeats, while the number of glutamine repeats is the same as both CAA and CAG codes for glutamine. The authors first confirm that a shorter uninterrupted CAG repeat number in human HD patients is associated with developing the first clinical signs of HD later. They predict that introducing a further CAA-CAG doublet will result in years of delay of clinical onset. Based on this observation, the authors tested the hypothesis that turning CAG to CAA within a CAG repeat sequence using base editing techniques will benefit HD biology. They show that, indeed, in HD cell models (HEK293 cells expressing 16/17 CAG repeats; a single human stem cell line carrying a CAG repeat expansion in the fully penetrant range with 42 CAG repeats), their base editing strategies do induce the desired CAG-CAA conversion. The efficiency of conversion differed depending on the strategy used. In stem cells, delivery posed a problem, so to test allele specificity, the authors then used a HEK 293 cell line with 51 CAG repeats on the expanded allele. Conversion occurred in both alleles with huntingtin protein and mRNA levels; transcriptomics data was unchanged. In knock-in mice carrying 110 CAG repeats, however, base editing did not work as well for different, mainly technical, reasons.

Strengths:<br /> The authors use state-of-the-art methods and carefully and thoroughly designed experiments. The data support the conclusions drawn. This work is a very valuable translation from the insight gained from large GWAS studies into HD pathogenesis. It rightly emphasises the potential this has as a causal treatment in HD, while the authors also acknowledge important limitations.

Weaknesses:<br /> They could dedicate a little more to discussing several of the mentioned challenges. The reader will better understand where base editing is in HD currently and what needs to be done before it can be considered a treatment option. For instance,

-It is important to clarify what can be gained by examining again the relationship between uninterrupted CAG repeat length and age-at-onset. Could the authors clarify why they do this and what it adds to their already published GWAS findings? What is the n of datasets?<br /> -What do they think an ideal conversion rate would be, and how that could be achieved?<br /> -Is there a dose-effect relationship for base editing, and would it be realistic to achieve the ideal conversion rate in target cells, given the difficulties described by the authors in differentiated neurons from stem cells?<br /> - The liver is a good tool for in-vivo experiments examining repeat instability in mouse models. However, the authors could comment on why they did not examine the brain.<br /> - Is there a limit to judging the effects of base editing on somatic instability with longer repeats, given the difficulties in measuring long CAG repeat expansions?<br /> - Given the methodological challenges for assessing HTT fragments, are there other ways to measure the downstream effects of base editing rather than extrapolate what it will likely be?<br /> - Sequencing errors could mask low-level, but biologically still relevant, off-target effects (such as gRNA-dependent and gRNA-independent DNA, Off-targets, RNA off-targets, bystander editing). How likely is that?<br /> - How worried are the authors about immune responses following base editing? How could this be assessed?

Author response:

Public Reviews:

Reviewer #1 (Public Review):

Summary: 

In the paper by Choi et al., the authors aimed to develop base editing strategies to convert CAG repeats to CAA repeats in the huntingtin gene (HTT), which causes Huntington's disease (HD). They hypothesized that this conversion would delay disease onset by shortening the uninterrupted CAG repeat. Using HEK-293T cells as a model, the researchers employed cytosine base editors and guide RNAs (gRNAs) to efficiently convert CAG to CAA at various sites within the CAG repeat. No significant indels, off-target edits, transcriptome alterations, or changes in HTT protein levels were detected. Interestingly, somatic CAG repeat expansion was completely abolished in HD knock-in mice carrying CAA-interrupted repeats. 

Correction of factual errors

We analyzed HEK293 cells, not "HEK-293T".

Strengths: 

This study represents the first proof-of-concept exploration of the cytosine base editing technique as a potential treatment for HD and other repeat expansion disorders with similar mechanisms. 

Weaknesses: 

Given that HD is a neurodegenerative disorder, it is crucial to determine the efficiency of the base editing strategies tested in this manuscript and their feasibility in relevant cells affected by HD and the brain, which needed to be improved in this manuscript. 

We appreciate the reviewer's constructive recommendations. Our genetic investigation focused on understanding observations in HD patients to develop genetic-based treatment strategies and test their feasibility. We agree with the reviewer regarding the importance of data from relevant cell types. Unfortunately, the levels of CAG-to-CAA conversion in the patient-derived neurons were modest, as described in our manuscript (approximately 2%). In addition, AAV did not produce detectable conversions in the brain of HD knock-in mice (data not shown), which was somewhat expected from the literature (PMID: 31937940). We believe some technical hurdles can be overcome by developing efficient delivery methods. Nonetheless, it will be an important follow-up study to perform preclinical studies employing optimized base editing strategies and efficient brain delivery methods to fully demonstrate the therapeutic potential of BE strategies. 

Reviewer #2 (Public Review):

Summary: 

In a proof-of-concept study with the aspiration of developing a treatment to delay HD onset, Choi et al. design and test an A>G DNA base editing strategy to exploit the recently established inverse relationship between the number of uninterrupted CAG repeats in polyglutamine repeat expansions and the age-of-onset of Huntington's Disease (HD). Most of the study is devoted to optimizing a base editing strategy typified by BE4max and gRNA2. The base editing is performed in human HEK293 cells engineered with a 51 CAG canonical repeat and in HD knock-in mice harboring 105+ CAG repeats. 

Correction of factual errors

We tested base editing strategies aimed at C > T conversion, not A > G DNA base editing. In addition to HEK293 and knock-in mice, we tested base editing strategies in patient-derived iPSC and neurons.

Weaknesses: 

Genotypic data on DNA editing are not portrayed in a clear manner consistent with the study's goal, namely reducing the number of uninterrupted CAG repeats by a clinically relevant amount according to the authors' least square approximated mean age-at-onset. No phenotypic data are presented to show that editing performed in either model would lead to reduced hallmarks of HD onset. 

More evidence is needed to support the central claims and therapeutic potential needs to be more adequate. 

Our strategies for converting CAG to CAA in model systems resulted in quantitative DNA modification in a population of cells. Consequently, individual cells may carry different genotypes, some harboring CAA and others CAG at the same genomic location. Therefore, using a standard genotype format for DNA to present base editing outcomes may not be ideal. Instead, we presented the resulting genotype data in a quantitative fashion to provide the percentage of conversion at each site. This approach allows for an intuitive interpretation of both the extent of repeat length reduction and the proportion of such modifications.

Currently, genetically precise HD mouse models with robust motor and behavioral phenotypes are unavailable. While some HD mouse models, such as the BAC and YAC models, feature pronounced behavioral phenotypes, they consist of interrupted CAG repeat sequences, making them unsuitable for base conversion studies due to their inherently short uninterrupted repeats. Although genetically precise HD knockin mouse models exist, they do not manifest motor symptom-like phenotypes. Given that CAG repeat expansion is the primary driver of the disease and knock-in mice recapitulate such phenomenon, our genetic investigation focused on assessing the effects of base conversion on CAG repeat instability in knock-in mice. However, as emphasized by the reviewer, subsequent preclinical studies to evaluate the therapeutic efficacy of CAG-to-CAA conversion strategies using mouse models harboring uninterrupted adult-onset CAG repeats and robust HD-like phenotypes remain crucial.

Reviewer #3 (Public Review):

Summary: 

In human patients with Huntington's disease (HD), caused by a CAG repeat expansion mutation, the number of uninterrupted CAG repeats at the genomic level influences age-at-onset of clinical signs independent of the number of polyglutamine repeats at the protein level. In most patients, the CAG repeat terminates with a CAACAG doublet. However, CAG repeat variants exist that either do not have that doublet or have two doublets. These variants consequently differ in their number of uninterrupted CAG repeats, while the number of glutamine repeats is the same as both CAA and CAG codes for glutamine. The authors first confirm that a shorter uninterrupted CAG repeat number in human HD patients is associated with developing the first clinical signs of HD later. They predict that introducing a further CAA-CAG doublet will result in years of delay of clinical onset. Based on this observation, the authors tested the hypothesis that turning CAG to CAA within a CAG repeat sequence using base editing techniques will benefit HD biology. They show that, indeed, in HD cell models (HEK293 cells expressing 16/17 CAG repeats; a single human stem cell line carrying a CAG repeat expansion in the fully penetrant range with 42 CAG repeats), their base editing strategies do induce the desired CAG-CAA conversion. The efficiency of conversion differed depending on the strategy used. In stem cells, delivery posed a problem, so to test allele specificity, the authors then used a HEK 293 cell line with 51 CAG repeats on the expanded allele. Conversion occurred in both alleles with huntingtin protein and mRNA levels; transcriptomics data was unchanged. In knock-in mice carrying 110 CAG repeats, however, base editing did not work as well for different, mainly technical, reasons. 

Correction of factual errors

"HD cell models (HEK293 cells expressing 16/17 CAG repeats" is an incorrect description. It should be "HD cell models (HEK293 cells expressing 51/17 CAG repeats".

Strengths: 

The authors use state-of-the-art methods and carefully and thoroughly designed experiments. The data support the conclusions drawn. This work is a very valuable translation from the insight gained from large GWAS studies into HD pathogenesis. It rightly emphasises the potential this has as a causal treatment in HD, while the authors also acknowledge important limitations. 

Weaknesses: 

They could dedicate a little more to discussing several of the mentioned challenges. The reader will better understand where base editing is in HD currently and what needs to be done before it can be considered a treatment option. For instance, 

- It is important to clarify what can be gained by examining again the relationship between uninterrupted CAG repeat length and age-at-onset. Could the authors clarify why they do this and what it adds to their already published GWAS findings? What is the n of datasets? 

Published HD GWAS (PMID: 31398342) compared the onset age of duplicated interruption and loss of interruption to that of canonical repeats to determine whether uninterrupted CAG repeat or polyglutamine determines age at onset. However, GWAS findings did not quantify the magnitude of the unexplained remaining variance in age at onset in duplicated interruption and loss of interruption. Our study further investigated to gain insights into the amount of additional impact of duplicated interruption to estimate the maximum clinical benefits of base editing strategies for CAG-to-CAA conversion. Since the purpose of this genetic analysis is described in the result section already, we added the following sentence in the introduction section to bring up what is unknown. 

"Still, age at onset of loss of interruption and duplicated interruption was not fully accounted for by uninterrupted CAG repeat, suggesting additional effects of non-canonical repeats."

We added sample size for the least square approximation analysis in the text and corresponding figure legend. Sample sizes for molecular and animal experiments can be found in the corresponding figure legend.

- What do they think an ideal conversion rate would be, and how that could be achieved? 

It is a very important question. However, speculating the ideal conversion levels is out of the scope of this genetic investigation. A series of preclinical studies using relevant models may generate data that may shed light on the conversion rate levels that are required to produce meaningful clinical benefits. In the discussion section, we added the following sentence. 

"Currently, the ideal levels of CAG-to-CAA conversion that produce significant clinical benefits are unknown. A series of preclinical studies using relevant model systems may generate data that may shed light on the optimal conversion rate levels that are required to produce significant clinical benefits."

- Is there a dose-effect relationship for base editing, and would it be realistic to achieve the ideal conversion rate in target cells, given the difficulties described by the authors in differentiated neurons from stem cells? 

We observed a clear dose-response relationship between the amount of BE reagents and the levels of conversion in non-neuronal cells. Unfortunately, the conversion rate was low in neuronal cells, potentially due to limited delivery, as speculated in the result section. As described in the discussion sections, we predict that efficient delivery methods will be crucial to produce significant CAG-to-CAA conversion to achieve therapeutic benefits.

- The liver is a good tool for in-vivo experiments examining repeat instability in mouse models. However, the authors could comment on why they did not examine the brain.

We focused on liver instability because of 1) the expectation that delivery/targeting efficiency is significantly lower in the brain (PMID: 31937940) and 2) shared underlying mechanisms between the brain and liver (described in the result section). The following sentence was added in the method section to provide a rationale for liver analysis. 

"Since significantly lower delivery/targeting efficiency was expected in the brain 34, we focused on analyzing liver instability."

- Is there a limit to judging the effects of base editing on somatic instability with longer repeats, given the difficulties in measuring long CAG repeat expansions? 

Determining the levels of base conversion using sequencing technologies gets harder as repeats become longer. Fragment analysis can overcome such technical difficulty if conversion efficiency is high. As pointed out, the repeat expansion measure is also challenging because amplification is biased toward shorter alleles. However, if repeat sizes are relatively similar, the levels of repeat expansion as a function of base conversion can be determined relatively precisely without a significant bias by a standard fragment analysis approach. 

- Given the methodological challenges for assessing HTT fragments, are there other ways to measure the downstream effects of base editing rather than extrapolate what it will likely be?

Our CAG-to-CAA conversion strategies are not expected to directly generate fragments of huntingtin DNA, RNA, or protein. In contrast, immediate downstream effects of CAG-to-CAA conversion include sequence changes (DNA and RNA) and alteration of repeat instability, which are presented in the manuscript. If repeat instability is associated with HTT exon 1A fragment, base conversion strategies may indirectly alter the levels of such putative toxic species, which remains to be determined.  

- Sequencing errors could mask low-level, but biologically still relevant, off-target effects (such as gRNAdependent and gRNA-independent DNA, Off-targets, RNA off-targets, bystander editing). How likely is that? 

We agree with the reviewer that increased editing efficiency is expected to increase the levels of off-target editing. However, the field is actively developing base editors with minimal off-target effect (PMID: 35941130), which will increase the safety aspects of this technology for clinical use. We added the following sentence.  "In addition, developing base editors with high level on-target gene specificity and minimal off-target effects is a critical aspect to address 100."

- How worried are the authors about immune responses following base editing? How could this be assessed? 

We added the following sentence in the discussion section as the reviewer raised an important safety issue.  

"Thorough assessments of immune responses against base editing strategies (e.g., development of antibody, B cell, and T cell-specific immune responses) and subsequent modification (e.g., immunosilencing) 101 will be critical to address immune response-associated safety issues of BE strategies."

Recommendations for the authors:

Reviewer #1 (Recommendations For The Authors):

The following points could be considered to improve the overall quality of the manuscript: 

(1) The authors mentioned that the reason for checking repeat instability in the nonneuronal cells was due to the availability of specific types of AAV; there are other subtypes of AAVs available to infect neurons and iPSCs. 

Our pilot experiments testing several AAV serotypes in patient-derived iPSC and HD knock-in mice showed that only AAV9 converted CAG to CAA at detectable levels in the liver, not in the brain or neurons. We also speculate that difficulties in targeting the CAG repeat region due to GC-rich sequence contributed to low conversion efficiency. Therefore, subsequent optimization of base editor and delivery may improve BE strategies for HD, permitting robust conversion at the challenging locus. 

(2) Despite its bold nature, minimal data in the manuscript demonstrate that this gene editing strategy is disease-modifying.

Resources required to demonstrate the therapeutic benefits of CAG-to-CAA conversion strategies are not fully available. Especially, relevant HD mouse models that carry uninterrupted adult onset CAG repeat and that permit measuring the levels of disease-modifying are lacking, as described in our response to the second reviewer. Given that CAG repeat expansion is the primary driver of the disease, this genetic investigation focused on determining the impacts of base editing strategies on CAG repeat expansion. Still, as indicated by the reviewer, follow-up preclinical studies to evaluate the levels of disease-modifying of CAG-to-CAA conversion strategies using relevant mouse models represent important next steps.

(3) Off-target analysis at the DNA level was limited to "predicted" off-target sites. What about possible translocations that can result from co-nicking on different chromosomes, as a large number of potential targets exist? 

Among gRNAs we tested, we focused on gRNAs 1 and 2, which predicted small numbers of off-target. Therefore, our off-target analysis at the DNA level was focused on validating those predicted off-targets. As pointed out, thoroughly evaluating off-target effects will be necessary when candidate BE strategies take the next steps for therapeutic development.

Genomic translocation caused by double-strand breaks can produce negative consequences, such as cancer. Importantly, although paired nicks efficiently induced translocations, translocations were not detected when a single nick was introduced on each chromosome (PMID: 25201414). Therefore, it is predicted that BE strategies using nickase confers little risk of translocation.

(4) For in vivo work, somatic repeat expansion was analyzed only in peripheral tissue samples. Since the main affected cellular population in HD is the brain, the outcome of this treatment on a disease-relevant organ still needs to be determined. 

Challenges in delivery to the brain made us determine instability in the liver since many mechanistic components of somatic CAG repeat instability are shared between the liver and striatum, as rationalized in the manuscript. However, we agree with the reviewer regarding the importance of determining the effects of base conversion on brain instability. We added the following sentence in the method section to provide a rationale. "Since significantly lower delivery/targeting efficiency was expected in brain 34, we focused on analyzing liver instability."

Reviewer #2 (Recommendations For The Authors):

Throughout the manuscript, the authors apologize for techniques that do not work when workarounds seem readily apparent to an expert in the field. In its current form, the manuscript reads verbose, speculative, apologetic, and preliminary. 

Drug development programs that are supported by human genetics data show increased success rates in clinical trials (PMID: 26121088, 31827124, 31830040). This is why this genetic study focused on 1) investigating observations in HD subjects and 2) subsequently developing treatment strategies that are supported by patient genetics. As the first illustration of base editing in HD, the main scope of our manuscript is to justify the genetic rationale of CAG-to-CAA conversion and demonstrate the feasibility of therapeutic strategies rooted in patient genetics. As our study was not aimed at entirely demonstrating the clinical benefits of base editing strategies in HD, some of our data were based on tools and approaches that were not fully optimal. We agree with the reviewer that it will be an important next step to employ optimized approaches to evaluate the efficacy of base editing strategies in model systems. Nevertheless, our novel base conversion strategies derived from HD patient genetics represent a significant advancement as they may contribute to developing effective treatments for this devastating disorder. 

Reviewer#3 (Recommendations For The Authors):

It would make for an easier read if abbreviations were kept to a minimum. 

As recommended, we decreased the use of abbreviations. The following has been spelled out throughout the manuscript: CR (canonical repeat), LI (loss of interruption), DI (duplicated interruption), and CBE (cytosine base editor). Other abbreviations with infrequent usage (e.g., ABE, SS, QC) were also spelled out in the text.

eLife assessment

This study provides a valuable contribution to our understanding of the mechanisms underlying the limited capacity to process rapid sequences of visual stimuli by reporting convincing evidence that the attentional blink affects neurally separable processes of visual detection and discrimination. The motivation for some of the analyses and the connection to previous empirical and theoretical work can be improved. The study will be of interest to neuroscientists and psychologists investigating perception and attention.

Reviewer #1 (Public Review):

Summary:

In this study, the authors used a multi-alternative decision task and a multidimensional signal-detection model to gain further insight into the cause of perceptual impairments during the attentional blink. The model-based analyses of behavioural and EEG data show that such perceptual failures can be unpacked into distinct deficits in visual detection and discrimination, with visual detection being linked to the amplitude of late ERP components (N2P and P3) and discrimination being linked to the coherence of fronto-parietal brain activity.

Strengths:

The main strength of this paper lies in the fact that it presents a novel perspective on the cause of perceptual failures during the attentional blink. The multidimensional signal-detection modelling approach is explained clearly, and the results of the study show that this approach offers a powerful method to unpack behavioural and EEG data into distinct processes of detection and discrimination.

Weaknesses:

While the model-based analyses are compelling, the paper also features some analyses that seem misguided, or, at least, insufficiently motivated and explained. Specifically, in the introduction, the authors raise the suggestion that the attentional blink could be due to a reduction in sensitivity or a response bias. The suggestion that a response bias could play a role seems misguided, as any response bias would be expected to be constant across lags, while the attentional blink effect is only observed at short lags. Thus, it is difficult to understand why the authors would think that a response bias could explain the attentional blink.

A second point of concern regards the way in which the measures for detection and discrimination accuracy were computed. If I understand the paper correctly, a correct detection was defined as either correctly identifying T2 (i.e., reporting CW or CCW if T2 was CW or CCW, respectively, see Figure 2B), or correctly reporting T2's absence (a correct rejection). Here, it seems that one should also count a misidentification (i.e., incorrect choice of CW or CCW when T2 was present) as a correct detection, because participants apparently did detect T2, but failed to judge/remember its orientation properly in case of a misidentification. Conversely, the manner in which discrimination performance is computed also raises questions. Here, the authors appear to compute accuracy as the average proportion of T2-present trials on which participants selected the correct response option for T2, thus including trials in which participants missed T2 entirely. Thus, a failure to detect T2 is now counted as a failure to discriminate T2. Wouldn't a more proper measure of discrimination accuracy be to compute the proportion of correct discriminations for trials in which participants detected T2?

My last point of critique is that the paper offers little if any guidance on how the inferred distinction between detection and discrimination can be linked to existing theories of the attentional blink. The discussion mostly focuses on comparisons to previous EEG studies, but it would be interesting to know how the authors connect their findings to extant, mechanistic accounts of the attentional blink. A key question here is whether the finding of dissociable processes of detection and discrimination would also hold with more meaningful stimuli in an identification task (e.g., the canonical AB task of identifying two letters shown amongst digits). There is evidence to suggest that meaningful stimuli are categorized just as quickly as they are detected (Grill-Spector & Kanwisher, 2005; Grill-Spector K, Kanwisher N. Visual recognition: as soon as you know it is there, you know what it is. Psychol Sci. 2005 Feb;16(2):152-60. doi: 10.1111/j.0956-7976.2005.00796.x. PMID: 15686582.). Does that mean that the observed distinction between detection and discrimination would only apply to tasks in which the targets consist of otherwise meaningless visual elements, such as lines of different orientations?

Reviewer #2 (Public Review):

Summary:

The authors had two aims: First, to decompose the attentional blink (AB) deficit into the two components of signal detection theory; sensitivity and bias. Second, the authors aimed to assess the two subcomponents of sensitivity; detection and discrimination. They observed that the AB is only expressed in sensitivity. Furthermore, detection and discrimination were doubly dissociated. Detection modulated N2p and P3 ERP amplitude, but not frontoparietal beta-band coherence, whereas this pattern was reversed for discrimination.

Strengths:

The experiment is elegantly designed, and the data - both behavioral and electrophysiological - are aptly analyzed. The outcomes, in particular the dissociation between detection and discrimination blinks, are consistently and clearly supported by the results. The discussion of the results is also appropriately balanced.

Weaknesses:

The lack of an effect of stimulus contrast does not seem very surprising from what we know of the nature of AB already. Low-level perceptual factors are not thought to cause AB. This is fine, as there are also other, novel findings reported, but perhaps the authors could bolster the importance of these (null) findings by referring to AB-specific papers, if there are indeed any, that would have predicted different outcomes in this regard.

On an analytical note, the ERP analysis could be finetuned a little more. The task design does not allow measurement of the N2pc or N400 components, which are also relevant to the AB, but the N1 component could additionally be analyzed. In doing so, I would furthermore recommend selecting more lateral electrode sites for both the N1, as well as the P1. Both P1 and N1 are likely not maximal near the midline, where the authors currently focused their P1 analysis.

Impact & Context:

The results of this study will likely influence how we think about selective attention in the context of the AB phenomenon. However, I think its impact could be further improved by extending its theoretical framing. In particular, there has been some recent work on the nature of the AB deficit, showing that it can be discrete (all-or-none) and gradual (Sy et al., 2021; Karabay et al., 2022, both in JEP: General). These different faces of target awareness in the AB may be linked directly to the detection and discrimination subcomponents that are analyzed in the present paper. I would encourage the authors to discuss this potential link and comment on the bearing of the present work on these previous behavioral findings.

Reviewer #3 (Public Review):

Summary:

In the present study, the authors aimed to achieve a better understanding of the mechanisms underlying the attentional blink, that is, a deficit in processing the second of two target stimuli when they appear in rapid succession. Specifically, they used a concurrent detection and identification task in- and outside of the attentional blink and decoupled effects of perceptual sensitivity and response bias using a novel signal detection model. They conclude that the attentional blink selectively impairs perceptual sensitivity but not response bias, and link established EEG markers of the attentional blink to deficits in stimulus detection (N2p, P3) and discrimination (fronto-parietal high-beta coherence), respectively. Taken together, their study suggests distinct mechanisms mediating detection and discrimination deficits in the attentional blink.

Strengths:

Major strengths of the present study include its innovative approach to investigating the mechanisms underlying the attentional blink, an elegant, carefully calibrated experimental paradigm, a novel signal detection model, and multifaceted data analyses using state-of-the-art model comparisons and robust statistical tests. The study appears to have been carefully conducted and the overall conclusions seem warranted given the results. In my opinion, the manuscript is a valuable contribution to the current literature on the attentional blink. Moreover, the novel paradigm and signal detection model are likely to stimulate future research.

Weaknesses:

Weaknesses of the present manuscript mainly concern the negligence of some relevant literature, unclear hypotheses, potentially data-driven analyses, relatively low statistical power, potential flaws in the EEG methods, and the absence of a discussion of limitations. In the following, I will list some major and minor concerns in detail.

Major points

Hypotheses:<br /> I appreciate the multifaceted, in-depth analysis of the given dataset including its high amount of different statistical tests. However, neither the Introduction nor the Methods contain specific statistical hypotheses. Moreover, many of the tests (e.g., correlations) rely on selected results of previous tests. It is unclear how many of the tests were planned a priori, how many more were performed, and how exactly corrections for multiple tests were implemented. Thus, I find it difficult to assess the robustness of the results.

Power:<br /> Some important null findings may result from the rather small sample sizes of N = 24 for behavioral and N = 18 for ERP analyses. For example, the correlation between detection and discrimination d' deficits across participants (r=0.39, p=0.059) (p. 12, l. 263) and the attentional blink effect on the P1 component (p=0.050, no test statistic) (p. 14, 301) could each have been significant with one more participant. In my opinion, such results should not be interpreted as evidence for the absence of effects.

Neural basis of the attentional blink:<br /> The introduction (e.g., p. 4, l. 56-76) and discussion (e.g., p. 19, 427-447) do not incorporate the insights from the highly relevant recent review by Zivony & Lamy (2022), which is only cited once (p. 19, l. 428). Moreover, the sections do not mention some relevant ERP studies of the attentional blink (e.g., Batterink et al., 2012; Craston et al., 2009; Dell'Acqua et al., 2015; Dellert et al., 2022; Eiserbeck et al., 2022; Meijs et al., 2018).

Detection versus discrimination:<br /> Concerning the neural basis of detection versus discrimination (e.g., p. 6, l. 98-110; p. 18, l. 399-412), relevant existing literature (e.g., Broadbent & Broadbent, 1987; Hillis & Brainard, 2007; Koivisto et al., 2017; Straube & Fahle, 2011; Wiens et al., 2023) is not included.

Pooling of lags and lag 1 sparing:<br /> I wonder why the authors chose to include 5 different lags when they later pooled early (100, 300 ms) and late (700, 900 ms) lags, and whether this pooling is justified. This is important because T2 at lag 1 (100 ms) is typically "spared" (high accuracy) while T2 at lag 3 (300 ms) shows the maximum AB (for reviews, see, e.g., Dux & Marois, 2009; Martens & Wyble, 2010). Interestingly, this sparing was not observed here (p. 43, Figure 2). Nevertheless, considering the literature and the research questions at hand, it is questionable whether lag 1 and 3 should be pooled.

Discrimination in the attentional blink<br /> Concerning the claims that previous attentional blink studies conflated detection and discrimination (p. 6, l. 111-114; p. 18, l. 416), there is a recent ERP study (Dellert et al., 2022) in which participants did not perform a discrimination task for the T2 stimuli. Moreover, since the relevance of all stimuli except T1 was uncertain in this study, irrelevant distractors could not be filtered out (cf. p. 19, l. 437). Under these conditions, the attentional blink was still associated with reduced negativities in the N2 range (cf. p. 19, l. 427-437) but not with a reduced P3 (cf. p. 19, l 439-447).

General EEG methods:<br /> While most of the description of the EEG preprocessing and analysis (p. 31/32) is appropriate, it also lacks some important information (see, e.g., Keil et al., 2014). For example, it does not include the length of the segments, the type and proportion of artifacts rejected, the number of trials used for averaging in each condition, specific hypotheses, and the test statistics (in addition to p-values).

EEG filters:<br /> P. 31, l. 728: "The data were (...) bandpass filtered between 0.5 to 18 Hz (...). Next, a bandstop filter from 9-11 Hz was applied to remove the 10 Hz oscillations evoked by the RSVP presentation." These filter settings do not follow common recommendations and could potentially induce filter distortions (e.g., Luck, 2014; Zhang et al., 2024). For example, the 0.5 high-pass filter could distort the slow P3 wave. Mostly, I am concerned about the bandstop filter. Since the authors commendably corrected for RSVP-evoked responses by subtracting T2-absent from T2-present ERPs (p. 31, l. 746), I wonder why the additional filter was necessary, and whether it might have removed relevant peaks in the ERPs of interest.

Coherence analysis:<br /> P. 33, l. 786: "For subsequent, partial correlation analyses of coherence with behavioral metrics and neural distances (...), we focused on a 300 ms time period (0-300 ms following T2 onset) and high-beta frequency band (20-30 Hz) identified by the cluster-based permutation test (Fig. 5A-C)." I wonder whether there were any a priori criteria for the definition and selection of such successive analyses. Given the many factors (frequency bands, hemispheres) in the analyses and the particular shape of the cluster (p. 49, Fig 5C), this focus seems largely data-driven. It remains unclear how many such tests were performed and whether the results (e.g., the resulting weak correlation of r = 0.22 in one frequency band and one hemisphere in one part of a complexly shaped cluster; p. 15, l. 327) can be considered robust.

References<br /> Batterink, L., Karns, C. M., & Neville, H. (2012). Dissociable mechanisms supporting awareness: The P300 and gamma in a linguistic attentional blink task. Cerebral Cortex, 22(12), 2733-2744. https://doi.org/10.1093/cercor/bhr346<br /> Broadbent, D. E., & Broadbent, M. H. P. (1987). From detection to identification: Response to multiple targets in rapid serial visual presentation. Perception & Psychophysics, 42(2), 105-113. https://doi.org/10.3758/BF03210498<br /> Craston, P., Wyble, B., Chennu, S., & Bowman, H. (2009). The attentional blink reveals serial working memory encoding: Evidence from virtual and human event-related potentials. Journal of Cognitive Neuroscience, 21(3), 550-566. https://doi.org/10.1162/jocn.2009.21036<br /> Dell'Acqua, R., Dux, P. E., Wyble, B., Doro, M., Sessa, P., Meconi, F., & Jolicœur, P. (2015). The attentional blink impairs detection and delays encoding of visual information: Evidence from human electrophysiology. Journal of Cognitive Neuroscience, 27(4), 720-735. https://doi.org/10.1162/jocn_a_00752<br /> Dellert, T., Krebs, S., Bruchmann, M., Schindler, S., Peters, A., & Straube, T. (2022). Neural correlates of consciousness in an attentional blink paradigm with uncertain target relevance. NeuroImage, 264C, 119679. https://doi.org/10.1016/j.neuroimage.2022.119679<br /> Dux, P. E., & Marois, R. (2009). The attentional blink: A review of data and theory. Attention, Perception, & Psychophysics, 71(8), 1683-1700. https://doi.org/10.3758/APP.71.8.1683<br /> Hillis, J. M., & Brainard, D. H. (2007). Distinct mechanisms mediate visual detection and identification. Current Biology, 17(19), 1714-1719. https://doi.org/10.1016/j.cub.2007.09.012<br /> Keil, A., Debener, S., Gratton, G., Junghöfer, M., Kappenman, E. S., Luck, S. J., Luu, P., Miller, G. A., & Yee, C. M. (2014). Committee report: Publication guidelines and recommendations for studies using electroencephalography and magnetoencephalography. Psychophysiology, 51(1), 1-21. https://doi.org/10.1111/psyp.12147<br /> Koivisto, M., Grassini, S., Salminen-Vaparanta, N., & Revonsuo, A. (2017). Different electrophysiological correlates of visual awareness for detection and identification. Journal of Cognitive Neuroscience, 29(9), 1621-1631. https://doi.org/10.1162/jocn_a_01149<br /> Luck, S. J. (2014). An introduction to the event-related potential technique. MIT Press.<br /> Martens, S., & Wyble, B. (2010). The attentional blink: Past, present, and future of a blind spot in perceptual awareness. Neuroscience & Biobehavioral Reviews, 34(6), 947-957. https://doi.org/10.1016/j.neubiorev.2009.12.005<br /> Meijs, E. L., Slagter, H. A., de Lange, F. P., & Gaal, S. van. (2018). Dynamic interactions between top-down expectations and conscious awareness. Journal of Neuroscience, 38(9), 2318-2327. https://doi.org/10.1523/JNEUROSCI.1952-17.2017<br /> Straube, S., & Fahle, M. (2011). Visual detection and identification are not the same: Evidence from psychophysics and fMRI. Brain and Cognition, 75(1), 29-38. https://doi.org/10.1016/j.bandc.2010.10.004<br /> Wiens, S., Andersson, A., & Gravenfors, J. (2023). Neural electrophysiological correlates of detection and identification awareness. Cognitive, Affective, & Behavioral Neuroscience. https://doi.org/10.3758/s13415-023-01120-5<br /> Zhang, G., Garrett, D. R., & Luck, S. J. (2024). Optimal filters for ERP research II: Recommended settings for seven common ERP components. Psychophysiology, n/a(n/a), e14530. https://doi.org/10.1111/psyp.14530

Author response:

Reviewer #1: 

Summary:

In this study, the authors used a multi-alternative decision task and a multidimensional signal-detection model to gain further insight into the cause of perceptual impairments during the attentional blink. The model-based analyses of behavioural and EEG data show that such perceptual failures can be unpacked into distinct deficits in visual detection and discrimination, with visual detection being linked to the amplitude of late ERP components (N2P and P3) and discrimination being linked to the coherence of fronto-parietal brain activity.

Strengths:

The main strength of this paper lies in the fact that it presents a novel perspective on the cause of perceptual failures during the attentional blink. The multidimensional signaldetection modelling approach is explained clearly, and the results of the study show that this approach offers a powerful method to unpack behavioural and EEG data into distinct processes of detection and discrimination.

Weaknesses:

(1.1) While the model-based analyses are compelling, the paper also features some analyses that seem misguided, or, at least, insufficiently motivated and explained. Specifically, in the introduction, the authors raise the suggestion that the attentional blink could be due to a reduction in sensitivity or a response bias. The suggestion that a response bias could play a role seems misguided, as any response bias would be expected to be constant across lags, while the attentional blink effect is only observed at short lags. Thus, it is difficult to understand why the authors would think that a response bias could explain the attentional blink.

A deficit in T2 identification accuracy could arise from either sensitivity or criterion effects; the criterion effect may manifest as a choice bias. For example, in short T1-T2 lag trials, when T2 closely follows T1, participants may adopt a more conservative choice criterion for reporting the presence of T2. Moreover, criterion effects need not be uniform across lags: A participant could infer the T1-T2 lag interval based on various factors, including trial length, thereby permitting them to adjust their choice criterion variably across different lags. We will provide a more detailed illustration of this claim in the revision.

(1.2) A second point of concern regards the way in which the measures for detection and discrimination accuracy were computed. If I understand the paper correctly, a correct detection was defined as either correctly identifying T2 (i.e., reporting CW or CCW if T2 was CW or CCW, respectively, see Figure 2B), or correctly reporting T2's absence (a correct rejection). Here, it seems that one should also count a misidentification (i.e., incorrect choice of CW or CCW when T2 was present) as a correct detection, because participants apparently did detect T2, but failed to judge/remember its orientation properly in case of a misidentification. Conversely, the manner in which discrimination performance is computed also raises questions. Here, the authors appear to compute accuracy as the average proportion of T2-present trials on which participants selected the correct response option for T2, thus including trials in which participants missed T2 entirely. Thus, a failure to detect T2 is now counted as a failure to discriminate T2. Wouldn't a more proper measure of discrimination accuracy be to compute the proportion of correct discriminations for trials in which participants detected T2?

Detection and discrimination accuracies were computed with precisely the same procedure, and under the same conditions, as described by the Reviewer (underlined text, above). We regret our poor description; we will improve upon it in the revised manuscript.

(1.3) My last point of critique is that the paper offers little if any guidance on how the inferred distinction between detection and discrimination can be linked to existing theories of the attentional blink. The discussion mostly focuses on comparisons to previous EEG studies, but it would be interesting to know how the authors connect their findings to extant, mechanistic accounts of the attentional blink. A key question here is whether the finding of dissociable processes of detection and discrimination would also hold with more meaningful stimuli in an identification task (e.g., the canonical AB task of identifying two letters shown amongst digits). There is evidence to suggest that meaningful stimuli are categorized just as quickly as they are detected (Grill-Spector & Kanwisher, 2005; Grill-Spector K, Kanwisher N. Visual recognition: as soon as you know it is there, you know what it is. Psychol Sci. 2005 Feb;16(2):152-60. doi: 10.1111/j.0956-7976.2005.00796.x. PMID: 15686582.). Does that mean that the observed distinction between detection and discrimination would only apply to tasks in which the targets consist of otherwise meaningless visual elements, such as lines of different orientations?

Our results are consistent with previous literature suggested by the Reviewer. Specifically, we do not claim that detection and discrimination are sequential processes; in fact, we modeled them as concurrent computations (Figs. 3A-B). Yet, our results suggest that these processes possess distinct neural bases. We have discussed this idea briefly in the Discussion section (e.g., “Yet, we found no evidence for these two computations being sequential…”). We will discuss this further in the revised manuscript in the context of previous literature.

Reviewer #2:

Summary:

The authors had two aims: First, to decompose the attentional blink (AB) deficit into the two components of signal detection theory; sensitivity and bias. Second, the authors aimed to assess the two subcomponents of sensitivity; detection and discrimination. They observed that the AB is only expressed in sensitivity. Furthermore, detection and discrimination were doubly dissociated. Detection modulated N2p and P3 ERP amplitude, but not frontoparietal beta-band coherence, whereas this pattern was reversed for discrimination.

Strengths:

The experiment is elegantly designed, and the data - both behavioral and electrophysiological - are aptly analyzed. The outcomes, in particular the dissociation between detection and discrimination blinks, are consistently and clearly supported by the results. The discussion of the results is also appropriately balanced.

Weaknesses:

(2.1) The lack of an effect of stimulus contrast does not seem very surprising from what we know of the nature of AB already. Low-level perceptual factors are not thought to cause AB. This is fine, as there are also other, novel findings reported, but perhaps the authors could bolster the importance of these (null) findings by referring to AB-specific papers, if there are indeed any, that would have predicted different outcomes in this regard.

While there is consensus that the low-level perceptual factors are not affected by the attentional blink, other studies may suggest evidence to the contrary (e.g., Chua et al, Percept. Psychophys., 2005). We will highlight the significance of our findings in the context of such conflicting evidence in literature, in the revised manuscript.

(2.2) On an analytical note, the ERP analysis could be finetuned a little more. The task design does not allow measurement of the N2pc or N400 components, which are also relevant to the AB, but the N1 component could additionally be analyzed. In doing so, I would furthermore recommend selecting more lateral electrode sites for both the N1, as well as the P1. Both P1 and N1 are likely not maximal near the midline, where the authors currently focused their P1 analysis.

We will incorporate these additional analyses in the revised manuscript.

(2.3) Impact & Context:

The results of this study will likely influence how we think about selective attention in the context of the AB phenomenon. However, I think its impact could be further improved by extending its theoretical framing. In particular, there has been some recent work on the nature of the AB deficit, showing that it can be discrete (all-or-none) and gradual (Sy et al., 2021; Karabay et al., 2022, both in JEP: General). These different faces of target awareness in the AB may be linked directly to the detection and discrimination subcomponents that are analyzed in the present paper. I would encourage the authors to discuss this potential link and comment on the bearing of the present work on these behavioural findings.

Thank you. We will discuss our findings in the context of these recent studies.

Reviewer #3:

Summary:

In the present study, the authors aimed to achieve a better understanding of the mechanisms underlying the attentional blink, that is, a deficit in processing the second of two target stimuli when they appear in rapid succession. Specifically, they used a concurrent detection and identification task in- and outside of the attentional blink and decoupled effects of perceptual sensitivity and response bias using a novel signal detection model. They conclude that the attentional blink selectively impairs perceptual sensitivity but not response bias, and link established EEG markers of the attentional blink to deficits in stimulus detection (N2p, P3) and discrimination (fronto-parietal high-beta coherence), respectively. Taken together, their study suggests distinct mechanisms mediating detection and discrimination deficits in the attentional blink.

Strengths:

Major strengths of the present study include its innovative approach to investigating the mechanisms underlying the attentional blink, an elegant, carefully calibrated experimental paradigm, a novel signal detection model, and multifaceted data analyses using state-of-theart model comparisons and robust statistical tests. The study appears to have been carefully conducted and the overall conclusions seem warranted given the results. In my opinion, the manuscript is a valuable contribution to the current literature on the attentional blink. Moreover, the novel paradigm and signal detection model are likely to stimulate future research.

Weaknesses:

Weaknesses of the present manuscript mainly concern the negligence of some relevant literature, unclear hypotheses, potentially data-driven analyses, relatively low statistical power, potential flaws in the EEG methods, and the absence of a discussion of limitations. In the following, I will list some major and minor concerns in detail.

Major points

(3.1) Hypotheses:

I appreciate the multifaceted, in-depth analysis of the given dataset including its high amount of different statistical tests. However, neither the Introduction nor the Methods contain specific statistical hypotheses. Moreover, many of the tests (e.g., correlations) rely on selected results of previous tests. It is unclear how many of the tests were planned a priori, how many more were performed, and how exactly corrections for multiple tests were implemented. Thus, I find it difficult to assess the robustness of the results.

As outlined in the Introduction, we hypothesized that neural computations associated with target detection would be characterized by regional neuronal markers (e.g., parietal or occipital ERPs), whereas computations linked to feature discrimination may involve neural coordination across multiple brain regions (e.g. fronto-parietal coherence). We planned and conducted our statistical tests based on this hypothesis. All multiple comparison corrections (e.g., Bonferroni-Holm correction, see Methods) were performed separately for each class of analyses. We will clarify these hypotheses and provide further details in the revised manuscript.

(3.2) Power:

Some important null findings may result from the rather small sample sizes of N = 24 for behavioral and N = 18 for ERP analyses. For example, the correlation between detection and discrimination d' deficits across participants (r=0.39, p=0.059) (p. 12, l. 263) and the attentional blink effect on the P1 component (p=0.050, no test statistic) (p. 14, 301) could each have been significant with one more participant. In my opinion, such results should not be interpreted as evidence for the absence of effects.

We agree and will revise the manuscript accordingly. We will also report Bayes factor (BF) values, where relevant, to further evaluate these claims.

(3.3) Neural basis of the attentional blink:

The introduction (e.g., p. 4, l. 56-76) and discussion (e.g., p. 19, 427-447) do not incorporate the insights from the highly relevant recent review by Zivony & Lamy (2022), which is only cited once (p. 19, l. 428). Moreover, the sections do not mention some relevant ERP studies of the attentional blink (e.g., Batterink et al., 2012; Craston et al., 2009; Dell'Acqua et al., 2015; Dellert et al., 2022; Eiserbeck et al., 2022; Meijs et al., 2018).

We will motivate and discuss our study in the context of these previous studies. 

(3.4) Detection versus discrimination:

Concerning the neural basis of detection versus discrimination (e.g., p. 6, l. 98-110; p. 18, l. 399-412), relevant existing literature (e.g., Broadbent & Broadbent, 1987; Hillis & Brainard, 2007; Koivisto et al., 2017; Straube & Fahle, 2011; Wiens et al., 2023) is not included.

Thank you for these suggestions. We will include these important studies in our discussion.

(3.5) Pooling of lags and lags 1 sparing:

I wonder why the authors chose to include 5 different lags when they later pooled early (100, 300 ms) and late (700, 900 ms) lags, and whether this pooling is justified. This is important because T2 at lag 1 (100 ms) is typically "spared" (high accuracy) while T2 at lag 3 (300 ms) shows the maximum AB (for reviews, see, e.g., Dux & Marois, 2009; Martens & Wyble, 2010). Interestingly, this sparing was not observed here (p. 43, Figure 2). Nevertheless, considering the literature and the research questions at hand, it is questionable whether lag 1 and 3 should be pooled.

Lag-1 sparing is not always observed in attentional blink studies; there are notable exceptions that do not report such sparing (Hommel et al., Q. J. Exp. Psychol., 2005; Livesay et al., Attention, Percept. Psychophys., 2011). Our statistical tests revealed no significant difference in accuracies between short lag (100 and 300 ms) trials or between long lag (700 and 900 ms) trials but did reveal significant differences between the short and long lag trials (ANOVA, followed by post-hoc tests). To simplify the presentation of the findings, we pooled together the short lag (100 and 300 ms) and, separately, the long lag (700 and 900 ms) trials. We will present these analyses, and clarify the motivation for pooling in the revised manuscript. 

(3.6) Discrimination in the attentional blink

Concerning the claims that previous attentional blink studies conflated detection and discrimination (p. 6, l. 111-114; p. 18, l. 416), there is a recent ERP study (Dellert et al., 2022) in which participants did not perform a discrimination task for the T2 stimuli. Moreover, since the relevance of all stimuli except T1 was uncertain in this study, irrelevant distractors could not be filtered out (cf. p. 19, l. 437). Under these conditions, the attentional blink was still associated with reduced negativities in the N2 range (cf. p. 19, l. 427-437) but not with a reduced P3 (cf. p. 19, l 439-447).

We will address the difference between our findings and those of Dellert et al (2022) in the revised manuscript.

(3.7) General EEG methods:

While most of the description of the EEG preprocessing and analysis (p. 31/32) is appropriate, it also lacks some important information (see, e.g., Keil et al., 2014). For example, it does not include the length of the segments, the type and proportion of artifacts rejected, the number of trials used for averaging in each condition, specific hypotheses, and the test statistics (in addition to p-values).

We regret the oversight. We will include these details in the revised Methods.

(3.8) EEG filters:

P. 31, l. 728: "The data were (...) bandpass filtered between 0.5 to 18 Hz (...). Next, a bandstop filter from 9-11 Hz was applied to remove the 10 Hz oscillations evoked by the RSVP presentation." These filter settings do not follow common recommendations and could potentially induce filter distortions (e.g., Luck, 2014; Zhang et al., 2024). For example, the 0.5 high-pass filter could distort the slow P3 wave. Mostly, I am concerned about the bandstop filter. Since the authors commendably corrected for RSVP-evoked responses by subtracting T2-absent from T2-present ERPs (p. 31, l. 746), I wonder why the additional filter was necessary, and whether it might have removed relevant peaks in the ERPs of interest.

Thank you for this suggestion. We will repeat this analysis by removing these additional filters.

(3.9) Coherence analysis:

P. 33, l. 786: "For subsequent, partial correlation analyses of coherence with behavioral metrics and neural distances (...), we focused on a 300 ms time period (0-300 ms following T2 onset) and high-beta frequency band (20-30 Hz) identified by the cluster-based permutation test (Fig. 5A-C)." I wonder whether there were any a priori criteria for the definition and selection of such successive analyses. Given the many factors (frequency bands, hemispheres) in the analyses and the particular shape of the cluster (p. 49, Fig 5C), this focus seems largely data-driven. It remains unclear how many such tests were performed and whether the results (e.g., the resulting weak correlation of r = 0.22 in one frequency band and one hemisphere in one part of a complexly shaped cluster; p. 15, l. 327) can be considered robust.

Please see responses to comments #3.1 and #3.2 (above). In addition to reporting further details regarding statistical tests and multiple comparisons corrections, we will compute and report Bayes factors to quantify the strength of the evidence for correlations, as appropriate.

eLife assessment

This study employs a modified protocol for single-nuclei RNA sequencing of adipose tissue that preserves RNA quality and nuclei integrity. Using this protocol, the study provides valuable insights into the cellular heterogeneity and molecular landscape of murine adipose tissue from lean mice and mice with diet-induced obesity. The study is solid in its approach and analysis, providing a comprehensive description of a dysfunctional hypertrophic adipocyte subpopulation that emerges in association with obesity.

Reviewer #1 (Public Review):

Summary:

This manuscript from So et al. describes what is suggested to be an improved protocol for single-nuclei RNA sequencing (snRNA-seq) of adipose tissue. The authors provide evidence that modifications to the existing protocols result in better RNA quality and nuclei integrity than previously observed, with ultimately greater coverage of the transcriptome upon sequencing. Using the modified protocol, the authors compare the cellular landscape of murine inguinal and perigonadal white adipose tissue (WAT) depots harvested from animals fed a standard chow diet (lean mice) or those fed a high-fat diet (mice with obesity).

Strengths:

Overall, the manuscript is well-written, and the data are clearly presented. The strengths of the manuscript rest in the description of an improved protocol for snRNA-seq analysis. This should be valuable for the growing number of investigators in the field of adipose tissue biology that are utilizing snRNA-seq technology, as well as those other fields attempting similar experiments with tissues possessing high levels of RNAse activity.

Moreover, the study makes some notable observations that provide the foundation for future investigation. One observation is the correlation between nuclei size and cell size, allowing for the transcriptomes of relatively hypertrophic adipocytes in perigonadal WAT to be examined. Another notable observation is the identification of an adipocyte subcluster (Ad6) that appears "stressed" or dysfunctional and likely localizes to crown-like inflammatory structures where pro-inflammatory immune cells reside.

Weaknesses:

Analogous studies have been reported in the literature, including a notable study from Savari et al. (Cell Metabolism). This somewhat diminishes the novelty of some of the biological findings presented here. Moreover, a direct comparison of the transcriptomic data derived from the new vs. existing protocols (i.e. fully executed side by side) was not presented. As such, the true benefit of the protocol modifications cannot be fully understood.

Reviewer #2 (Public Review):

Summary:

In the present manuscript So et al utilize single-nucleus RNA sequencing to characterize cell populations in lean and obese adipose tissues.

Strengths:

The authors utilize a modified nuclear isolation protocol incorporating VRC that results in higher-quality sequencing reads compared with previous studies.

Weaknesses:

The use of VRC to enhance snRNA-seq has been previously published in other tissues. The snRNA-seq snRNA-seq data sets presented in this manuscript, when compared with numerous previously published single-cell analyses of adipose tissue, do not represent a significant scientific advance.

Figure 1-3: The snRNA-seq data obtained by the authors using their enhanced protocol does not represent a significant improvement in cell profiling for the majority of the highlighted cell types including APCs, macrophages, and lymphocytes. These cell populations have been extensively characterized by cytoplasmic scRNA-seq which can achieve sufficient sequencing depth, and thus this study does not contribute meaningful additional insight into these cell types. The authors note an increase in the number of rare endothelial cell types recovered, however this is not translated into any kind of functional analysis of these populations.

Figure 4: The authors did not provide any evidence that the relative fluorescent brightness of GFP and mCherry is a direct measure of the nuclear size, and the nuclear size is only a moderate correlation with the cell size. Thus sorting the nuclei based on GFP/mCherry brightness is not a great proxy for adipocyte diameter. Furthermore, no meaningful insights are provided about the functional significance of the reported transcriptional differences between small and large adipocyte nuclei.

Figure 5-6: The Ad6 population is highly transcriptionally analogous to the mAd3 population from Emont et al, and is thus not a novel finding. Furthermore, in the present data set, the authors conclude that Ad6 are likely stressed/dying hypertrophic adipocytes with a global loss of gene expression, which is a well-documented finding in eWAT > iWAT, for which the snRNA-seq reported in the present manuscript does not provide any novel scientific insight.

Reviewer #3 (Public Review):

Summary:

The authors aimed to improve single-nucleus RNA sequencing (snRNA-seq) to address current limitations and challenges with nuclei and RNA isolation quality. They successfully developed a protocol that enhances RNA preservation and yields high-quality snRNA-seq data from multiple tissues, including a challenging model of adipose tissue. They then applied this method to eWAT and iWAT from mice fed either a normal or high-fat diet, exploring depot-specific cellular dynamics and gene expression changes during obesity. Their analysis included subclustering of SVF cells and revealed that obesity promotes a transition in APCs from an early to a committed state and induces a pro-inflammatory phenotype in immune cells, particularly in eWAT. In addition to SVF cells, they discovered six adipocyte subpopulations characterized by a gradient of unique gene expression signatures. Interestingly, a novel subpopulation, termed Ad6, comprised stressed and dying adipocytes with reduced transcriptional activity, primarily found in eWAT of mice on a high-fat diet. Overall, the methodology is sound, the writing is clear, and the conclusions drawn are supported by the data presented. Further research based on these findings could pave the way for potential novel interventions in obesity and metabolic disorders, or for similar studies in other tissues or conditions.

Strengths:

• The authors developed a robust snRNA-seq technique that preserves the integrity of the nucleus and RNA across various tissue types, overcoming the challenges of existing methods.

• They identified adipocyte subpopulations that follow adaptive or pathological trajectories during obesity.

• The study reveals depot-specific differences in adipose tissues, which could have implications for targeted therapies.

Weaknesses:

• The adipose tissues were collected after 10 weeks of high-fat diet treatment, lacking the intermediate time points for identifying early markers or cell populations during the transition from healthy to pathological adipose tissue.

• The expansion of the Ad6 subpopulation in obese iWAT and gWAT is interesting. The author claims that Ad6 exhibited a substantial increase in eWAT and a moderate rise in iWAT (Figure 4C). However, this adipocyte subpopulation remains the most altered in iWAT upon obesity. Could the authors elaborate on why there is a scarcity of adipocytes with ROS reporter and B2M in obese iWAT?

• While the study provides extensive data on mouse models, the potential translation of these findings to human obesity remains uncertain.