Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Transparent Peer Review

---

Download the complete Review Process [PDF] including:

• reviews
• authors' reply
• editorial decisions

</br>

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Below is our point-by-point response:

---

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

The manuscript of Lalanne and coworkers address the cellular responses to varied translation termination factor expression in Bacillus subtilis. The authors set-up a system to fine-tune the expression of release factor RF1, RF2 as well as PrmC that post-translationally modifies RF1/RF2 to maximize their catalytic hydrolysis activity. They then monitor the fitness costs associated with overexpression or depletion of the factor by following the changes in growth rate. The set-up is nicely illustrated in Figure 1. The results in Figure 2 show that overexpression of RF1 and RF2 has relatively modest effect on the growth rate compared to overexpression of PrmC that leads to dramatic growth rate reduction. By contrast, depletion of RF1 has a strong negative influence on fitness, whereas a similar level of depletion of RF2 had little influence on fitness. PrmC overexpression appears to be correlated with the induction of the sigmaB regulon, however, the authors do not manage to ascertain why this is. By contrast, RF2 depletion also results in the induction of the sigmaB regulon and the authors demonstrate convincingly that this is due to a termination defect within the rsbQ-rsbV operon that contains an overlapping start-stop AUGA

A few points that the authors might consider discussing

1. The natural abundance of each RF in bacteria in relation to the usage of different stop codons in different organisms.

Response: We thank the reviewer for their suggestion. A correlation between RF abundance and stop codon usage across bacterial species has been previously reported (Korkmaz et al., 2014; Wei et al., 2016), which is corroborated by our quantification (see below). This correlation provides further evidence that the RF expression may be optimized to meet their demands in translation termination. We now include a new discussion in the main text (p. 9, lines 410-415): "Our data thus corroborate several previous lines of evidence suggesting that RF expression might be precisely tuned. First, it was found that the relative expression between RF1 and RF2 correlates with stop codon usage between different species (Korkmaz et al., 2014; Wei et al., 2016). For instance, B. subtilis has a higher abundance of RF1 and more frequent UAG usage compared to E. coli, suggesting that RF1’s expression setpoint meets translational demand (Methods).”

Below we include additional analyses that may be of interests to the Reviewer.

From our ribosome profiling quantification in E. coli, B. subtilis, C. crescentus, and V. natriegens (Lalanne et al., 2018), we can compare the relative usage of the three stop codons (frequency of stop codons weighted by expression) with abundances of RF1 and RF2:

Despite the limited sample size, we find reasonable agreement with the expected correlation between codon usage and cognate RF abundance. In species with substantial differences between RF1 and RF2 abundances (E. coli and B. subtilis), the most heavily used non-UAA stop corresponds to the most highly expressed RF. This argues in favor of expression tuning of these important enzymes and is consistent with the growth optimization we directly observe.

As a word of caution, although the low usage of UAG in E. coli and low expression of RF1 (reported long ago, e.g., (Adamski et al., 1994)) is well established, it should be noted that strain MG1655’s RF2 factor harbors a debilitating missense A246T mutation near its active site (Dinçbas-Renqvist et al., 2000), which potentially complicates interpretation of the expression of E. coli’s release factors [interestingly, we do not see any difference in RF1 and RF2 expression from ribosome profiling data in strain NCM3722, which contains the RF2 variant without the A246T mutation (JBL, unpublished data)].

The role of the frameshifting mechanism in RF2 and how then RF1 levels are regulated.

Response: We thank the reviewer to raising the interesting topic of release factor expression regulation. We have added a section in our discussion to comment on RF2 regulation (p. 9, lines 415-420).

“Second, the gene encoding RF2 has a broadly conserved UGA-based frameshift event that autoregulates the expression based on its own activity (Baranov et al., 2002; Craigen and Caskey, 1986; Craigen et al., 1985). Interestingly, there are no reports of RF1 autoregulation to our knowledge, and we found that ectopic over- or under-expression does not affect its own promoter activity (Fig. S7). Therefore, a lack of autoregulation does not necessarily indicate that cells are less sensitive to small perturbations on its expression.”

The statement above includes an additional analysis on RF1 regulation that was motivated by the Reviewer’s comment. In contrast to RF2, no definitive evidence exists on autoregulatory mechanisms for RF1. Following the Reviewer’s comment, we realized that our dataset allowed us to search for evidence of endogenous regulation in B. subtilis: our RF1 expression strain has a markerless deletion of prfA and prmC genes, leaving the surrounding regions, and notably the promoter, intact. As such, possible unbeknownst regulatory mechanisms at the promoter level could be identified in our RNA-seq data under steady-state perturbation of RF1 levels. Quantifying the expression of the 5’ untranslated region and operonic gene ywkF at the ablated prfAlocus (presented in Fig. S7, reproduced below), we find no significant changes in expression across over 30-fold range in RF1 expression, arguing against such transcriptional regulatory mechanisms. Although this does not rule out other regulatory mechanisms at the post-transcriptional level, no such mechanisms have been documented for RF1 to our knowledge.

The authors observe queuing in front of the relevant stop codons upon RF depletion, however, do not discuss about readthrough events, which are usually competing with termination. Surprisingly, in this context the authors don't discuss the work from Mankin and coworkers showing sequestration of RFs from termination by peptides such as apideacin leads to translational readthrough.

Response: We concur with the Reviewer about the importance of the recent work from Mankin et al. This paper was referenced in our original submission, but our literature management software improperly formatted its citation. The corrected reference to (Mangano et al., 2020) is now included in the revised manuscript.

Translational readthrough is indeed clearly visible in our ribosome profiling data from acute CRISPRi knockdown of RF1/PrmC and RF2. Using an approach analogous to Mangano and Florin et al, we quantified readthrough as the ribosome footprint density downstream of the stop codon (+5 to +45 bp) to the density in the gene body for isolated genes (no codirectional genes within 55 bp). We find five-fold increase in the median readthrough for genes that are terminated by the RF under perturbation (shown in a new panel in the main text, Fig. 4b, reproduced below). This new analysis is included in the section regarding translational phenotypes identified from ribosome profiling under RF depletion, p. 7, lines 309-312.

“The stop-codon-specific queuing is associated with translational readthrough downstream (Fig. 4b), consistent with a recent observation based on inhibition of peptide release by the antimicrobial apidaecin in E. coli (Mangano et al., 2020).”

This additional analysis, in conjunction with (Mangano et al., 2020), also allows us to calibrate the depletion of RFs in our non steady-state CRISPRi perturbation. Given that apidaecin treatment (shown to lead to a nearly complete depletion of free RF in the cell) causes a >100-fold increase in readthrough, this suggests that our CRISPRi perturbation experiments only led to partial RF depletion at the moment of cell harvesting.

The efficiency of translation termination is well-known to be dependent on the context of the stop codon. Do the authors also observe such a trend. Especially, UGAC for RF2, one would expect to observe high levels of readthrough upon RF2 depletion.

Response: Further assessment of the sequence determinants that dictate susceptibility of certain genes and regulatory elements to RF perturbation is of great interest. We now include additional analyses for the effect of stop codon context on readthrough.

In our RF2 CRISPRi knockdown data, stratifying the translational readthrough (data from Fig. 4b) by stop codon and its next nucleotide, we observe only a modest (≈2×, p“We also observed a trend of tetranucleotide-dependent (UGAN) readthrough for RF2 knockdowns (Methods, Appendix Fig. 2) consistent with previous characterizations (Poole et al., 1995).”

As an additional point of interest, the importance of the 4th nucleotide in termination has not been studied outside of E. coli. Although indirect, one way to assess the influence of the 4th nucleotide is to determine the aggregated usage of each tetranucleotide stop signal by ribosome profiling. Interestingly, and as pointed out by the Reviewer, whereas E. coli (MG1655) displays a 16× increase in usage between the maximum UGAU (tetranucleotide usage 0.064) and minimum UGAC (tetranucleotide usage 0.004), no such difference is observed in B. subtilis (usage for UGAU and UGAC both at 0.015), suggesting that the immediate sequence context surrounding stop codons could have different consequences in different species.

Reviewer #2 (Significance (Required)):

Overall, the experiments are clearly performed and beautifully illustrated. Clearly, a lot of work has gone into this study but the end message that the cell regulates carefully RF concentrations is not surprising. Especially given that RF2 carefully regulates its own levels using an autoregulatory frameshifting mechanism. The major finding that the rsbQ-rsbV operon with the RF2 dependence leading to induction of the sigmaB regulon is in the end rather trivial since these regulators depend on RF2 for termination. Therefore, this manuscript is unlikely to have general interest to people in the translation field (such as myself) but rather those working in the field of synthetic biology.

Response: We thank the Reviewer for their positive assessment of our presentation and experimental methods, and for their judgment that our work will be of interest to synthetic biologists.

In our study, we used translation as a well-characterized system to interrogate the cellular response when enzyme concentrations are perturbed. Because the system is so well characterized, it allowed to ask whether the fitness effects are due to perturbations to the translation flux itself, or rather driven by spurious distal connections in the regulatory network. The end message we wish to convey is that enzyme expression is entrenched by spurious regulatory connections, suggesting that predictive bottom-up models of expression-fitness landscapes will require near-exhaustive characterization of parts.

Although our focus is on the cellular response, there are several interesting findings related to translation. First, we show that even though RF1 and PrmC are not subject to the strict autoregulation as RF2 is, cell growth is similarly or even more sensitive to RF1 and PrmC abundance. Second, among the numerous regulators that depend on RF2 for termination, RbsV/RbsW is exceptionally sensitive to RF2 depletion (Fig. 4e). This result not only points to our incomplete understanding of translation regarding what makes this pair particularly susceptible, and further underscores the spurious nature of the cellular response to perturbations. We have expanded the discussions on the implication of these findings in the revised manuscript.

---

Reviewer #3 (Evidence, reproducibility and clarity (Required)):

In this paper, the authors use a combination of RNA sequencing, ribosome profiling and measurements of cellular composition and growth rate to gain insight into the multi-scale affects that perturbations to translation termination factors have on general physiological states and reproductive fitness using Bacillus subtilis as their model organism. Specifically, they find that perturbing the expression levels of peptide chain release factors in any direction has a negative effect on growth-rate. This negative effect was not due to a direct impact of the gene on the cell, but instead due to a chain of regulatory interactions that cause the activation of the general stress regulon. This leads to upregulation of a large chunk of the genome and an indirect impact on the expression of all other genes. Critically, the knock-on effects observed for the specific perturbations studied suggest that it may be difficult to predict expression-fitness landscapes of a cell, without carrying out a detailed mapping of all genes and the cell's physiological state.

Overall, the core findings in the paper are well justified by the data presented and the experiments appear to have been rigorously carried out.

Response: We thank the reviewer for their positive assessment.

My only concern is that it is unclear if biological replicates of the ribosome profiling were performed. Also, biological replicates are mentioned for the RNA-seq data, but no data is shown. Even a simple graph demonstrating the expression levels across these would be useful to be assured of no issues in reproducibility given the complex processing of the data involved.

Response: We now include additional analyses for biological replicates of RNA-seq and ribosome profiling experiments, which show the same high degree of reproducibility as we have demonstrated in previous studies (Johnson et al., 2020; Lalanne et al., 2018; Li et al., 2014).

With respect to RNA-seq quantification, we compared our 6 wild-type datasets (biological replicates except for different inert inducer concentrations, using the same batch of conditioned MCC medium) against each other in all possible pairs. The data is now included as Appendix Fig. 1a (referred to in the main text, p. 4, line 138), and is reproduced below. Across pairs, the mRNA level quantification shows a median FC1090 (10th and 90th percentile in fold-change) between 0.86 to 1.16, and median R2 of log-transformed data at 0.99. These statistics showcasing reproducibility of our RNA-seq methodology are now included in our description of our RNA-seq approach in the Methods, p. S8, lines 313-320.

Regarding ribosome profiling quantification, we now include comparisons between pairs of two replicates for wild type cells, and pairs of replicates wild-type with inert fluorescent protein expression, each pair of samples with their own batch of conditioned MCC medium. These samples were taken under different inducer concentrations, which are expected to affect the expression of two genes and not others. As indicated in Appendix Fig. 1b and reproduced below, the Pearson correlation of log-transformed footprint density is respectively of R2=0.98 and 0.99 (genes with >100 reads mapped), with a 10th to 90th percentile of fold-changes between 0.83 to 1.17, and 0.91 to 1.12. These results are described in the Methods, p. S9, lines 339-345.

Related to this, I see no mention of data availability in the paper. For this study to be useful to others, providing the raw data (unprocessed) would be essential (ideally in a public repository).

Response: We are sorry that the statement on data availability was buried in the original Methods section that was not a part of the merged PDF file. The raw sequencing data were submitted to Gene Expression Omnibus under the accession number GSE162169. The processed data, including fitness scores, mRNA levels, protein synthesis rates, were included as Supplementary Data Tables 1-9. We now moved the data availability statement to the main document at p. 12, lines 512-516.

The presentation of the work is excellent, with very clear figures and text that helped guide the reader through the results. There were a few minor comments:

1. Abstract: "in bacterium Bacillus subtilis" should read "in the bacterium Bacillus subtilis".

Response: This typo is now corrected.

Page 4: "found that under numerous ways" should read "found that under the numerous ways".

Response: This typo is now corrected.

The authors mention that changes in the expression level of RF1 impacted motility and biofilm genes, but not how this impacts fitness. Would they be able to experimentally identify origin of RF1 growth defects in the same way they did for PrmC? This is not essential for the main findings but would help strengthen the work.

Response: The cause of the growth defect under RF1 knockdown is indeed interesting. We now present evidence ruling out the hypothesis that the growth defect is caused by the expression decrease for motility and biofilm genes.

This hypothesis is driven by our result that ablation of SigB regulon rescues the fitness defect during PrmC overexpression (Fig. 3g) and by the observed downregulation of motility and lyt operons and upregulation of the eps operon during RF1 knockdown. To test this hypothesis, we used a strain without sigD (the motility sigma factor), which displays similar expression changes to what we observed in RF1 knockdown (Chai et al., 2009). Comparing the growth rates of wild-type to DsigD, we found only a slight difference (30% growth defect measured upon RF1 knockdown, it appears that transcriptional changes to the motility regulon can only partially explain of the RF1 growth defect. These results are discussed on p. 10, lines 459-463. Further assessment will constitute interesting future research avenues.

It is difficult to know how generalisable the findings of this work are due to the very limited scope. It could be helpful for the authors in the discussion to consider and comment on how such approaches might be scaled-up to enable broader and more general studies of expression-fitness landscapes and where they will find most use.

Response: Indeed, the spurious nature of the expression-fitness landscape makes it difficult to generalize the exact mechanisms that we described here to other proteins. However, what is generalizable is our conclusion that such spurious connections limit the feasibility of bottom-up models for predicting fitness landscapes unless one has near-exhaustive characterization of all parts.

Our approach of mechanistic profiling of cell states under perturbations therefore provides a path forward that can be scaled up by recent developments in multiscale measurements. We now include a discussion for broader and more general studies on p. 11, lines 473-480.

“Various strategies can now generate expression-fitness landscapes for a large number of genes in parallel, for example using suites of promoters (Keren et al., 2016), genome-scale library of inducible gene expression (Arita et al., 2021), or tunable CRISPR perturbations (Hawkins et al., 2020; Jost et al., 2020; Mathis et al., 2021). Together with the advent of single-cell transcriptomics in bacteria (Blattman et al., 2020; Imdahl et al., 2020; Kuchina et al., 2020), these methods open the possibility of dissecting the molecular underpinnings of expression-fitness landscapes genome-wide, and to comprehensively identify instances of regulatory entrenchment.”

Reviewer #3 (Significance (Required)):

This work has a number of contributions. Firstly, it demonstrates how to combine several complementary sequencing approaches to characterize in detail the transcriptional and translational state of a cell, as well as its overall growth rate to generate comprehensive expression-fitness maps. Secondly, it shows how the interwoven nature of cellular regulatory networks and the molecular interactions encoded within the genome can lead to cryptic responses in cellular behavior and fitness at a system-level that can only be understood by taking a detailed "bottom-up" approach. Finally, it suggests that some of these regulatory interactions may in fact "entrench" an organism's evolutionary path, by causing small genetic perturbations to propagate and potentially amplify their negative effect. While the results are compelling and well supported by experiments, the limited scope of the work makes it difficult to know whether this is in fact a rare or common occurrence. However, I do believe there is significance to these findings and that it will likely spur further studies to assess the generality of these findings.

Overall, I believe the work will have a wide appeal covering areas such as Systems Biology, Gene Regulation, Evolution, Quantitative Biology, Sequencing, High-throughput Technologies.

Response: We thank the reviewer for their assessment that our work will be of appeal to a broad audience.

My field of expertise is in the quantitative measurement of core cellular processes (e.g. transcription and translation) using novel sequencing techniques and the application of this knowledge to biological design. As such, I believe I have sufficient expertise to review this paper in detail.

Response references

Adamski, F.M., McCaughan, K.K., Jørgensen, F., Kurland, C.G., and Tate, W.P. (1994). The concentration of polypeptide chain release factors 1 and 2 at different growth rates of Escherichia coli. J. Mol. Biol. 238, 302–308.

Arita, Y., Kim, G., Li, Z., Friesen, H., Turco, G., Wang, R.Y., Climie, D., Usaj, M., Hotz, M., Stoops, E., et al. (2021). A genome-scale yeast library with inducible expression of individual genes. BioRxiv 2020.12.30.424776.

Baranov, P. V, Gesteland, R.F., and Atkins, J.F. (2002). Release factor 2 frameshifting sites in different bacteria. 3, 373–377.

Blattman, S.B., Jiang, W., Oikonomou, P., and Tavazoie, S. (2020). Prokaryotic single-cell RNA sequencing by in situ combinatorial indexing. Nat. Microbiol. 5, 1192–1201.

Chai, Y., Normam, T., Kolter, R., and Losick, R. (2009). An epigenetic switch governing daughter cell separation in Bacillus subtilis. Genes Dev. 7824, 754–765.

Craigen, W.J., and Caskey, C.T. (1986). Expression of peptide chain release factor 2 requires high-efficiency frameshift. Nature 322, 273–275.

Craigen, W.J., Cook, R.G., Tate, W.P., and Caskey, C.T. (1985). Bacterial peptide chain release factors: Conserved primary structure and possible frameshift regulation of release factor 2. Proc. Natl. Acad. Sci. U. S. A. 82, 3616–3620.

Dinçbas-Renqvist, V., Engström, Å., Mora, L., Heurgué-Hamard, V., Buckingham, R., and Ehrenberg, M. (2000). A post-translational modification in the GGQ motif of RF2 from Escherichia coli stimulates termination of translation. EMBO J. 19, 6900–6907.

Hawkins, J.S., Silvis, M.R., Koo, B.M., Peters, J.M., Osadnik, H., Jost, M., Hearne, C.C., Weissman, J.S., Todor, H., and Gross, C.A. (2020). Mismatch-CRISPRi Reveals the Co-varying Expression-Fitness Relationships of Essential Genes in Escherichia coli and Bacillus subtilis. Cell Syst. 11, 523-535.e9.

Imdahl, F., Vafadarnejad, E., Homberger, C., Saliba, A.E., and Vogel, J. (2020). Single-cell RNA-sequencing reports growth-condition-specific global transcriptomes of individual bacteria. Nat. Microbiol. 5, 1202–1206.

Johnson, G.E., Lalanne, J.B., Peters, M.L., and Li, G.W. (2020). Functionally uncoupled transcription–translation in Bacillus subtilis. Nature 585, 124–128.

Jost, M., Santos, D.A., Saunders, R.A., Horlbeck, M.A., Hawkins, J.S., Scaria, S.M., Norman, T.M., Hussmann, J.A., Liem, C.R., Gross, C.A., et al. (2020). Titrating gene expression using libraries of systematically attenuated CRISPR guide RNAs. Nat. Biotechnol. 38, 355–364.

Keren, L., Hausser, J., Lotan-Pompan, M., Vainberg Slutskin, I., Alisar, H., Kaminski, S., Weinberger, A., Alon, U., Milo, R., and Segal, E. (2016). Massively Parallel Interrogation of the Effects of Gene Expression Levels on Fitness. Cell 166, 1282-1294.e18.

Korkmaz, G., Holm, M., Wiens, T., and Sanyal, S. (2014). Comprehensive analysis of stop codon usage in bacteria and its correlation with release factor abundance. J. Biol. Chem. 289, 30334–30342.

Kuchina, A., Brettner, L.M., Paleologu, L., Roco, C.M., Rosenberg, A.B., Carignano, A., Kibler, R., Hirano, M., William DePaolo, R., and Seelig, G. (2020). Microbial single-cell RNA sequencing by split-pool barcoding. Science (80-. ).

Lalanne, J.B., Taggart, J.C., Guo, M.S., Herzel, L., Schieler, A., and Li, G.W. (2018). Evolutionary Convergence of Pathway-Specific Enzyme Expression Stoichiometry. Cell 749–761.

Li, G.W., Burkhardt, D., Gross, C., and Weissman, J.S. (2014). Quantifying absolute protein synthesis rates reveals principles underlying allocation of cellular resources. Cell 157, 624–635.

Mangano, K., Florin, T., Shao, X., Klepacki, D., Chelysheva, I., Ignatova, Z., Gao, Y., Mankin, A.S., and Vázquez-Laslop, N. (2020). Genome-wide effects of the antimicrobial peptide apidaecin on translation termination in bacteria. Elife 9, 1–24.

Mathis, A.D., Otto, R.M., and Reynolds, K.A. (2021). A simplified strategy for titrating gene expression reveals new relationships between genotype, environment, and bacterial growth. Nucleic Acids Res. 49, e6.

Poole, E.S., Brown, C.M., and Tate, W.P. (1995). The identity of the base following the stop codon determines the efficiency of in vivo translational termination in Escherichia coli. EMBO J. 14, 151–158.

Wei, Y., Wang, J., and Xia, X. (2016). Coevolution between Stop Codon Usage and Release Factors in Bacterial Species. Mol. Biol. Evol. 33, 2357–2367.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

In this paper, the authors use a combination of RNA sequencing, ribosome profiling and measurements of cellular composition and growth rate to gain insight into the multi-scale affects that perturbations to translation termination factors have on general physiological states and reproductive fitness using Bacillus subtilis as their model organism. Specifically, they find that perturbing the expression levels of peptide chain release factors in any direction has a negative effect on growth-rate. This negative effect was not due to a direct impact of the gene on the cell, but instead due to a chain of regulatory interactions that cause the activation of the general stress regulon. This leads to upregulation of a large chunk of the genome and an indirect impact on the expression of all other genes. Critically, the knock-on effects observed for the specific perturbations studied suggest that it may be difficult to predict expression-fitness landscapes of a cell, without carrying out a detailed mapping of all genes and the cell's physiological state.

Overall, the core findings in the paper are well justified by the data presented and the experiments appear to have been rigorously carried out. My only concern is that it is unclear if biological replicates of the ribosome profiling were performed. Also, biological replicates are mentioned for the RNA-seq data, but no data is shown. Even a simple graph demonstrating the expression levels across these would be useful to be assured of no issues in reproducibility given the complex processing of the data involved. Related to this, I see no mention of data availability in the paper. For this study to be useful to others, providing the raw data (unprocessed) would be essential (ideally in a public repository).

The presentation of the work is excellent, with very clear figures and text that helped guide the reader through the results. There were a few minor comments:

1. Abstract: "in bacterium Bacillus subtilis" should read "in the bacterium Bacillus subtilis".
2. Page 4: "found that under numerous ways" should read "found that under the numerous ways".
3. The authors mention that changes in the expression level of RF1 impacted motility and biofilm genes, but not how this impacts fitness. Would they be able to experimentally identify origin of RF1 growth defects in the same way they did for PrmC? This is not essential for the main findings but would help strengthen the work.
4. It is difficult to know how generalisable the findings of this work are due to the very limited scope. It could be helpful for the authors in the discussion to consider and comment on how such approaches might be scaled-up to enable broader and more general studies of expression-fitness landscapes and where they will find most use.

Significance

This work has a number of contributions. Firstly, it demonstrates how to combine several complementary sequencing approaches to characterize in detail the transcriptional and translational state of a cell, as well as its overall growth rate to generate comprehensive expression-fitness maps. Secondly, it shows how the interwoven nature of cellular regulatory networks and the molecular interactions encoded within the genome can lead to cryptic responses in cellular behavior and fitness at a system-level that can only be understood by taking a detailed "bottom-up" approach. Finally, it suggests that some of these regulatory interactions may in fact "entrench" an organism's evolutionary path, by causing small genetic perturbations to propagate and potentially amplify their negative effect. While the results are compelling and well supported by experiments, the limited scope of the work makes it difficult to know whether this is in fact a rare or common occurrence. However, I do believe there is significance to these findings and that it will likely spur further studies to assess the generality of these findings.

Overall, I believe the work will have a wide appeal covering areas such as Systems Biology, Gene Regulation, Evolution, Quantitative Biology, Sequencing, High-throughput Technologies.

My field of expertise is in the quantitative measurement of core cellular processes (e.g. transcription and translation) using novel sequencing techniques and the application of this knowledge to biological design. As such, I believe I have sufficient expertise to review this paper in detail.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

The manuscript of Lalanne and coworkers address the cellular responses to varied translation termination factor expression in Bacillus subtilis. The authors set-up a system to fine-tune the expression of release factor RF1, RF2 as well as PrmC that post-translationally modifies RF1/RF2 to maximize their catalytic hydrolysis activity. They then monitor the fitness costs associated with overexpression or depletion of the factor by following the changes in growth rate. The set-up is nicely illustrated in Figure 1. The results in Figure 2 show that overexpression of RF1 and RF2 has relatively modest effect on the growth rate compared to overexpression of PrmC that leads to dramatic growth rate reduction. By contrast, depletion of RF1 has a strong negative influence on fitness, whereas a similar level of depletion of RF2 had little influence on fitness. PrmC overexpression appears to be correlated with the induction of the sigmaB regulon, however, the authors do not manage to ascertain why this is. By contrast, RF2 depletion also results in the induction of the sigmaB regulon and the authors demonstrate convincingly that this is due to a termination defect within the rsbQ-rsbV operon that contains an overlapping start-stop AUGA

A few points that the authors might consider discussing

1. The natural abundance of each RF in bacteria in relation to the usage of different stop codons in different organisms.
2. The role of the frameshifting mechanism in RF2 and how then RF1 levels are regulated.
3. The authors observe queuing in front of the relevant stop codons upon RF depletion, however, do not discuss about readthrough events, which are usually competing with termination. Surprisingly, in this context the authors don't discuss the work from Mankin and coworkers showing sequestration of RFs from termination by peptides such as apideacin leads to translational readthrough.
4. The efficiency of translation termination is well-known to be dependent on the context of the stop codon. Do the authors also observe such a trend. Especially, UGAC for RF2, one would expect to observe high levels of readthrough upon RF2 depletion.

Significance

Overall, the experiments are clearly performed and beautifully illustrated. Clearly, a lot of work has gone into this study but the end message that the cell regulates carefully RF concentrations is not surprising. Especially given that RF2 carefully regulates its own levels using an autoregulatory frameshifting mechanism. The major finding that the rsbQ-rsbV operon with the RF2 dependence leading to induction of the sigmaB regulon is in the end rather trivial since these regulators depend on RF2 for termination. Therefore, this manuscript is unlikely to have general interest to people in the translation field (such as myself) but rather those working in the field of synthetic biology.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

This reviewer did not leave any comments

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Reviewer response

We thank the reviews for the careful reviews, and were delighted to see that they assessed both the quality and significance of the work so highlty.

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

The authors investigated the cross-neutralization capacity of serum antibodies to past and future 229E coronaviruses using 229E spikes isolated from five time points and sera from two different periods (1985-1995 and 2020). They demonstrated a general pattern of asymmetric cross-neutralization, with sera cross-reactive to historical but not future strains. Using chimeras, the authors showed this pattern was mostly driven by antibodies to the evolving RBD. The rate of change in the neutralization titer, a possible measure of antigenic evolution, was estimated to be on par with that of flu B viruses. Interesting differences in individuals' cross-neutralization capacity were observed. The main take-away is that reinfection with 229E is enabled by antigenic escape, not "weak" immunity after infection (as proposed by others).

Thanks for the excellent summary of the paper. We agree with it, although we would note that our work does not exclude “weak immunity” as a possible compounding explanation for re-infection in addition to the antigenic evolution we demonstrate.

**Major comments:**

The key conclusions are convincing and justified by the data. The work is clearly presented and presented with sufficient detail for reproducibility. Characteristically and laudably, the authors have made all the code and data publicly available on GitHub.

Thanks for the favorable summary.

**Minor comments:**

p 3: Perhaps it is clearer to write that 229E has been identified/isolated in humans for >50 y? Or do you really mean to imply (by contrast with "circulated") that NL63 emerged very recently?

This is a good suggestion. We really do not know how long either CoV-229E or CoV-NL63 have been circulating humans, only that CoV-229E was first isolated >50 years ago whereas CoV-NL63 was first identified only in 2003. It is possible both viruses have been circulating for longer than that. We have made the suggested change to clarify.

p 3: An important citation for the antigenic implications of the ladder-like phylogeny AND phylogenetic clustering by date is the classic paper introducing phylodynamics by Grenfell et al. (2004, Science).

Thanks for pointing out this citation; we have added it.

p 4: I might not be like all readers, but I prefer to see a bit in the main text about the source of sera for this kind of study. (I wonder about age, if donors are healthy, etc.)

This is a good question, and we have expanded on it in both the main text and the methods. Briefly, the sera were all from apparently healthy individuals, and no information about recent respiratory virus infections were available. We have provided the age of the serum donor (at the time of serum collection) above the title of each plot showing person-specific neutralization data.

p 4: "Our reason for focusing..." stops short. Is the idea that these are probably people who were recently infected?

This is a good question, and we have elaborated in the revised text. We don’t have any direct information on whether the individuals had recent infections, although that seems plausible. More pragmatically, we reasoned that sera that had reasonably high initial titers would provide better dynamic range to see how titers changed as the virus evolved given our assay has a lower limit of detection.

p 5: Probably my biggest suggestion for the paper is that it mention another relevant study. In 1980, Anne Underwood demonstrated similar asymmetric cross-immunity among early strains of H3N2 (but using rabbits, not human sera), finding that antibodies raised to one strain reacted more strongly by HAI to past strains than to later strains (doi: 10.1128/IAI.27.2.397-404.1980). This relates to the significance of the paper (next section).

Thanks, this is a good and relevant citation, and we have added it when we discuss the possible asymmetry of antigenic change with respect to time.

Obviously, there are citations to update throughout due to the booming SARS-CoV-2 literature.

We have updated the other citations to keep pace with the fast-changing literature!

Reviewer #1 (Significance (Required)):

This study, if anything, undersells itself. Obviously it is a huge contribution to our understanding of how a seasonal coronavirus that bears important phenotypic resemblance to SARS-CoV-2 evolves, but I think it is also providing a foundational piece of evidence--a mechanism--of how rapid viral turnover (by antigenic evolution) occurs. There is no reason to think this should be limited to the coronaviruses, and I suspect the evidence here will go a long way to unifying the evolutionary and epidemiological dynamics of fast-evolving viruses.

Thanks for the praise of the manuscript. Indeed, we were surprised to find that no similarly designed studies have been done even for influenza virus, and so are now interested in expanding our future work to do that as we fully agree it could provide insight more broadly.

Asymmetric competition is nearly an ecological requirement for one strain to successfully invade and displace another. It is thought (unsure how widely?) that flu evolves antigenically, with new strains eventually displacing old ones, by mutating at key epitopes in ways that the immune system does not immediately pick up. That is, immune memory is biased to recall responses to conserved epitopes, which on average are probably less neutralizing. This will induce competition between mutant and resident viruses, but it would be symmetric, since infection with either would induce responses to conserved epitopes on the other. But if on infection with the mutant, immune memory sometimes reuses (boosts) antibody responses to target the mutated epitopes, those recycled antibodies might be less effective against the mutant, making the competition asymmetric.

What this paper and Underwood (1980) suggest is that we can get this asymmetric, antibody-mediated competition fairly easily and without extensive memory. Underwood showed this more powerfully in rabbits, but in this paper too we see an indirect suggestion of asymmetry in relatively inexperienced children (Fig. 3). Mutants (future strains) successfully invade when they can trigger presumably recalled antibodies that are more harmful to the resident (soon historical) strain than the mutant. If this is so easy to do, as judged by the extensive data here, then it could be common.

I've gone off on a theoretical limb here, but the paper is still important without these considerations. This work will be of interest to evolutionary biologists, epidemiologists, vaccinologists, and everyone else wondering what SARS-CoV-2 will do next and how immunity to antigenically variable pathogens works.

We completely agree with the ideas mentioned above, and appreciate having it put in this nice context, particularly alongside the Underwood paper (with which we were not previously familiar). That said, we believe that the small number of recent children sera samples in the current study preclude us from drawing strong conclusions about the asymmetry--as the reviewer says, our data provides an indirect suggestion too. So overall we have not tried to expand this angle here because as the reviewer says, the paper is still important without these considerations. However, we are actively working to see if we can design a similar study with more children sera in the future to separately address the questions about asymmetry.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

An important question in coronavirology is what governs their ability to seemingly reinfect people regularly (within 2 or 3 years). While waning protective immunity has been proposed and is of current concern for SARS CoV-2, the role of antigenic drift driven by escape from neutralizing antibodies has not been well characterized. The authors have attempted to look at this through examining historical Spike proteins from HCoV-229E over a period of 30-odd years. The authors show that 229E evolves along a linear trajectory consistent with yearly selection by pre-existing immunity. Taking representative spike proteins from different time points into pseudovirus neut assays, they find that older spike proteins are less sensitive to neut by more recent sera. Conversely, spike proteins from prior to the birth of an individual display markedly less sensitivity to neut that those prevalent during the persons lifetime. Sequence analysis of the spike shows variation accruing in both N-termina regions and the RBD, parts of spike predominantly targeted by nABs. Lastly producing early spikes with chimeric RBDs from late viruses enhances the sensitivity to more recent sera.

This is a potentially important MS that addresses a pertinent question that is of wide interest for the CoV2 pandemic. While it is limited in addressing the relative contribution of antigenic escape vs waning Ab titers because of the nature of the sample, the MS makes a strong case for Spike evolution being driven by antigenic escape.

Thanks for the summary. We agree that our paper does not really address waning immunity because we don’t have sequential serum samples from the same individual. However, it does clearly show that antigenic evolution is important independent of waning immunity, because all of the experiments (e.g., Figure 2 and 3) show the same serum sample tested against newer spikes, and neutralization titers definitely decrease as the spike evolves. The reviewer is correct that this doesn’t rule out the possibility of waning immunity as a separate phenomenon, and we have been sure to emphasize that in the revised text.

Reviewer #2 (Significance (Required)):

While the Figs 1-3 are clear, the data in Fig 4 is somewhat preliminary. In all likelihood many people are making neutralizing antibodies both against RBD and the N-terminal region and the relative proportion probably underlies the variability in the data in Fig 4B. I think the MS would benefit from the following:

A comparison of NTD vs RBD vs NTD/RBD chimeras in Fig 4B to give a fuller picture of antigenic escape with statistical support.

The reviewer is correct that our manuscript does not provide a decisive answer on the relative role of NTD versus RBD targeting antibodies, although the data in Fig. 4B clearly show that RBD antibodies are important for many individuals as simply changing the RBD to that of newer viruses recapitulates the full spike antigenic evolution without any changes in the NTD or elsewhere (e.g., subject SD87_2 or SD85_3 in Fig 4B). However, for some other individuals NTD antibodies may play a role.

In general, full dissection of the role of RBD versus NTD antibodies is beyond the scope of our study (and in some cases not even possible with the available volumes of the older serum). In any case, the major point of our study—the first experimental demonstration that seasonal coronaviruses undergo antigenic evolution—does not depend on dissecting the relative roles of RBD and NTD antibodies. We have therefore added new text explaining that we cannot fully parse the relative role of antibodies to these domains beyond knowing that RBD antibodies play n important role. We have added text to emphasize that antibodies to other regions including the NTD could also be important.

A figure to map the polymorphic residues in Fig 4A onto the 229E spike structure to visualise their position and special relatedness, with perhaps a comparison with the latest knowledge of SASR CoV-2 epitopes.

We agree that visualizing the variable sites on the structure is useful and have added such a visualization as a new panel in Figure 4. This allows us to more clearly show the clustering of variability in the RBD and NTD. This clustering of mutations in those regions is consistent with what is currently being seen with the emergence of SARS-CoV-2 variants with mutations in those regions of spike. However, given the divergence between SARS-CoV-2 and CoV-229E, we are not able to do a more fine-grained comparison of epitope sites as many important sites in the RBD and NTD do not have a clear one-to-one alignment (for instance, the RBD’s don’t even bind the same receptor).

Additional discussion to reflect the new SARS CoV-2 variants and their potential selection by escape in the light of the authors data.

We have updated the manuscript to describe the new SARS-CoV-2 variants (which mostly emerged after submission of our original manuscript) and how this emerging antigenic evolution of SARS-CoV-2 is consistent with what we saw in CoV-229E.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

An important question in coronavirology is what governs their ability to seemingly reinfect people regularly (within 2 or 3 years). While waning protective immunity has been proposed and is of current concern for SARS CoV-2, the role of antigenic drift driven by escape from neutralizing antibodies has not been well characterized. The authors have attempted to look at this through examining historical Spike proteins from HCoV-229E over a period of 30-odd years. The authors show that 229E evolves along a linear trajectory consistent with yearly selection by pre-existing immunity. Taking representative spike proteins from different time points into pseudovirus neut assays, they find that older spike proteins are less sensitive to neut by more recent sera. Conversely, spike proteins from prior to the birth of an individual display markedly less sensitivity to neut that those prevalent during the persons lifetime. Sequence analysis of the spike shows variation accruing in both N-termina regions and the RBD, parts of spike predominantly targeted by nABs. Lastly producing early spikes with chimeric RBDs from late viruses enhances the sensitivity to more recent sera.

This is a potentially important MS that addresses a pertinent question that is of wide interest for the CoV2 pandemic. While it is limited in addressing the relative contribution of antigenic escape vs waning Ab titers because of the nature of the sample, the MS makes a strong case for Spike evolution being driven by antigenic escape.

Significance

While the Figs 1-3 are clear, the data in Fig 4 is somewhat preliminary. In all likelihood many people are making neutralizing antibodies both against RBD and the N-terminal region and the relative proportion probably underlies the variability in the data in Fig 4B. I think the MS would benefit from the following:

• A comparison of NTD vs RBD vs NTD/RBD chimeras in Fig 4B to give a fuller picture of antigenic escape with statistical support.

• A figure to map the polymorphic residues in Fig 4A onto the 229E spike structure to visualise their position and special relatedness, with perhaps a comparison with the latest knowledge of SASR CoV-2 epitopes.

• Additional discussion to reflect the new SARS CoV-2 variants and their potential selection by escape in the light of the authors data.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

Evidence, reproducibility and clarity

The authors investigated the cross-neutralization capacity of serum antibodies to past and future 229E coronaviruses using 229E spikes isolated from five time points and sera from two different periods (1985-1995 and 2020). They demonstrated a general pattern of asymmetric cross-neutralization, with sera cross-reactive to historical but not future strains. Using chimeras, the authors showed this pattern was mostly driven by antibodies to the evolving RBD. The rate of change in the neutralization titer, a possible measure of antigenic evolution, was estimated to be on par with that of flu B viruses. Interesting differences in individuals' cross-neutralization capacity were observed. The main take-away is that reinfection with 229E is enabled by antigenic escape, not "weak" immunity after infection (as proposed by others).

Major comments:

The key conclusions are convincing and justified by the data. The work is clearly presented and presented with sufficient detail for reproducibility. Characteristically and laudably, the authors have made all the code and data publicly available on GitHub.

Minor comments:

p. 3: Perhaps it is clearer to write that 229E has been identified/isolated in humans for >50 y? Or do you really mean to imply (by contrast with "circulated") that NL63 emerged very recently?

p. 3: An important citation for the antigenic implications of the ladder-like phylogeny AND phylogenetic clustering by date is the classic paper introducing phylodynamics by Grenfell et al. (2004, Science).

p. 4: I might not be like all readers, but I prefer to see a bit in the main text about the source of sera for this kind of study. (I wonder about age, if donors are healthy, etc.)

p. 4: "Our reason for focusing..." stops short. Is the idea that these are probably people who were recently infected?

p. 5: Probably my biggest suggestion for the paper is that it mention another relevant study. In 1980, Anne Underwood demonstrated similar asymmetric cross-immunity among early strains of H3N2 (but using rabbits, not human sera), finding that antibodies raised to one strain reacted more strongly by HAI to past strains than to later strains (doi: 10.1128/IAI.27.2.397-404.1980). This relates to the significance of the paper (next section).

Obviously, there are citations to update throughout due to the booming SARS-CoV-2 literature.

Significance

This study, if anything, undersells itself. Obviously it is a huge contribution to our understanding of how a seasonal coronavirus that bears important phenotypic resemblance to SARS-CoV-2 evolves, but I think it is also providing a foundational piece of evidence--a mechanism--of how rapid viral turnover (by antigenic evolution) occurs. There is no reason to think this should be limited to the coronaviruses, and I suspect the evidence here will go a long way to unifying the evolutionary and epidemiological dynamics of fast-evolving viruses.

Asymmetric competition is nearly an ecological requirement for one strain to successfully invade and displace another. It is thought (unsure how widely?) that flu evolves antigenically, with new strains eventually displacing old ones, by mutating at key epitopes in ways that the immune system does not immediately pick up. That is, immune memory is biased to recall responses to conserved epitopes, which on average are probably less neutralizing. This will induce competition between mutant and resident viruses, but it would be symmetric, since infection with either would induce responses to conserved epitopes on the other. But if on infection with the mutant, immune memory sometimes reuses (boosts) antibody responses to target the mutated epitopes, those recycled antibodies might be less effective against the mutant, making the competition asymmetric.

What this paper and Underwood (1980) suggest is that we can get this asymmetric, antibody-mediated competition fairly easily and without extensive memory. Underwood showed this more powerfully in rabbits, but in this paper too we see an indirect suggestion of asymmetry in relatively inexperienced children (Fig. 3). Mutants (future strains) successfully invade when they can trigger presumably recalled antibodies that are more harmful to the resident (soon historical) strain than the mutant. If this is so easy to do, as judged by the extensive data here, then it could be common.

I've gone off on a theoretical limb here, but the paper is still important without these considerations. This work will be of interest to evolutionary biologists, epidemiologists, vaccinologists, and everyone else wondering what SARS-CoV-2 will do next and how immunity to antigenically variable pathogens works.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

**Summary:**

This study of Nils Halberg and colleagues aims to characterize tumor-associated immune cell

infiltrates in a mouse model of diet-induced obesity. Authors compared different syngeneic

tumor cell lines for mammary adenocarcinoma and pancreatic ductal adenocarcinoma. Tumor

infiltrating leukocytes were analyzed by a 36-parametric mass cytometry protocol. The authors

put a lot of efforts in the generation of high-quality data by applying state-of-the-art methods for

sample barcoding and batch analyses, removal of batch-specific variations and in the

subsequent pipeline of data analysis. The clinical relevance of the topic addressed is well

documented in several studies, showing a clear association between obesity and the

development of several tumors, including those tumors investigated in this study.

Main findings of this study can be summarized that in the model system used tumor-dependent

differences in the qualitative and quantitative composition of immune cell infiltrates were

observed. Unfortunately, the mouse model system used obviously did not reveal convincing

data whether obesity may modulate the process of tumor infiltration.

The manuscript is well written, quantity of figures is appropriately and of excellent quality and

prior studies were referenced appropriately.

In conclusion, authors made tremendous methodological and technical efforts to generate

robust and high-quality mass cytometry data, but the overall outcome of the study remains

limited in respect to shedding some new light how obesity is possibly involved in the qualitative

and quantitative modulation of tumor-related immune cell infiltration.

Authors’ Response: We thank the reviewer for their constructive and positive feedback as well

as appreciation of our rigorous approach. We would however argue that our data significantly

contributes to the understanding of how obesity affects tumor immunity. We believe that our

systemic approach across multiple tumor systems highlights that i) it matters what model you

choose, as each model have a separate response to the obesity challenge ii) for one model, the

E0771 model, our data reflect obesity-dependent alterations to the CD8+ T-cells population.

This was corroborated by a parallel publication by Rigel et al., 2020 as highlighted by the 2nd

reviewer. That being said, we too, were surprised that the pro-inflammatory obese environment

did not have more pronounced effects on the tumor immune infiltrates across the five models.

**Major comments:**

Due to the limited data really showing an association between obesity and immune cell

infiltration of tumors investigated I would suggest that authors should change emphasis of their

results more closely related to the findings of tumor-dependent immune cell infiltrations than

obesity-related associations. So, the title of the study should be appropriately changed since

"High dimensional immunotyping of the obese tumor micro-environment" rather implies

analyses of spatial relationships of immune, tumor and fat cells by immunohistological analyses,

which would indeed help to strengthen the outcome of this mass cytometry study.

Authors’ Response: We appreciate the constructive suggestion. We did not intend the title to

infer immunohistochemical analysis and apologize that was the case. We have therefore

changed the title to “High dimensional immunotyping of tumors grown in obese and non-obese

mice” in the revised version (line 1).

Although all the efforts made in mass cytometry data generation are quite commendable in this

study, basic statistical issues are not clearly addressed regarding the number of biological

replicates. How many mice were treated per tumor cell line? According to figure 1B nine chow

and eight HFD animals were used: does this mean that only one or two mice were analyzed per

cell line, respectively? Please explain how many animals belong to each of the seven mouse

cohorts.

Authors’ Response: We agree that this was not clearly defined in the manuscript. We have

updated Figure 1A and the corresponding legend to make it clearer. The mouse numbers,

referred to as tumors, are also located in Table 3. In total 69 mice were used, distributed as:

E0771_1 consists of 4 chow fed mice and 4 HFD mice (N=4/4, where N=chow/HFD, for a total

of 8 mice)

E0771_2 is N=5/4. Wnt1 is N=6/6. TeLi is N=5/5. C11_1 is N=5/4. C11_2 is N=5/5. UN-KC is N=5/6.

Figure 1B shows representative mouse weights only. The female mice are from breast cancer

cohort E0771_1 and the male mice are from pancreatic cancer cohort C11_1. We chose to only

show representative data since diet-induced obesity is well established in the C57Bl/6 strain.

Obviously, cell lines E07771 and C11 were analyzed as duplicates only. Regarding E0771,

tumor growth was 31 and 23 days, respectively. So, large inter-individual differences in tumor

growth were obvious and how this is reflected at the level of tumor infiltration? Therefore, please

explain which criteria were used to decide when the tumors had to be removed. Furthermore,

please indicate weight, viability and absolute cell number of each tumor sample in a

supplementary table to get an impression about variability in tumor growth.

Authors’ Response: The reviewer brings up an important point. The E0771 and C11 cohorts are

included in the paper as combined cohorts. The individual C11 cohorts had too few tumors

remaining after removal of samples with too low viability (as discussed below) to analyze

separately. The E0771 cohorts are presented together as a representation of that tumor model.

Data analysis for the E0771 cohorts separately shows comparable population abundance

differences and obesity-dependent changes between chow and HFD tumors. The metacluster

fold change for non-obese and obese tumors between E0771_1 and E0771_2 correlated with a

R2 = 0.8586. Presenting the data combined provides a more concise view of the model.

Removal of E0771 and C11 tumors in each individual cohort were time matched. E0771 tumors

were continuously measured by caliper and removed before they reached 1 cm3 according to the

local ethical guidelines. The E0771_2 cohort tumors had to be removed sooner as one tumor

reached 1 cm3 earlier. We have reported the tumor mass in Figure 1C as that is a more accurate

measurement of final tumor burden. Pancreatic tumors were removed based on optical imaging

of luciferase expressing cancer cells and careful monitoring of mouse distress based on the

grimace scale. The material and methods section has been updated to reflect this (line 556-558).

Only pancreatic tumors had viability poor enough that they had to be excluded from analysis. A

cutoff of CD45+ 5000 cells was set and applied to cells remaining following the gating strategy

shown in Figure 1D. Therefore, CyTOF data for tumors with fewer than 5000 CD45+/Cisplatin

negative cells were excluded from analysis as indicated with an X in Figure1C. As requested, we

have included tumor weights and available viability measurements in new Table 5.

**Minor comments:**

The generation of orthotopic pancreatic cancer mouse models is technically challenging, and

needs more complex imaging methods to monitor the growth of the implanted tumor cells.

Furthermore, orthotopic implantation of tumor cells into the pancreas by surgery can also inflict

significant physical trauma to the recipient animals. How authors have monitored tumor cell

implantation?

Authors’ Response: We agree that tracking tumor growth in the orthotopic pancreas cancer

model is challenging. As mentioned above, these cells were engineered to express luciferase

and optical imaging was used to monitor growth of the implanted cells. We did not report these

numbers as we were unable to convincingly correct for possible light absorption by the

enhanced adipose tissue mass in the high fat group. As such, these scans were used to

estimate the end point.

The number of CD45-positive cells per tumor sample is not given in the manuscript, but this

information would be important to know, because it can be expected that most of the samples

showed less than 20.000 cells. This relatively low number of total leukocytes would not allow a

statistically significant profiling of rare cell subsets, such as DC's or MDSC's. This limitation

should at least clearly addressed in the discussion section.

Authors’ Response: The reviewer raises a great point. Since the cells were live cryopreserved

and thawed before measuring CD45, we did not determine the total immune cell infiltrate. After

thawing, the CD45+ cells accounted for roughly 1-12% of the total events collected across all

batches leading to a total number of CD45+ cells ranging between 54,317 and 1,102,767 per

batch. Numbers for each batch can be found in Table 3. After gating and exclusion of tumors

with less than 5000 CD45+ cells, the remaining tumor data were equally sampled and 5206

CD45+ cells were included in further analysis from each tumor. Overall, we were focused on

broad phenotyping of the immune infiltrate and not on rare subsets. Some subsets had low

abundance in some tumors and high abundance in others. Because the analysis was performed

altogether, the overall phenotyping and clustering did not find any truly rare subsets. DCs and

MDSC were not rare when assessed across the datasets. While we cannot characterize the

subsets that are small in a specific tumor type, we can be confident in the characterization

provided by the streamlined analysis of the data as a whole.

According to table 2 authors have used 36 immune cell-related antigens including casp3, which

was only used to exclude apoptotic cells from downstream analyses. But as written in the

results section only 26 phenotyping markers were used to generate the viSNE map shown in

Figure 3. In Figure 3C-F 30 markers were shown. Please explain this obvious inconsistency of

markers used.

Authors’ Response: Thank you for this question. Our goal here was to generate a viSNE map

that best separated out immune cells by phenotype. Lineage markers and well-established

phenotyping markers were therefore included to create the well separated viSNE map. It follows

that some markers were not included: i) Markers that were used to gate the population of

interest (CD45 and c-Cas3) were excluded from the viSNE input parameters.; ii) Markers that

had relatively low signal were also excluded such as MHC-1 and CD117. Including negative

markers is computationally costly, provides limited biological insight, and can produce a worse

viSNE map by reducing cell separation due to shared lack of signal (Diggins et al., 2015); iii)

Activation/ exhaustion markers were excluded from the viSNE analysis because the focus of the

phenotyping was on major cell subsets and not on activation states. The hope was to observe

differences in exhaustion marker expression between chow and HFD; and iv) CD5 was

excluded because having two bright T cell markers skewed the map towards a more T cell

dominant view. Markers with meaningful expression were reintroduced in the MEM analysis

after the viSNE map was made. Exclusion of markers from viSNE analysis is a generally

accepted practice and has been applied previously (Wogsland et al., 2017, Cheng et al., 2016,

Huse et al., 2019, Leelatian et al., 2020, Doxie et al., 2018, Okamato et al., 2021, Henderson et

al., 2020). The reasoning behind using the 26 phenotyping markers have been included in the

revised manuscript (line 754 – 757)

How viability of tumor samples was determined?

Authors’ Response: Viability was measured at three points using membrane exclusion assays.

Viability was first measured upon tumor dissociation using trypan blue and a Countess cell

counter on the single cell suspension before freezing. Values were used to guide cell aliquoting

for cryopreservation. Viability was again measured with trypan blue upon thaw in order to

barcode and stain 3 million live cells per sample. Before fixation, cells were again stained for

viability, this time with cisplatin, to exclude dead cells after data collection with gating. This has

been added to the methods section (line 559-562)

Cells were additionally stained for cleaved-Cas3 as an indicator of cells undergoing apoptosis.

Only pancreatic tumors had viability poor enough that they had to be excluded form analysis.

Tumors with fewer than 5000 CD45+ Cisplatin negative cells were excluded from analysis as

indicated with an open X in Figure1C. The tumor count in parentheses in Table 3 indicates the

tumors that were not excluded.

Please indicate cell loss caused by cryopreservation of dispersed tumor tissue samples.

Authors state that mainly neutrophilic granulocytes will be lost during cryopreservation, and that

this would help to the "definitive identification and characterization of G-MDSC". But there are

several reports showing that MDSC-subsets also behave very sensitive during cryopreservation

and that it is recommended to analyze fresh samples if MDSC's are of particular interest (DOI:

10.1177/1753425912463618; DOI: 10.1177/1753425912463618). This possible limitation

should be discussed in the manuscript and not only highlighted as advantage on the way to

identify MDSC-subsets.

Authors’ Response: We thank the reviewer for this insightful comment. We agree that we likely

lost some MDSC during the cryopreservation process as shown in the reference above. But

since no neutrophils survive standard cryopreservation (Graham-Pole et al., 1977), the Ly6G

positive cells in our analysis are G-MDSC and not neutrophils. We assume that any cell death

related to cryopreservation would be consistent across samples, so although cell totals may be

lower than in the tumor, abundance differences and phenotype can still be evaluated. We have

included a discussion of this in the revised manuscript

(line 408 – 410).

In the Figure 1D X-axis named by "193Ir-NA" should be replaced by "193Ir-DNA".

Authors’ Response: NA is shorthand for nucleic acid since the iridium intercalates into DNA

and RNA. The figure legend has been updated to make this clear.

Furthermore, please explain "(T)" in the figure legend. Percentages in the last two dot plots

related to "all previous gates" are confusing: 20,44% of all DNA-containing single cells were

finally intact, living CD45+ cells, i.e. almost 80% of cells were excluded because they were dead

or apoptotic and this corresponds to 57,06% of intact, living CD45-positive cells related to all

CD45-positive cells? How these percentages are related to the "Percent of CD45/total raw

events" in the last column of Table 3?

Authors’ Response: These are great points. Thank you for bringing them to our attention. This

confusing notation has been removed since Figure 1D is a representative gating strategy. “All

previous gates” means that the previous gates were all applied to the population showing in that

plot. CyTOF data requires thorough gating to remove the events that are not representative of

actual cells so yes, many events were removed before analysis. Even more cells were excluded

here since our focus was on the CD45+ cells and not the cancer cells. The CD45+ cells

indicated in Table 3 and visualized in Figure 1E can be calculated by summing the total gated

CD45+ cells per Figure 1D for each batch and dividing that by the total number of events

collected per batch. The summed CD45+ values and the total collected events are also in Table

3.

Authors claimed that "155Gd_IRF4" was changed to "155Gd", but it is not clear why to mention

that IRF4 has been NOT used throughout the study? Please provide only those technical

details, which are necessary to understand what has been done.

Authors’ Response: We apologize for any confusion. This change was mentioned because

most cohorts included the IRF4 channel while a two (C11_2 and Wnt_1) did not. The FCS files

were changed to allow for simultaneous analysis. The IRF4 antibody did not work so there

shouldn’t be any bleed into other channels in the samples that were stained with IRF4.

According to general practice, we believe that it is important to make note of any manipulation to

the FCS files.

Re Figure 6: please explain the abbreviation "TNBC".

Authors’ Response: We apologize for not explaining this abbreviation. TNBC is short for triple

negative breast cancer. This has been corrected in the resubmitted version.

Experiments done with TKO mice are not described in the Materials and Methods section. In

particular, it would be important to know the number of replicates and the number of tumors

grown in this model. It should be also discussed that the growth kinetics of tumors in chow and

HFD TKO mice seem to be much faster as compared to wild type mice. Principally, the TKO

model used here is only of limited value to clarify especially the role of CD8 cells since all other

T- and B- cell subsets including NK cells are also absent in this knockout model and indirect

effects caused by these cells cannot be excluded.

Authors’ Response: We deeply apologize that the TKO experiments were not included in the

Triple knockout (Rag2-/-::CD47-/-::Il2rg-/-; TKO) mice were purchased from Jackson Laboratories (Stock No: 025730).

We agree with the reviewer it is an important point that the E0771 tumors overall grew faster in the TKO model. Ringel et al. 2020 saw similar results when depleting CD8 T cells in their MC38 model. Comparably, the most striking difference observed was that the tumor growth between obese and non-obese mice disappeared in the TKO mice.

We have modified the results section to include these points (line 309-310). Reviewer #1 (Significance (Required)):

material and methods section.

experiment was performed with N=5/5. The description of the TKO model has been added to

Orthotopic implantation and

monitoring of E0771 and C11 cells were performed as with the wild type C57BL/6 mice. Each

the methods section (line 520- 533) and number of mice used has been added to the figure

legend.

did not observe any major growth changes (overall growth rate and growth differences between

obese and non-obese mice) in the TKO mice compared to the wild type mice.

In the C11 model, interestingly, we

We agree that the combined lack of B- and NK- cells in combination to the lack of T-cells

exclude a direct conclusion on the effect of obesity-dependent alteration in T-cell phenotypes.

Altogether, this study is a paragon that a single technology-based study alone, even when well-

designed, is not sufficient to explore complex tumor microenvironment-immune cell interactions

and that additional information on spatial relationships of cells and possibly single cell-based

RNAseq techniques are necessary to shed new light on this ambitious topic. But there is no

doubt that the potential of mass cytometry has been not fully exploited in this study and that a

more focused view on particular cell types identified so far, such as macrophages or CD8 cells,

by using as many immunophenotypic and functionally-related parameters as necessary would

allow a more in depth-phenotyping of particular immune cell compartments.

The significance of this subject would have been tremendously increased if human samples will

be analyzed in a future confirmative study.

Authors’ Response: Again, these are important insights. To what extend we have taken full

advantage of the suspension mass cytometry technology is of course debatable. When we set

out to perform these studies, we were compelled to take a broad approach rather than focusing

on a single cell type for the following reasons: i) we had noticed extreme variability in immune

targeted analysis through FACS of murine cancer models. Since we set out to demonstrate

systemic effects of the obese environment rather than model-specific effects, the broad antibody

panel made the most sense and ii) tumor immune infiltrates are known to be composed of

multiple cell types and the effect of the obese state would likely affect multiple of these. To not

bias ourselves this prompted us to design a rather broad immune panel. With the knowledge

derived from this study and others (For example Rigel et al, Cell 2020 and Chung et all, Cell

2020), new and more focused panels could be developed and implemented for future studies.

We agree that the inclusion of human data would be of great value. We were, however, unable

to obtain suitable human material that could be used for this suspension mass cytometry

analysis. This was largely due to large inconsistencies in reported patient BMI and inadequate

tumor freezing conditions.

Even when I'm not a specialist in tumor biology, based on my expertise in the fields of chronic

inflammation and cytometry, I'm convinced that the outlined way of generating

immunophentypic data by single cell-based mass cytometry is of major interest not only for

tumor biologists, but will be for sure recognized by a broad scientific community interested in the

generation of single cell-based immunophenotypic data.

Authors’ Response: Thank you for your helpful and supportive feedback. It is indeed our hope

and motivation that the immunophenotyping platform presented herein will be broadly applicable

to other cancer immunologists and fields.

Reviewer #2

(Evidence, reproducibility and clarity (Required)):

Wogsland et al. apply herein mass cytometry (CyTOF) to investigate how obesity affects tumor

immune infiltrates. They use several models of murine breast and pancreatic cancers and

analyse their immune landscape thanks to an extended panel of 36 markers. They notably

describe a decrease in CD8 T cells in one breast cancer model fed with high fat diet inducing

obesity which favors tumor development.

Overall, the report is clearly written and follows a very logical plan. Figures are also clear and

nicely support the text. The mass cytometry approach appears quite original and could be

relevant for many readers.

Authors’ Response: We thank the referee for their constructive and positive comments on our work.

The referee raises the general criticism that our study is descriptive.

Nevertheless, some concerns have to be made and would need to be acknowledged by

authors:

-First of all, the paper appears very descriptive. Except at the end of the last figures, authors

only establish of catalog of immune cells in different tumors. Even if the trueness of such

observations is undisputable, their relevance to improve our understanding of tumor biology is

clearly questionable.

Authors’ Response:

While we agree that the majority of the manuscript is focused on establishing a robust immune

atlas in multiple tumor models grown in obese and non-obese mice, we believe that such work

has important merit: i) our immune cell atlas of 5 transplant models will be a valuable resource

for other cancer researchers interested in the immune-oncology field (as also highlighted by the

first referee); ii) our findings clearly underscore the critical need to apply multiple cell lines in

experimental setups when studying the interaction between tumors and immune cells –

particularly in the obese setting; iii) we have implemented an analysis pipeline that is broadly

applicable for high dimensional mass cytometry data that will be useful for future high-

dimensional immunotyping efforts, iv) through our unbiased analysis pipeline we did identify

obesity-dependent alterations to the CD8 cell population in the E0771 model. This finding was

corroborated by the studies by Ringel et al., 2020. Collectively, we strongly believe that our

studies will contribute to the advancement of our understanding of tumor biology.

-Moreover, the major finding claimed in this study (CD8 T cells decrease in tumors from HFD

mice) has been very recently published paper also providing mechanistic insights (Ringel et al.,

Cell 2020). Authors could legitimately be disappointed but the interest of their study is sadly

severely impacted by this prior publication. This key paper should be at least discussed and

included in references.

Authors’ Response: The paper by Ringel et al., was published after we originally submitted our

manuscript for review and was therefore not referenced or discussed (Ringel et al., 2020). In the

resubmitted version of the manuscript, we have included a thorough discussion of the paper’s

findings in terms of consistencies and inconsistencies with our conclusions (line 545-463).

-Finally, even if the initial strategy of integration of breast and pancreatic cancers was

indubitably a good one, results reported in figures 5 & 6 clearly show a quite specific

observation in the E0771 model. So in this context, integrating all these datasets do not improve

the understanding of this phenotype

Authors’ Response: We thank the referee for bringing up this point. Regardless of the outcome, we would strongly argue that the integrated approach to be advantageous to individual analysis. The integrated approach did not hinder new discoveries in any of the datasets – if anything, the integrated analysis pipeline developed herein would facilitate new discoveries that would be missed by repeating individual analysis. By integrating the datasets, we enabled the robust identification of more cell subsets. In particular cell types that displayed low abundance in some models. Those cells would have likely been hidden or even missed in a larger subset had the models been analyzed separately. As such, we maintain that the integrated approach is the correct and most biological meaningful to follow when given the possibility.

Besides these quite general comments, few more specific points:

-In Fig 2, the F4/80 signal appears very weak in all datasets except one (TeLi) with an almost

flat curve for all the other ones. It asks the question of the reproducibility of the staining that

could be only partially corrected with batch correction algorithms.

Authors’ Response: The F4/80 peak in the TeLi cohort is indicative of a large F4/80

population rather than a sign of signal intensity differences. TeLi tumors had much higher

abundance of F4/80+ cells than did the other tumor types as can be seen in Figure 4B. For

each mass cytometry run, we included a control sample to ensure equal staining patterns

between the antibodies in each run.

-Obesity is clearly known to be sex/hormone dependent as confirmed by authors themselves in

their Fig 1B so again the global integration (both sex and 2 organs) strategy is disputable here.

It is hard to know if there is no effect in the pancreas because of tissue or sex specificities.

Authors’ Response: Thank you for the feedback. We specifically tried to show the different

tumors side by side without making too many comparisons across tumor types because of the

sex and tissue differences (as was noted in the results section of the manuscript, line 267). Both

breast and pancreatic tumor models are relevant for studying the obesity cancer connection

which is why we have worked to develop these models with different cancer types. Even with

the sex and tissue differences, male and female mice became obese on a high fat diet, and

tumors from both tissues grew larger on a high fat diet

. It is our hope that this work will pave the

way for future studies to interrogate these differences.

-On Fig 1C, red dots are closed or open but explanation of this is lacking.

Authors’ Response: Thank you for pointing this out. The figure legend has been updated. The

X indicates a tumor that had too few live CD45+ cells to be included in the data CyTOF

analysis. We apologize this was not clear.

-Authors use 36 markers in their CyTOF panel but use only 26 for the dimension reduction

without clearly explaining this choice. Should be amended. For example, why excluding CD5?

Authors’ Response: Thank you for bringing this up. We have addressed these concerns in

response to Reviewer #1 above.

Reviewer #2 (Significance (Required)):

Severly impaired by Ringel et al., Cell 2020

Authors’ Response:

It is clear that the study by Ringel et al., demonstrate new and important mechanistic insights

into the connection between obesity, T-cell biology and tumor behavior. Our studies share many

of the same conclusions on tumor immune cell infiltrate in obesity – particularly the T-cell finding

in our E0771 model. However, we stipulate that our experimental approach and scientific

questions differ. Our approach was to generate high-dimensional immune phenotyping atlas

across multiple models to identify overarching obesity-dependent effects. The manuscript by

Ringel et al., has a more mechanistic focus. The field would benefit from the additive insights

from the two papers combined.

Rebuttal References:

CHENG, Y., WONG, M. T., VAN DER MAATEN, L. & NEWELL, E. W. 2016. Categorical Analysis of Human T Cell Heterogeneity with One-Dimensional Soli-Expression by Nonlinear Stochastic Embedding. The Journal of Immunology, 196, 924-932.

DIGGINS, K. E., FERRELL, P. B., JR. & IRISH, J. M. 2015. Methods for discovery and characterization of cell subsets in high dimensional mass cytometry data. Methods, 82, 55-63.

DOXIE, D. B., GREENPLATE, A. R., GANDELMAN, J. S., DIGGINS, K. E., ROE, C. E., DAHLMAN, K. B., SOSMAN, J. A., KELLEY, M. C. & IRISH, J. M. 2018. BRAF and MEK inhibitor therapy eliminates Nestin-expressing melanoma cells in human tumors. Pigment Cell & Melanoma Research, 31, 708-719.

GRAHAM-POLE, J., DAVIE, M. & WILLOUGHBY, M. L. 1977. Cryopreservation of human granulocytes in liquid nitrogen. Journal of Clinical Pathology, 30, 758.

HENDERSON, L. A., HOYT, K. J., LEE, P. Y., RAO, D. A., JONSSON, A. H., NGUYEN, J. P., RUTHERFORD, K., JULÉ, A. M., CHARBONNIER, L.-M., CASE, S., CHANG, M. H., COHEN, E. M., DEDEOGLU, F., FUHLBRIGGE, R. C., HALYABAR, O., HAZEN, M. M., JANSSEN, E., KIM, S., LO, J., LO, M. S., MEIDAN, E., SON, M. B. F., SUNDEL, R. P., STOLL, M. L., NUSBAUM, C., LEDERER, J. A., CHATILA, T. A. & NIGROVIC, P. A. 2020. Th17 reprogramming of T cells in systemic juvenile idiopathic arthritis. JCI Insight, 5.

HUSE, K., WOGSLAND, C. E., POLIKOWSKY, H. G., DIGGINS, K. E., SMELAND, E. B., MYKLEBUST, J. H. & IRISH, J. M. 2019. Human Germinal Center B Cells Differ from Naïve and Memory B Cells in CD40 Expression and CD40L-Induced Signaling Response. Cytometry Part A, 95, 442-449.

LEELATIAN, N., SINNAEVE, J., MISTRY, A. M., BARONE, S. M., BROCKMAN, A. A., DIGGINS, K. E., GREENPLATE, A. R., WEAVER, K. D., THOMPSON, R. C., CHAMBLESS, L. B., MOBLEY, B. C., IHRIE, R. A. & IRISH, J. M. 2020. Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells. eLife, 9.

OKAMATO, Y., GHOSH, T., OKAMOTO, T., SCHUYLER, R. P., SEIFERT, J., CHARRY, L. L., VISSER, A., FESER, M., FLEISCHER, C., PEDRICK, C., AUGUST, J., MOSS, L., BEMIS, E. A., NORRIS, J. M., KUHN, K. A., DEMORUELLE, M. K., DEANE, K. D., GHOSH, D., HOLERS, V. M. & HSIEH, E. W. Y. 2021. Subjects at-risk for future development of rheumatoid arthritis demonstrate a PAD4-and TLR-dependent enhanced histone H3 citrullination and proinflammatory cytokine production in CD14hi monocytes. Journal of Autoimmunity, 117, 102581.

RINGEL, A. E., DRIJVERS, J. M., BAKER, G. J., CATOZZI, A., GARCÍA-CAÑAVERAS, J. C., GASSAWAY, B. M., MILLER, B. C., JUNEJA, V. R., NGUYEN, T. H., JOSHI, S., YAO, C.-H., YOON, H., SAGE, P. T., LAFLEUR, M. W., TROMBLEY, J. D., JACOBSON, C. A., MALIGA, Z., GYGI, S. P., SORGER, P. K., RABINOWITZ, J. D., SHARPE, A. H. & HAIGIS, M. C. 2020. Obesity Shapes Metabolism in the Tumor Microenvironment to Suppress Anti-Tumor Immunity. Cell, 183, 1848-1866.e26.

WOGSLAND, C. E., GREENPLATE, A. R., KOLSTAD, A., MYKLEBUST, J. H., IRISH, J. M. & HUSE, K. 2017. Mass Cytometry of Follicular Lymphoma Tumors Reveals Intrinsic Heterogeneity in Proteins Including HLA-DR and a Deficit in Nonmalignant Plasmablast and Germinal Center B-Cell Populations. Cytometry B Clin Cytom, 92, 79-87.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

Wogsland et al. apply herein mass cytometry (CyTOF) to investigate how obesity affects tumor immune infiltrates. They use several models of murine breast and pancreatic cancers and analyse their immune landscape thanks to an extended panel of 36 markers. They notably describe a decrease in CD8 T cells in one breast cancer model fed with high fat diet inducing obesity which favors tumor development.

Overall, the report is clearly written and follows a very logical plan. Figures are also clear and nicely support the text. The mass cytometry approach appears quite original and could be relevant for many readers.

Nevertheless, some concerns have to be made and would need to be acknowledged by authors:

-First of all, the paper appears very descriptive. Except at the end of the last figures, authors only establish of catalog of immune cells in different tumors. Even if the trueness of such observations is undisputable, their relevance to improve our understanding of tumor biology is clearly questionable.

-Moreover, the major finding claimed in this study (CD8 T cells decrease in tumors from HFD mice) has been very recently published paper also providing mechanistic insights (Ringel et al., Cell 2020). Authors could legitimately be disappointed but the interest of their study is sadly severely impacted by this prior publication. This key paper should be at least discussed and included in references.

-Finally, even if the initial strategy of integration of breast and pancreatic cancers was indubitably a good one, results reported in figures 5 & 6 clearly show a quite specific observation in the E0771 model. So in this context, integrating all these datasets do not improve the understanding of this phenotype

Besides these quite general comments, few more specific points: -In Fig 2, the F4/80 signal appears very weak in all datasets except one (TeLi) with an almost flat curve for all the other ones. It asks the question of the reproducibility of the staining that could be only partially corrected with batch correction algorithms.

-Obesity is clearly known to be sex/hormone dependent as confirmed by authors themselves in their Fig 1B so again the global integration (both sex and 2 organs) strategy is disputable here. It is hard to know if there is no effect in the pancreas because of tissue or sex specificities.

-On Fig 1C, red dots are closed or open but explanation of this is lacking.

-Authors use 36 markers in their CyTOF panel but use only 26 for the dimension reduction without clearly explaining this choice. Should be amended. For example, why excluding CD5?

Significance

Severly impaired by Ringel et al., Cell 2020

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

Evidence, reproducibility and clarity

Summary:

This study of Nils Halberg and colleagues aims to characterize tumor-associated immune cell infiltrates in a mouse model of diet-induced obesity. Authors compared different syngeneic tumor cell lines for mammary adenocarcinoma and pancreatic ductal adenocarcinoma. Tumor infiltrating leukocytes were analyzed by a 36-parametric mass cytometry protocol. The authors put a lot of efforts in the generation of high-quality data by applying state-of-the-art methods for sample barcoding and batch analyses, removal of batch‐specific variations and in the subsequent pipeline of data analysis. The clinical relevance of the topic addressed is well documented in several studies, showing a clear association between obesity and the development of several tumors, including those tumors investigated in this study.

Main findings of this study can be summarized that in the model system used tumor-dependent differences in the qualitative and quantitative composition of immune cell infiltrates were observed. Unfortunately, the mouse model system used obviously did not reveal convincing data whether obesity may modulate the process of tumor infiltration. The manuscript is well written, quantity of figures is appropriately and of excellent quality and prior studies were referenced appropriately. In conclusion, authors made tremendous methodological and technical efforts to generate robust and high-quality mass cytometry data, but the overall outcome of the study remains limited in respect to shedding some new light how obesity is possibly involved in the qualitative and quantitative modulation of tumor-related immune cell infiltration.

Major comments:

Due to the limited data really showing an association between obesity and immune cell infiltration of tumors investigated I would suggest that authors should change emphasis of their results more closely related to the findings of tumor-dependent immune cell infiltrations than obesity-related associations. So, the title of the study should be appropriately changed since "High dimensional immunotyping of the obese tumor micro-environment" rather implies analyses of spatial relationships of immune, tumor and fat cells by immunohistological analyses, which would indeed help to strengthen the outcome of this mass cytometry study. Although all the efforts made in mass cytometry data generation are quite commendable in this study, basic statistical issues are not clearly addressed regarding the number of biological replicates. How many mice were treated per tumor cell line? According to figure 1B nine chow and eight HFD animals were used: does this mean that only one or two mice were analyzed per cell line, respectively? Please explain how many animals belong to each of the seven mouse cohorts. Obviously, cell lines E07771 and C11 were analyzed as duplicates only. Regarding E0771, tumor growth was 31 and 23 days, respectively. So, large inter-individual differences in tumor growth were obvious and how this is reflected at the level of tumor infiltration? Therefore, please explain which criteria were used to decide when the tumors had to be removed. Furthermore, please indicate weight, viability and absolute cell number of each tumor sample in a supplementary table to get an impression about variability in tumor growth.

Minor comments:

The generation of orthotopic pancreatic cancer mouse models is technically challenging, and needs more complex imaging methods to monitor the growth of the implanted tumor cells. Furthermore, orthotopic implantation of tumor cells into the pancreas by surgery can also inflict significant physical trauma to the recipient animals. How authors have monitored tumor cell implantation? The number of CD45-positive cells per tumor sample is not given in the manuscript, but this information would be important to know, because it can be expected that most of the samples showed less than 20.000 cells. This relatively low number of total leukocytes would not allow a statistically significant profiling of rare cell subsets, such as DC's or MDSC's. This limitation should at least clearly addressed in the discussion section. According to table 2 authors have used 36 immune cell-related antigens including casp3, which was only used to exclude apoptotic cells from downstream analyses. But as written in the results section only 26 phenotyping markers were used to generate the viSNE map shown in Figure 3. In Figure 3C-F 30 markers were shown. Please explain this obvious inconsistency of markers used. How viability of tumor samples was determined? Please indicate cell loss caused by cryopreservation of dispersed tumor tissue samples. Authors state that mainly neutrophilic granulocytes will be lost during cryopreservation, and that this would help to the "definitive identification and characterization of G-MDSC". But there are several reports showing that MDSC-subsets also behave very sensitive during cryopreservation and that it is recommended to analyze fresh samples if MDSC's are of particular interest (DOI: 10.1177/1753425912463618; DOI: 10.1177/1753425912463618). This possible limitation should be discussed in the manuscript and not only highlighted as advantage on the way to identify MDSC-subsets. In the Figure 1D X-axis named by "193Ir-NA" should be replaced by "193Ir-DNA". Furthermore, please explain "(T)" in the figure legend. Percentages in the last two dotplots related to "all previous gates" are confusing: 20,44% of all DNA-containing single cells were finally intact, living CD45+ cells, i.e. almost 80% of cells were excluded because they were dead or apoptotic and this corresponds to 57,06% of intact, living CD45-positive cells related to all CD45-positive cells? How these percentages are related to the "Percent of CD45/total raw events" in the last column of Table 3 ?

Authors claimed that "155Gd_IRF4" was changed to "155Gd", but it is not clear why to mention that IRF4 has been NOT used throughout the study? Please provide only those technical details, which are necessary to understand what has been done.

Re Figure 6: please explain the abbreviation "TNBC". Experiments done with TKO mice are not described in the Materials and Methods section. In particular, it would be important to know the number of replicates and the number of tumors grown in this model. It should be also discussed that the growth kinetics of tumors in chow and HFD TKO mice seem to be much faster as compared to wild type mice. Principally, the TKO model used here is only of limited value to clarify especially the role of CD8 cells since all other T- and B- cell subsets including NK cells are also absent in this knockout model and indirect effects caused by these cells cannot be excluded.

Significance

Altogether, this study is a paragon that a single technology-based study alone, even when well-designed, is not sufficient to explore complex tumor microenvironment-immune cell interactions and that additional information on spatial relationships of cells and possibly single cell-based RNAseq techniques are necessary to shed new light on this ambitious topic. But there is no doubt that the potential of mass cytometry has been not fully exploited in this study and that a more focused view on particular cell types identified so far, such as macrophages or CD8 cells, by using as many immunophenotypic and functionally-related parameters as necessary would allow a more in depth-phenotyping of particular immune cell compartments. The significance of this subject would have been tremendously increased if human samples will be analyzed in a future confirmative study.

Even when I'm not a specialist in tumor biology, based on my expertise in the fields of chronic inflammation and cytometry, I'm convinced that the outlined way of generating immunophentypic data by single cell-based mass cytometry is of major interest not only for tumor biologists, but will be for sure recognized by a broad scientific community interested in the generation of single cell-based immunophenotypic data.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

We thank all three reviewers for their very useful and constructive comments. Below is our point-by-point response.

Reviewer #1 (Evidence, reproducibility and clarity (Required)):

The manuscript by Viais R et al describes a novel role for augmin complex in apoptosis prevention during brain development. Augmin complex recruits g TuRC to microtubule lattices to nucleate microtubule branches. The authors show how -in its absence- neural progenitors have elevated p53 activity and apoptotic rate, with severe consequences on overall brain development. In particular, augmin-deleted neural progenitors display spindle abnormalities and mitotic delay, which induce DNA damage accountable for p53-induced apoptosis.

One point that I personally found very interesting is the role of augmin-dependent MT nucleation depletion in interphase. The authors mention (line 152) that at stage E13.5, besides the number of neurons being reduced, a few neurons were misplaced in the apical region, indicating a role for augmin-driven MT nucleation in cell migration. Moreover, the authors showed that p53 genetic deletion in the Haus6 cKO rescues the apoptosis phenotype but not the tissue disorganisation, suggesting that augmin-dependent microtubule might play a role in tissue polarity. While this is well presented in the discussion, the title in line 268 narrowly refers to mitotic augmin roles. I would like here to see the authors referring to putative roles for augmin-mediated MT nucleation in interphase, by toning down the title in line 268.

We note that severe loss of tissue integrity is evident in the p53 KO background. In this background cells are allowed to repeatedly undergo defective cell divisions with aberrant chromosome segregation, producing increasingly abnormal daughter cells that may eventually fail to support epithelial integrity. Regarding possible neuronal migration defects, this has been previously observed in a study by the Hoogenraad group (Cunha-Ferreira et al., Cell Reports, 2018, 24, 791–800) and this is mentioned in our discussion. To account for the possibility that augmin may have roles beyond mitosis, we have changed the heading to a more neutral statement, not specifically referring to proliferation/mitosis: “Loss of augmin in p53 KO brains disrupts neuroepithelium integrity”.

Overall, the text is well written and flows easily. Figures are clear and legends provide sufficient information on experimental conditions, number of replicates and scale bars. I noticed that, although the number of repeats is specified, the number of cells scored per experiment is not always included. In my comments below I highlight cases where this missing information should be added.

**Specific points:**

1. In the Cep63 KO (Marjanovic et al, 2015) and the CenpJ KO mice (Insolera et al, 2014), as well as other recently published papers (e.g. Phan TP et al, EMBO Journal, 2020) part of the phenotypical characterisation of the KO mice displays pictures of the overall brain dissected from the mice. Could the author show these images?

The main difference between the cited studies (including our own on the role of CEP63 in brain development) and our current study is that in the previous studies brains are microcephalic but essentially intact, whereas in our current study brain development was aborted and accompanied by cell death and severe tissue disruption. As a result, these brains are very fragile and difficult/impossible to isolate. An additional challenge is the fact that brain disruption occurs at a very early developmental stage (before E13.5), where dissection is more difficult than at later stages. Indeed, we note that all the brains presented in the above cited studies were from later embryonic stages or newborn/adult mice. Therefore, instead of dissecting brains, we decided to present encephalic coronal and sagittal sections as shown in Fig. 1c, d, e, Fig. S1c, and Fig. 3b, e to show the overall impact of Haus6 cKO and Haus6 cKO p53 KO on embryonic brain morphology at E13.5 and E17.5.

Fig2d: do the insets correspond to higher magnification images? What is the zoom factor? I could not find it in the legend.

The zoom factor is 1.4 - we have added this information to the figure legend.

Fig2E,I and K graphs: how many cells were quantified here over how many experiments? I could not find information in the figure legend.

We have added the information regarding the number of embryos and counted cells to the figure legends.

The impact of Haus6 on mitotic spindle needs further clarification:

o Fig2F: here, the authors show quantification for abnormal and multipolar spindle together. Later on, the abnormal spindle phenotype is no longer discussed (Fig4). I was wondering what is the individual contribution of abnormal and multipolar spindle, separately. Which one of the two is more frequent? Could the authors explain in the text how they define an abnormal spindle? Is it the lack of MT with the condensed chromosome area?

We agree that our previous classification was somewhat confusing. The spindle defects in Haus6 cKO cells are directly linked to the spindle pole fragmentation phenotype shown in Fig. 2d, e. Association of spindle microtubules with these scattered PCM fragments causes spindles to appear overall disorganized. In some cases, multiple smaller asters are present, which is what we had termed “multipolar”. However, this does not always involve multipolar DNA configurations, which we separately quantify in Fig. 4. To avoid confusion, we now classify spindle morphologies based on tubulin staining simply as “normal” (bipolar configuration, two robust and focused asters) or “disorganized” (lack of bipolar configuration, in some cases multiple smaller asters). We have included a better description of this classification (lines 202-205).

o Could it be that augmin deletion induce an instability in MTs within the mitotic spindle, leading to the "empty" or with very few MTs spindles? Or could it be that more cold-sensitive MTs are affected by fixation? What is the percentage of the spindle with no MT in control?

It is possible that augmin-deficient spindles are less well-preserved during fixation due to compromised spindle microtubule stability. Indeed, in tissue culture cells augmin deficient spindle microtubules are more cold-sensitive than controls (Zhu et al., 2008, JCB, 183, 835-848). To address this we will determine the percentage of mitotic control and Haus6 cKO cells lacking microtubule staining.

o Did the authors quantify anaphase/telophase phenotypes as they did in Fig4f?

Yes, this quantification was already included in Fig. 4j, where we compared abnormal chromosome configurations between Haus6 cKO and Haus6 cKO p53 KO.

o How do authors explain PCM fragmentation here? Could this phenotype be due to an initial cytokinesis defect which led the cells to accumulate extra centrosomes? Or could this maybe be a product of aberrant PCM maturation/centrosome duplication? Could the authors add here a line to discuss the possible origin of pole fragmentation?

The PCM fragmentation phenotype has previously been described after augmin RNAi in cultured cells (Lawo et al., 2009, Curr Biol, 19, 816-826). We refer to this result in the discussion and we have added the above reference, to emphasize this point. The authors showed that this phenotype does not involve amplification of centriole number, but is caused by an imbalance in microtubule-dependent forces acting on the PCM and leading to its fragmentation. Thus, the extra poles were formed by acentriolar PCM fragments. We will clarify this issue by quantifying centriole numbers in mitotic cells (when centriole duplication is complete) in control and Haus6 cKO brains. We expect that this will confirm the data previously obtained in cell lines showing that in most cells the fragmented poles are not due to extra centrioles (see also below).

Apart from the PCM fragmentation phenotype that does not alter centriole number, previous work in cultured cells also described cytokinesis defects. Failed cytokinesis would indeed lead to increased centriole number. However, it would also increase DNA content, which would be visible by an increase in the size of interphase nuclei (which we observed in Haus6 cKO p53 KO cells and quantified in Fig. 4J) and a larger size of mitotic figures. We now refer to the possibility of cytokinesis defects and cite previous work in lines 272-274. In case we observe cells with increased centriole number, which we will quantify for the revised version of the manuscript (see above), we will also determine if this corelates with an increased size of the corresponding mitotic figures. If so, this would be consistent with failed cytokinesis as cause of extra centrosomes.

Fig 4 Did the authors quantify centrosome fragmentation and abnormal spindle here? As they characterised them for the Haus6 cKO mouse, it would be preferable to maintain the same characterisation for the Haus6 cKO p53KO.

We will quantify pole fragmentation and spindle defects also in Haus6 cKO p53 KO as shown for Haus6 cKO in Fig. 2.

Fig4c and d: how many replicates were done to obtain these graphs? I think the authors forgot to add this information in the figure legend.

This information has been included in the figure legend.

Fig4f,g, I and J: how many cells were counted per experiment? I appreciate the authors writing the n of experiments performed.

We have added this information to the figure legend.

Fig5d: how many cells were counted per experiment?

We have added this information to the figure legend.

Reviewer #1 (Significance (Required)):

While it was already known that mitotic delay affects the neuronal progenitor pool through activation of p53-dependent apoptosis (Pilaz L-J, Neuron 2016; Mitchell-Dick A, Dev Neurosci 2020), and that this can be triggered by depletion of centrosomal proteins as Cenpj and Cep63, the role of surface-dependent microtubule nucleation was not identified so far. Some insights come from a Haus6-KO mouse model which dies during blastocyst stage after several aberrant mitosis (Watanabe S, Cell Reports, 2016). In parallel, McKinley KL et al showed that Haus8 depletion in human cells (RPE1cells) triggered p53-dependent G1 arrest following mitotic defects (McKinley KL, Developmental Cell, 2017). Building on the Hause6 KO mouse and human cell line data, here Viais R et al discover a novel role for the augmin-mediated MT nucleation in neural progenitor growth and brain development in vivo, through prevention of p53-induced apoptosis.

Specifically, Viais R et al show that:

1. Surface-dependent microtubule nucleation depletion severely impacts brain development, disrupting partly or completely forebrain domains and cerebellum;
2. Surface-dependent microtubule nucleation depletion induce spindle abnormalities, resulting in mitotic delay in apical progenitors;

3. Mitotic delay results in DNA breaks, p53 activation and p53-induced apoptosis.

   This is a tidy, well-executed study with good quality data. These findings propose a novel mechanism that results essential for neural progenitor and overall brain development.

   In my opinion, a large audience will benefit from these discoveries: from developmental biologists to cell biologists focused on microtubule dynamics, cell cycle, differentiation, stem cells and cell polarity.

Key works describing my area of expertise: microtubule dynamics, centrosome function, cell cycle regulation and cell polarity.

Reviewer #2 (Evidence, reproducibility and clarity (Required)):

Viais, Lüders and colleagues here present an analysis of augmin's roles in neural stem cell development. They describe a dramatic impact of the conditional ablation of Haus6 on embryonic brain development in the mouse, with mitotic problems that lead to greatly-increased levels of apoptosis. The rescue of this apoptosis by mutation of the gene that encodes p53 did not restore brain development, which was still aberrant, due to mitotic errors.

The paper is clearly written, with well-designed and controlled experiments. Its conclusions are well supported by the data presented. I have few comments on the technical aspects of the work- it appears very solid to me.

**Specific comments**

1. Clearer explanation of the mouse strains used should be provided. The section describing the generation of the Haus6 conditional on p.5 should specify that this is the same as was already published in the 2016 Watanabe paper (this is in the Materials and Methods, but this should be more clearly specified. More specific details of the p53 knockout mice from Jackson should be included in the Materials and Methods.

We have included additional information describing the generation of the Haus6 cKO mice in the main text (line 137-140). It is not exactly the same as described in the Watanabe et al. paper. The previously published strain (Watanabe et al., 2016, Cell Reports, 15, 54-60) contained a floxed Haus6 cKO allele with a flanking neomycin cassette. For the current study the neomycin cassette was removed. Details are described in the method section and also shown in Fig. S1a. Specific information regarding the p53 KO strain has been added to the method section.

Figure 1a contains minimal information on the Haus6 locus. More detail should be included for information, if this Figure is to remain (although reference to the targeting details in the original description would be sufficient). It is unclear what the timeline diagram is to convey and it should be improved or deleted. A similar comment applies for the details in Figure 3a, although the colour scheme for the different genotypes is useful.

More detailed information on the Haus6 locus is shown in the schematic of Fig. S1a and in the referenced study (Watanabe et al., 2016, Cell Reports, 15, 54-60). Since the targeting of Haus6 exon1 was previously described, we feel that including this information as a supplementary figure and referring to the previous study is appropriate.

Regarding the schematics in Fig. 1a and Fig. 3a, we have improved these. The timeline shows the time points of Cre expression and of obtaining embryos for analysis.

The important PCR controls in Figure S1b have an unexplained 1000 bp band that appears only in the floxed heterozygote. It would be helpful if the authors explained this in the relevant Figure legend.

This band is an artifact and represents heteroduplexes of floxed (1080 bp) and wild type (530 bp) DNA strands due to extended regions of complementary. We have explained this in the figure legend.

Assuming the putative centrosome 'clusters' in Figure 6c are similar to the fragmented structures seen in thalamus in Figure 2d, a different description should be used to avoid confusion with multiple centrosomes, which is not a phenotype here. It is not clear how the loss of centrosomes from the ventricular surface was scored, whether it was based on total gamma-tubulin signal or individual centrosomes; how fragmented poles would affect that is unclear, so the legend and relevant details should clarify this point.

The fragmented spindle poles shown in Fig. 2d are different from the centrosome clusters in Fig. 6c. The fragmented poles are fragments of PCM rather than extra centrosomes. Fragmentation is specific to mitosis, involving forces exerted by spindle microtubules (Lawo et al., 2009, Curr Biol, 19, 816-826). In contrast, the centrosome clusters that we observed in Haus6 cKO p53 KO apical progenitors represent centrosomes from multiple cells in interphase, most likely as part of apical membrane patches that have delaminated form the ventricular surface. In the intact epithelium of controls these centrosomes line the ventricular surface. To avoid confusion, we now indicate in the text and legend that these centrosome clusters involve interphase cells.

Phospho-histone H2AX should be referred to as a marker of activation of the DNA damage response, rather than DNA repair.

We have changed the text accordingly.

**Minor points**

i. Figure 1b should include a scale bar.

We have added the scale bar.

ii. The labelling of Figure 1f should be revised.

The labels have been fixed.

iii. Figure 2k is not labelled in this Figure.

This has been fixed.

iv. Scale bars should be included in the blow-ups in Figure 6c.

We have added the scale bars.

Reviewer #2 (Significance (Required)):

While it is striking that they see complete disruption of brain development, rather than microcephaly, arguably the mechanistic novelty of the findings is moderate, in that the impacts of Haus6 deficiency on mitotic spindle assembly are well established. The authors only allude to potential additional and novel activities of augmin (in neural progenitors, potentially) that might explain this possibly-unexpected outcome of this study. The topic is likely to be of interest to people in the field of mitosis, genome stability and brain development.

My expertise is cell biology/ mitosis, less so on murine brain development.

Reviewer #3 (Evidence, reproducibility and clarity (Required)):

Jens Lüders &Co demonstrates the essential role of Augmin-mediated MT is critical for proper brain development in mice. The most striking point is that even p53 is eliminated, the microcephaly phenotypes of Haus6 KOs were not rescued. This could mean that the Augmin-mediated MT process is critical to cellular functions that are independent of p53. The authors claim that there are increased DNA damage and excessive mitotic errors. In these aspects, the current work is fascinating. Nevertheless, what causes massive damage to the neural epithelial tissues in the double mutant is not well explained or examined. Few questions appear in mind before I go into the detail. Are these animals still harbor functional centrosomes and their numerical status?

This is an important point that was also raised by the other reviewers. Based on previous work in cells lines (Lawo et al., 2009, Curr Biol, 19, 816-826), we do not expect that loss of augmin directly impairs centrosomes. Indeed, the authors showed that centriole number was unaffected. The only centrosome defect that the authors observed was fragmentation of the PCM during mitosis, but this was shown to be due to imbalanced forces exerted by spindle microtubules: fragmentation could be rescued by microtubule depolymerization or depletion of the cortical microtubule tethering factor NUMA (Lawo et al., 2009, Curr Biol, 19, 816-826). That being said, we will examine this issue also in our mouse model by staining and counting of centrioles in mitotic apical progenitors of control and Haus6 cKO embryos.

The microcephaly part of the introduction needs some more work. In particular, the authors need to explain apical progenitors' depletion, possibly the correct mechanisms in causing microcephaly. By saying cortical progenitors, it becomes vague. Indeed, there would also be cortical progenitors depleted. But, the fundamental mechanisms are the depletion of apical progenitors lined up at VZ's lumen. Two works in this connection generated brain tissues from microcephaly patients carrying mutations in CenpJ and CDK5RAP2 (Gabriel and Lancaster et al). Authors should cite their work and relate their findings to mouse brain data.

We have introduced text changes in the introduction to indicate the specific role of apical progenitor depletion in microcephaly and the differences in the underlying mechanism between mouse and human organoid models (line 63; lines 86-92). In this context we also cite the Gabriel et al. and Lancaster et al. studies.

-What makes me worry is, looking at figure 1E, there is pretty much no brain, and of course, authors have analyzed what is left over. How could one distinguish reduced PAX6 area and TUJ1 area is due to the gross defects in brain development. Clearly, Haus6 KO causes a severe defect in brain development. Thus, deriving a conclusion from the damaged brain can be misleading. One way to circumvent this problem is to perform 2D experiments with isolated cell types (let us say NPCs and testing if they can spontaneous differentiate).

We note that overall brain structures are only lost by E17.5, but brain structures (albeit defective) are still present at E13.5. Indeed, all of our quantifications were done at E13.5 or earlier stages. That being said, we understand the concern that quantifications in defective brain structures may be misleading. However, 2D cultures, for which cells are removed from their tissue context, may have similar issues. For this reason, we plan to provide two different type of analyses. We will measure PAX6 and TUJ1 layers in brains from embryos at E.11.5, since the relevant tissues will be less damaged at this earlier stage. In addition, we will use BrdU injection prior to fixation of embryos. Proliferating apical progenitors will incorporate the label during S phase and subsequently we will determine the relative amounts of BrdU-positive cell types (apical progenitors vs neurons) in control and Haus6 cKO brains. Tissue damage will have less impact in this short-term labelling experiment.

Figure 2: A nice illustration that Hau6 KO animals harbor many mitotic figures. The quantifications lack how many slices and how many cells were analyzed. Simply n=4 does not say much. 4 animals were considered but how many cells/slices would help identify mitotic cells/animals' distribution. A simple bar diagram does not tell a lot.

We have added this information to the figure legend.

As a minor point, how did the authors unambiguously scored prometaphase cells and other mitotic figures? Representative figures will help. Besides, what is the meaning of many prometaphase cells? At least a discussion would help.

This is a good suggestion and we will provide examples of the mitotic figures that we scored. We now explain the meaning of the increase in prometaphase cells in the description of this result (lines 178-180).

Can the authors probe centrosomes (not by using gamma-tubulin) and relate their presence or absence to p53 upregulation? This is an important point because a complete loss of centrosome is known to trigger p53 upregulation. This may be different in Haus6 KO. This could mean (i.e, centrosomes are normal in numbers or increase in numbers), p53 upregulation is regardless of centrosomes loss.

Indeed, we believe that p53 upregulation in Haus6-deficient brains is not caused by loss of centrosomes. Instead, our data suggest, as explained in the discussion, that mitotic delay caused by augmin deficiency is sufficient for p53 upregulation. We will further support this conclusion by counting centrioles in mitotic cells. At this point of the cell cycle centriole duplication is complete and we expect to observe largely normal centriole numbers. In some cells we may observe increased numbers due to cytokinesis failure (see response to reviewer #1).

I have a hard time to ascertain how the authors scored interphase cells that enriched with p53. Some representative images with identity markers will help.

Scoring p53-positive interphase cells is relatively straightforward since the p53 signal is nuclear and not observed in mitotic apical progenitors. We have included a magnified region of the tissue shown in Fig. 2j, displaying PAX6/p53-positive nuclei of individual cells.

Looking at the p53 status in Haus6 KO animals, it is intriguing that p53 upregulation is not unique to centrosome loss. At this point, it becomes essential to thoroughly analyze the centrosome status to cross-check if Haus6 loss abrogates centrosomes; if so, how much.

Since centrosome number is linked to centriole number, we will address this point by quantifying centriole numbers in mitotic apical progenitors (see above).

Double KO could subside the cell death, but not tissue growth is impressive. So what is going on there? Is there a premature differentiation that leads to NPCs depletion? I believe the authors should generate 2D experiments with cells derived from these double KO animals compared to Haus6 KO and test if there is a premature differentiation that can lead to malformation of the forebrain. Here staining for the forebrain progenitor markers will additionally help (Perhaps FOXG1).

As explained in response to reviewer #1, we prefer to analyse this issue in vivo rather than in cells that are removed from their native tissue context, which may affect cell fate decisions. To address whether cells prematurely differentiate, we will use injection of BrdU (incorporated by proliferating apical progenitors) prior to fixation, followed by staining for cell type-specific markers. If there is premature differentiation, this should be visible as an increase in BrdU-positive post-mitotic cells.

Looking at Figure 6, it becomes clear that the double KOs have severe issues in maintaining the apical progenitors suggesting that they undergo premature differentiation before attaining a sufficient pool of NPCs. Testing this will bridge the paper between descriptive findings to mechanisms.

This point relates to the reviewer’s previous point: do Haus6 cKO p53 KO apical progenitors prematurely differentiate? We believe that cell loss, tissue disruption, and aborted development may also be explained without premature differentiation. In the absence of p53, repeated abnormal mitoses and the resulting increasingly severe chromosomal aberrations including DNA damage (Fig. 5) may produce cells that eventually won’t be able to proliferate and function properly. However, we will test premature differentiation by BrdU injection and staining with appropriate markers as explained above.

The discussion section is excellent, but it should add some human relevance. In particular, are there p53 dependent cell deaths that have been described in human tissues. In my opinion, it seems specific in the mouse brain. The discussion can also have statements about why the human brain is so sensitive even for mild mutations. I am not sure if those human mutations can cause similar defects in the mouse brain. Most of the mice based studies have been focusing on eliminating complete genes of interest.

We have included a section in the discussion to relate our findings to human brain development and the differences with results obtained in mouse models regarding the role of apoptosis (lines 386-391).

Reviewer #3 (Significance (Required)):

Overall, this is a very well done work but requires some more experiments for mechanisms understanding. Addressing those will make the paper fit to get published.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

Jens Lüders &Co demonstrates the essential role of Augmin-mediated MT is critical for proper brain development in mice. The most striking point is that even p53 is eliminated, the microcephaly phenotypes of Haus6 KOs were not rescued. This could mean that the Augmin-mediated MT process is critical to cellular functions that are independent of p53. The authors claim that there are increased DNA damage and excessive mitotic errors. In these aspects, the current work is fascinating. Nevertheless, what causes massive damage to the neural epithelial tissues in the double mutant is not well explained or examined. Few questions appear in mind before I go into the detail. Are these animals still harbor functional centrosomes and their numerical status? The microcephaly part of the introduction needs some more work. In particular, the authors need to explain apical progenitors' depletion, possibly the correct mechanisms in causing microcephaly. By saying cortical progenitors, it becomes vague. Indeed, there would also be cortical progenitors depleted. But, the fundamental mechanisms are the depletion of apical progenitors lined up at VZ's lumen. Two works in this connection generated brain tissues from microcephaly patients carrying mutations in CenpJ and CDK5RAP2 (Gabriel and Lancaster et al). Authors should cite their work and relate their findings to mouse brain data.

-What makes me worry is, looking at figure 1E, there is pretty much no brain, and of course, authors have analyzed what is left over. How could one distinguish reduced PAX6 area and TUJ1 area is due to the gross defects in brain development. Clearly, Haus6 KO causes a severe defect in brain development. Thus, deriving a conclusion from the damaged brain can be misleading. One way to circumvent this problem is to perform 2D experiments with isolated cell types (let us say NPCs and testing if they can spontaneous differentiate)

Figure 2: A nice illustration that Hau6 KO animals harbor many mitotic figures. The quantifications lack how many slices and how many cells were analyzed. Simply n=4 does not say much. 4 animals were considered but how many cells/slices would help identify mitotic cells/animals' distribution. A simple bar diagram does not tell a lot.

As a minor point, how did the authors unambiguously scored prometaphase cells and other mitotic figures? Representative figures will help. Besides, what is the meaning of many prometaphase cells? At least a discussion would help.

Can the authors probe centrosomes (not by using gamma-tubulin) and relate their presence or absence to p53 upregulation? This is an important point because a complete loss of centrosome is known to trigger p53 upregulation. This may be different in Haus6 KO. This could mean (i.e, centrosomes are normal in numbers or increase in numbers), p53 upregulation is regardless of centrosomes loss.

I have a hard time to ascertain how the authors scored interphase cells that enriched with p53. Some representative images with identity markers will help.

Looking at the p53 status in Haus6 KO animals, it is intriguing that p53 upregulation is not unique to centrosome loss. At this point, it becomes essential to thoroughly analyze the centrosome status to cross-check if Haus6 loss abrogates centrosomes; if so, how much.

Double KO could subside the cell death, but not tissue growth is impressive. So what is going on there? Is there a premature differentiation that leads to NPCs depletion? I believe the authors should generate 2D experiments with cells derived from these double KO animals compared to Haus6 KO and test if there is a premature differentiation that can lead to malformation of the forebrain. Here staining for the forebrain progenitor markers will additionally help (Perhaps FOXG1).

Looking at Figure 6, it becomes clear that the double KOs have severe issues in maintaining the apical progenitors suggesting that they undergo premature differentiation before attaining a sufficient pool of NPCs. Testing this will bridge the paper between descriptive findings to mechanisms.

The discussion section is excellent, but it should add some human relevance. In particular, are there p53 dependent cell deaths that have been described in human tissues. In my opinion, it seems specific in the mouse brain. The discussion can also have statements about why the human brain is so sensitive even for mild mutations. I am not sure if those human mutations can cause similar defects in the mouse brain. Most of the mice based studies have been focusing on eliminating complete genes of interest.

Significance

Overall, this is a very well done work but requires some more experiments for mechanisms understanding. Addressing those will make the paper fit to get published.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

Viais, Lüders and colleagues here present an analysis of augmin's roles in neural stem cell development. They describe a dramatic impact of the conditional ablation of Haus6 on embryonic brain development in the mouse, with mitotic problems that lead to greatly-increased levels of apoptosis. The rescue of this apoptosis by mutation of the gene that encodes p53 did not restore brain development, which was still aberrant, due to mitotic errors.

The paper is clearly written, with well-designed and controlled experiments. Its conclusions are well supported by the data presented. I have few comments on the technical aspects of the work- it appears very solid to me.

Specific comments

1. Clearer explanation of the mouse strains used should be provided. The section describing the generation of the Haus6 conditional on p.5 should specify that this is the same as was already published in the 2016 Watanabe paper (this is in the Materials and Methods, but this should be more clearly specified. More specific details of the p53 knockout mice from Jackson should be included in the Materials and Methods.
2. Figure 1a contains minimal information on the Haus6 locus. More detail should be included for information, if this Figure is to remain (although reference to the targeting details in the original description would be sufficient). It is unclear what the timeline diagram is to convey and it should be improved or deleted. A similar comment applies for the details in Figure 3a, although the colour scheme for the different genotypes is useful.
3. The important PCR controls in Figure S1b have an unexplained 1000 bp band that appears only in the floxed heterozygote. It would be helpful if the authors explained this in the relevant Figure legend.
4. Assuming the putative centrosome 'clusters' in Figure 6c are similar to the fragmented structures seen in thalamus in Figure 2d, a different description should be used to avoid confusion with multiple centrosomes, which is not a phenotype here. It is not clear how the loss of centrosomes from the ventricular surface was scored, whether it was based on total gamma-tubulin signal or individual centrosomes; how fragmented poles would affect that is unclear, so the legend and relevant details should clarify this point.
5. Phospho-histone H2AX should be referred to as a marker of activation of the DNA damage response, rather than DNA repair.

Minor points

i. Figure 1b should include a scale bar.<br> ii. The labelling of Figure 1f should be revised. iii. Figure 2k is not labelled in this Figure. iv. Scale bars should be included in the blow-ups in Figure 6c.

Significance

While it is striking that they see complete disruption of brain development, rather than microcephaly, arguably the mechanistic novelty of the findings is moderate, in that the impacts of Haus6 deficiency on mitotic spindle assembly are well established. The authors only allude to potential additional and novel activities of augmin (in neural progenitors, potentially) that might explain this possibly-unexpected outcome of this study.

The topic is likely to be of interest to people in the field of mitosis, genome stability and brain development.

My expertise is cell biology/ mitosis, less so on murine brain development.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

Evidence, reproducibility and clarity

The manuscript by Viais R et al describes a novel role for augmin complex in apoptosis prevention during brain development. Augmin complex recruits g TuRC to microtubule lattices to nucleate microtubule branches. The authors show how -in its absence- neural progenitors have elevated p53 activity and apoptotic rate, with severe consequences on overall brain development. In particular, augmin-deleted neural progenitors display spindle abnormalities and mitotic delay, which induce DNA damage accountable for p53-induced apoptosis.

One point that I personally found very interesting is the role of augmin-dependent MT nucleation depletion in interphase. The authors mention (line 152) that at stage E13.5, besides the number of neurons being reduced, a few neurons were misplaced in the apical region, indicating a role for augmin-driven MT nucleation in cell migration. Moreover, the authors showed that p53 genetic deletion in the Haus6 cKO rescues the apoptosis phenotype but not the tissue disorganisation, suggesting that augmin-dependent microtubule might play a role in tissue polarity. While this is well presented in the discussion, the title in line 268 narrowly refers to mitotic augmin roles. I would like here to see the authors referring to putative roles for augmin-mediated MT nucleation in interphase, by toning down the title in line 268.

Overall, the text is well written and flows easily. Figures are clear and legends provide sufficient information on experimental conditions, number of replicates and scale bars. I noticed that, although the number of repeats is specified, the number of cells scored per experiment is not always included. In my comments below I highlight cases where this missing information should be added.

Specific points:

1. In the Cep63 KO (Marjanovic et al, 2015) and the CenpJ KO mice (Insolera et al, 2014), as well as other recently published papers (e.g. Phan TP et al, EMBO Journal, 2020) part of the phenotypical characterisation of the KO mice displays pictures of the overall brain dissected from the mice. Could the author show these images?
2. Fig2d: do the insets correspond to higher magnification images? What is the zoom factor? I could not find it in the legend.
3. Fig2E,I and K graphs: how many cells were quantified here over how many experiments? I could not find information in the figure legend.
4. The impact of Haus6 on mitotic spindle needs further clarification:

o Fig2F: here, the authors show quantification for abnormal and multipolar spindle together. Later on, the abnormal spindle phenotype is no longer discussed (Fig4). I was wondering what is the individual contribution of abnormal and multipolar spindle, separately. Which one of the two is more frequent? Could the authors explain in the text how they define an abnormal spindle? Is it the lack of MT with the condensed chromosome area?

o Could it be that augmin deletion induce an instability in MTs within the mitotic spindle, leading to the "empty" or with very few MTs spindles? Or could it be that more cold-sensitive MTs are affected by fixation? What is the percentage of the spindle with no MT in control?

o Did the authors quantify anaphase/telophase phenotypes as they did in Fig4f?

o How do authors explain PCM fragmentation here? Could this phenotype be due to an initial cytokinesis defect which led the cells to accumulate extra centrosomes? Or could this maybe be a product of aberrant PCM maturation/centrosome duplication? Could the authors add here a line to discuss the possible origin of pole fragmentation?

1. Fig 4 Did the authors quantify centrosome fragmentation and abnormal spindle here? As they characterised them for the Haus6 cKO mouse, it would be preferable to maintain the same characterisation for the Haus6 cKO p53KO.
2. Fig4c and d: how many replicates were done to obtain these graphs? I think the authors forgot to add this information in the figure legend.
3. Fig4f,g, I and J: how many cells were counted per experiment? I appreciate the authors writing the n of experiments performed.
4. Fig5d: how many cells were counted per experiment?

Significance

While it was already known that mitotic delay affects the neuronal progenitor pool through activation of p53-dependent apoptosis (Pilaz L-J, Neuron 2016; Mitchell-Dick A, Dev Neurosci 2020), and that this can be triggered by depletion of centrosomal proteins as Cenpj and Cep63, the role of surface-dependent microtubule nucleation was not identified so far. Some insights come from a Haus6-KO mouse model which dies during blastocyst stage after several aberrant mitosis (Watanabe S, Cell Reports, 2016). In parallel, McKinley KL et al showed that Haus8 depletion in human cells (RPE1cells) triggered p53-dependent G1 arrest following mitotic defects (McKinley KL, Developmental Cell, 2017). Building on the Hause6 KO mouse and human cell line data, here Viais R et al discover a novel role for the augmin-mediated MT nucleation in neural progenitor growth and brain development in vivo, through prevention of p53-induced apoptosis.

Specifically, Viais R et al show that:

1. Surface-dependent microtubule nucleation depletion severely impacts brain development, disrupting partly or completely forebrain domains and cerebellum;
2. Surface-dependent microtubule nucleation depletion induce spindle abnormalities, resulting in mitotic delay in apical progenitors;
3. Mitotic delay results in DNA breaks, p53 activation and p53-induced apoptosis.

This is a tidy, well-executed study with good quality data. These findings propose a novel mechanism that results essential for neural progenitor and overall brain development.

In my opinion, a large audience will benefit from these discoveries: from developmental biologists to cell biologists focused on microtubule dynamics, cell cycle, differentiation, stem cells and cell polarity.

Key works describing my area of expertise: microtubule dynamics, centrosome function, cell cycle regulation and cell polarity.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Response to reviewers: Woodcock et al. 2021

Reviewer 1 (Evidence, reproducibility, and clarity):

Summary The authors resolved the biosynthesis of trehalose and alpha-glucan in Pseudomonas aeruginosa and the role of these two compounds in osmotic and desiccation stress.

We thank the reviewer for their positive review of our manuscript. Our responses to their specific queries are interspersed below.

Major comments:

• Are the key conclusions convincing? Yes

• Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? No

• Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation. Not necessary, comprehensive coverage of research topic.

• • Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments. * Not applicable

• Are the data and the methods presented in such a way that they can be reproduced? Yes

• • Are the experiments adequately replicated and statistical analysis adequate? * Yes, everything is adequate but just one subtle concern: check the significance of the number of digits in the entries listed in Table S3. Revise Table S3.

Table S3 has been revised as requested. The data in this table is now presented correct to one decimal place.

Minor comments:

• Specific experimental issues that are easily addressable. Not applicable (Table S3: see above)

• • Are prior studies referenced appropriately? No. Refs. 18- 32: The subjects of 'trehalose' and 'osmotic stress' have already been addressed in the Pseudomonas field and should be referenced. The authors cite work carried out on trehalose and osmotic stress on phylogenetically distant microorganisms, but do not cite related work from the Pseudomonas field which I consider to be inappropriate. Similarly, trehalose biosynthesis in Pseudomonas* has not only been covered by refs. 47 and 48.

This is a fair comment. The focus of our introduction came from a desire to concentrate specifically on the metabolism and intracellular function of trehalose/α-glucan in Pseudomonas. In hindsight, we acknowledge that our introduction is a little too narrowly focussed. We have expanded the introduction and discussion sections to include additional discussion of trehalose in Pseudomonas and its regulation in the CF lung.

• Are the text and figures clear and accurate? Extremely well written manuscript and prepared figures

• • Do you have suggestions that would help the authors improve the presentation of their data and conclusions? Revise the list of references and discuss more thoroughly your novel findings in the light of existing knowledge in the Pseudomonas* field.

    Please see previous comment relating to the literature.

Reviewer 1 (Significance):

Significance

• • Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field. Conceptual advance: The authors identified and characterized the enzymatic pathway of trehalose and alpha-glucan biosynthesis in Pseudomonas aeruginosa and its role to cope with osmotic and desiccation stress. The authors' conclusions do not correspond with recently published peers' work; hence they should discuss in more detail why they consider their data to be more accurate to discern the role of trehalose to contain desiccation and osmotic stress in P. aeruginosa*.

    Please see previous comment relating to the literature. In general, the published work to date on trehalose in Pseudomonas spp. does not consider GlgE pathway-mediated link to α-glucan that we characterise in this paper. Our work demonstrates that synthesis and metabolism of the two molecules are implicitly linked in species where the GlgE pathway is present, and they cannot be considered in isolation. For this reason we are very confident that our study represents the most accurate model to date for trehalose and α-glucan metabolism and their associated phenotypes in P. aeruginosa. We have therefore emphasised that the role of trehalose in Pseudomonas spp. should be re-evaluated in light of our findings.

• • Place the work in the context of the existing literature (provide references, where appropriate). Existing literature focusing on trehalose, osmotic stress, desiccation stress in the Pseudomonas *field not cited by the authors:

• These papers are of variable scientific quality, but the conceptual work by Hallsworth and the work by Behrens on the PA metabolome in CF lungs are worth discussing. All other work provides pieces of information on function and biosynthesis of trehalose up to now known by the Pseudomonas community. The authors resolved the function of the GlgA operon which will be definitely appreciated.

  We thank the reviewer for these helpful suggestions. We have reviewed these papers carefully and have incorporated several, including the papers from Hallsworth and Behrens into the revised manuscript.

Strengths of the manuscript:

• Meticulously planned and carefully executed experiments, not a single experimental flaw
• Very high technical quality of experiments and primary data
• Comprehensive coverage of the research topic

• Excellent presentation in text and illustrations Only weakness:

• Insufficient consideration of peers' published work on trehalose and its role in stress response in P. aeruginosa

  Please see previous comment relating to the literature.

• • State what audience might be interested in and influenced by the reported findings. Scientists working in the fields of glycoconjugate and carbohydrate research, biochemists, microbiologists with interest in metabolic pathways, stress response and/or Pseudomonas. __Reviewer #2 (Evidence, reproducibility, and clarity):*__

It will be difficult for me to write a review of this paper and for the authors to make sense of my review because the manuscript's pages / lines are not numbered…

We apologise to the reviewer for this oversight.

Summary The authors carried out a comprehensive characterization of the metabolism of trehalose in Pseudomonas aeruginosa PA01, using techniques of biochemistry, reverse genetics, and bioinformatics. The main findings include that the disaccharide trehalose is synthesized in this organism from branched chain α-glucans and that the catabolism of trehalose proceeds via another disaccharide, maltose and is fed back into the synthesis of α-glucans. Trehalose and α-glucans have been implicated in conferring resistance to abiotic stresses in other organisms. The authors show that mutants that are blocked in the synthesis of trehalose are sensitive to high salinity but are normal with respect to their sensitivity to desiccation, whereas mutants impaired in the accumulation of α-glucans are sensitive to desiccation without being unduly sensitive to osmotic stress. These results indicate that trehalose and α-glucans have different roles in abiotic stress-tolerance.

Major points

This manuscript describes an impressive amount of careful work and presents new insights into the metabolism of trehalose, maltose, and α-glucans. However, the authors should address the following major comments before the paper is accepted.

We thank the reviewer for their thorough and positive assessment of the manuscript. We address their specific points below.

• Discussion: the authors state that "trehalose protects Pseudomonas ssp. against osmotic stress, most likely due to its role as a compatible solute." According to Table 2, P. aeruginosa grown in the medium of low osmolarity accumulated 0.13% trehalose per gram dry weight, i.e. ~4 μmol / g dry weight. Assuming that the dry weight / wet weight ratio of P. aeruginosa is the same as that of P. putida, which is ~1/3 (PMID: 6508285), the concentration of trehalose in the cells calculates to be ~2 mM. It is not plausible that trehalose could be significant as compatible solute at this low concentration.

• One way out could be if the accumulation of this disaccharide were increased by osmotic stress. The authors should also measure the trehalose content of cells grown in medium containing 0.85 M NaCl. In case of positive results in this experiment, it would be interesting to determine the effects of osmotic stress on the levels of trehalose biosynthetic and catabolic enzymes, but this would not be necessary for the acceptance of the paper.

  This is a fair point. To address this, we measured the trehalose and maltose-1-phosphate levels for PA01 grown in the presence of 0.85 M NaCl. We saw a highly significant increase in the abundance of trehalose, compared to growth on standard M9 media. This strongly suggests that trehalose accumulates under conditions of osmotic stress as suggested by the reviewer. These new results have been added to the relevant sections of the manuscript (M&M, results, table 2 and discussion). The student (Danny Ward) who conducted these new experiments has been added to the author list.

• However, there is also an extensive literature suggesting that trehalose has antioxidant functions e.g. PMID: 29241092 (the first paper that came up in Google search for "trehalose as antioxidant"). The authors should discuss this possible alternate role of trehalose.

  The reviewer is correct that trehalose has well-documented antioxidant functions in various species. We have modified the introduction to address this. To maintain the focus of our manuscript on bacteria we have used a different example to that suggested by the reviewer.

• It is not described adequately in the Materials and Methods how the cellular contents of trehalose and maltose-1-phosphate (M1P) were determined.

  The Materials and Methods section has been revised to include more details of this method.

• I found the growth curves in Figure 8, especially in panel B, to be uninterpretable. The authors should spread these data into more panels or use some other method to make them clearer.

  We have expanded the legend for Figure 8 to describe more fully what is going on in this figure. The results in Figure 8 are grouped according to the operons in which each set of genes is located. As such, the graphs contain unequal numbers of curves, with 8B containing the most and 8C only showing data for WT and ΔglgP.

• The statement "The GlgA and GlgE proteins . . . enable two alternate mechanisms for linear α-glucan biosynthesis", which is echoed a number of times in the manuscript, seems to create the impression that there are two de novo pathways of synthesis of these polysaccharides. However, as shown in Figure 1, the GlgA pathway is the only route to the net synthesis of α-glucans, and GlgE is only part of a recycling pathway. Therefore, it cannot be true that "the vast majority of α-glucan accumulated by P. aeruginosa will be produced by GlgE".

  We have revised this section to further clarify what we mean when we state that the majority of α-glucan accumulated by P. aeruginosa will be produced by GlgE. Our data suggest that there is a big difference between the generation of α-glucan (conducted by both GlgA and GlgE) and its accumulation (flux through GlgA generated α-glucan is high, so only GlgE generated α-glucan can accumulate to generate large polymers).

• The authors state that "MalQ disproportionates (sic) α-glucan with glucose to produce maltose." Figure 1 shows that GlgE uses an "acceptor", which I assume could be glucose. How is free glucose synthesized? Could cells grown on a non-carbohydrate as sole carbon source make free glucose?

  P. aeruginosa is able to carry out gluconeogenesis, so it can produce glucose from non-carbohydrate carbon sources if necessary.

Our data show that GlgE acceptor preference gets lower as the acceptor molecule gets shorter. It is possible to detect GlgE activity without an acceptor. In this case we see a lag, implying M1P hydrolyses slowly at first and priming with glucose is also slow. Eventually however, the products get long enough for the reaction to take off. MalQ will work with DP2 or longer as the donor and DP1 or longer as the acceptor, moving one glucose unit at a time.

• Pedantic point, but "disproportionation" means an oxidation-reduction reaction in which two identical molecules are used to produce two different molecules (https://en.wikipedia.org/wiki/Disproportionation). The reaction catalysed by MalQ does not involve electron transfer. Don't the authors mean that this enzyme is a glycosyl transferase?

  We have checked this, and our use of disproportionation in the manuscript is correct. The definition of disproportionation is any desymmetrizing reaction of the following type: 2 A → A' + A", and is not limited to redox reactions. MalQ carries out a reaction of this type when presented with a maltooligosaccharide.

• The authors state that TreS had "a very high Km for trehalose (>100 mM)". In view of the low concentration of trehalose (Point 1, above), the physiological relevance of this suggested activity is questionable.

  See response to question 1 above. As trehalose levels are elevated under osmostress conditions this concern becomes less critical. It is of course true that conditions in vitro may not fully reflect cellular conditions and that this activity may be higher in vivo, but this is a general limitation of all protein biochemistry studies. The important point here is that trehalose synthase activity is detected for PA01 TreS.

• Explain better what "predicted mean log10(CFU) means.

  The predicted mean refers to the value of log10(CFU) predicted by the statistical model we use. We have clarified this in the relevant sections of the manuscript.

• Can the authors suggest how "α-glucan protects PA01 against desiccation"?

  Without further investigation we can only speculate as to how α-glucan confers desiccation tolerance in PA01. One possibility is that α-glucan functions as a hydrogel, like the exopolysaccharide alginate, trapping water molecules and slowing their evaporation. Alternatively, it may confer a structural role akin to that of trehalose, preventing the loss of cell integrity as water levels decrease. We now address these possibilities in the discussion.

• Can P. aeruginosa metabolize exogenous trehalose or maltose? If the authors know either way, they should mention it. If they don't know, I am not suggesting that they should test this for this paper, but it would be interesting to know whether these compounds would induce the expression of the trehalose or maltose catabolic enzymes or repress the relevant biosynthetic enzymes. >P. aeruginosa is able to metabolise exogenous maltose and trehalose. While the experiments that the reviewer suggests are certainly interesting, in our view tre/glg gene regulation is beyond the scope of the current manuscript. This field is certainly worth investigating in the future, however.

Minor points

• First page under "Results": "phosphomutase" should be "phosphoglucomutase"?

  Changed as requested.

• Discussion: insert "P. syringae" before "Pto".

  Changed as requested.

• Materials and Methods: describe how ADP was quantified in the maltokinase assay.

  The materials and methods section has been updated as requested.

Reviewer 2 (Significance):

Significance

Until this work, the biosynthesis of trehalose has been most extensively characterized in Escherichia coli, in which it has been shown that this disaccharide is made by the reaction of glucose-6-phosphate and UDP-glucose to give trehalose-6-phosphate and dephosphorylation to trehalose, catalysed by OtsA and OtsB. The authors discovered a very different pathway in P. aeruginosa in which the synthesis of trehalose goes through α-glucans as intermediates.

Because trehalose and α-glucans are needed for osmotic stress- and desiccation-tolerance, respectively, this work is of significance to researchers studying abiotic stress resistance.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

Review of manuscript "Trehalose and α-glucan mediate distinct abiotic responses in Pseudomonas aeruginosa" by S. D. Woodcock et al.

It will be difficult for me to write a review of this paper and for the authors to make sense of my review because the manuscript's pages / lines are not numbered. I will do my best write a review, but for the future, I urge this Journal to print the text on pages in which the lines are numbered or require this of the authors.

Summary.

The authors carried out a comprehensive characterization of the metabolism of trehalose in Pseudomonas aeruginosa PA01, using techniques of biochemistry, reverse genetics, and bioinformatics. The main findings include that the disaccharide trehalose is synthesized in this organism from branched chain α-glucans and that the catabolism of trehalose proceeds via another disaccharide, maltose and is fed back into the synthesis of α-glucans. Trehalose and α-glucans have been implicated in conferring resistance to abiotic stresses in other organisms. The authors show that mutants that are blocked in the synthesis of trehalose are sensitive to high salinity but are normal with respect to their sensitivity to desiccation, whereas mutants impaired in the accumulation of α-glucans are sensitive to desiccation without being unduly sensitive to osmotic stress. These results indicate that trehalose and α-glucans have different roles in abiotic stress-tolerance.

Major points.

This manuscript describes an impressive amount of careful work and presents new insights into the metabolism of trehalose, maltose, and α-glucans. However, the authors should address the following major comments before the paper is accepted.

1. Discussion: the authors state that "trehalose protects Pseudomonas ssp. against osmotic stress, most likely due to its role as a compatible solute." According to Table 2, P. aeruginosa grown in the medium of low osmolarity accumulated 0.13% trehalose per gram dry weight, i.e. ~4 μmol / g dry weight. Assuming that the dry weight / wet weight ratio of P. aeruginosa is the same as that of P. putida, which is ~1/3 (PMID: 6508285), the concentration of trehalose in the cells calculates to be ~2 mM. It is not plausible that trehalose could be significant as compatible solute at this low concentration.<br> One way out could be if the accumulation of this disaccharide were increased by osmotic stress. The authors should also measure the trehalose content of cells grown in medium containing 0.85 M NaCl. In case of positive results in this experiment, it would be interesting to determine the effects of osmotic stress on the levels of trehalose biosynthetic and catabolic enzymes, but this would not be necessary for the acceptance of the paper.<br> However, there is also an extensive literature suggesting that trehalose has antioxidant functions e.g. PMID: 29241092 (the first paper that came up in Google search for "trehalose as antioxidant"). The authors should discuss this possible alternate role of trehalose.<br> It is not described adequately in the Materials and Methods how the cellular contents of trehalose and maltose-1-phosphate (M1P) were determined.
2. I found the growth curves in Figure 8, especially in panel B, to be uniterpretable. The authors should spread these data into more panels or use some other method to make them clearer.
3. The statement "The GlgA and GlgE proteins . . . enable two alternate mechanisms for linear α-glucan biosynthesis", which is echoed a number of times in the manuscript, seems to create the impression that there are two de novo pathways of synthesis of these polysaccharides. However, as shown in Figure 1, the GlgA pathway is the only route to the net synthesis of α-glucans, and GlgE is only part of a recycling pathway. Therefore, it cannot be true that "the vast majority of α-glucan accumulated by P. aeruginosa will be produced by GlgE".
4. The authors state that "MalQ disproportionates (sic) α-glucan with glucose to produce maltose." Figure 1 shows that GlgE uses an "acceptor", which I assume could be glucose.<br> How is free glucose synthesized? Could cells grown on a non-carbohydrate as sole carbon source make free glucose? Pedantic point, but "disproportionation" means an oxidation-reduction reaction in which two identical molecules are used to produce two different molecules (https://en.wikipedia.org/wiki/Disproportionation). The reaction catalyzed by MalQ does not involve electron transfer. Don't the authors mean that this enzyme is a glycosyl transferase?
5. The authors state that TreS had "a very high Km for trehalose (>100 mM)". In view of the low concentration of trehalose (Point 1, above), the physiological relevance of this suggested activity is questionable.
6. Explain better what "predicted mean log10(CFU) means.
7. Can the authors suggest how "α-glucan protects PA01 against desiccation"?
8. Can P. aeruginosa metabolize exogenous trehalose or maltose? If the authors know either way, they should mention it. If they don't know, I am not suggesting that they should test this for this paper, but it would be interesting to know whether these compounds would induce the expression of the trehalose or maltose catabolic enzymes or repress the relevant biosynthetic enzymes.

Minor points.

1. First page under "Results": "phosphomutase" should be "phosphoglucomutase"?
2. Discussion: insert "P. syringae" before "Pto".
3. Materials and Methods: describe how ADP was quantified in the maltokinase assay.

Significance

Significance.

Until this work, the biosynthesis of trehalose has been most extensively characterized in Escherichia coli, in which it has been shown that this disaccharide is made by the reaction of glucose-6-phosphate and UDP-glucose to give trehalose-6-phosphate and dephosphorylation to trehalose, catalyzed by OtsA and OtsB. The authors discovered a very different pathway in P. aeruginosa in which the synthesis of trehalose goes through α-glucans as intermediates.<br> Because trehalose and α-glucans are needed for osmotic stress- and desiccation-tolerance, respectively, this work is of significance to researchers studying abiotic stress resistance.

The Reviewers' guidelines stipulate that Reviewers should define their fields of expertise.

My credentials are: a) I have been solicited to review this paper, and b) I have publications in osmotic stress adaptation and trehalose biosynthesis in Enterobacteriaceae.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

Evidence, reproducibility and clarity

Summary:

Provide a short summary of the findings and key conclusions (including methodology and model system(s) where appropriate). Please place your comments about significance in section 2.

The authors resolved the biosynthesis of trehalose and alpha-glucan in Pseudomonas aeruginosa and the role of these two compounds in osmotic and desiccation stress.

Major comments:

• Are the key conclusions convincing?

yes

• Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether?

no

• Would additional experiments be essential to support the claims of the paper? Request additional experiments only where necessary for the paper as it is, and do not ask authors to open new lines of experimentation.

Not necessary, comprehensive coverage of research Topic

• Are the suggested experiments realistic in terms of time and resources? It would help if you could add an estimated cost and time investment for substantial experiments.

Not applicable

• Are the data and the methods presented in such a way that they can be reproduced?

yes

• Are the experiments adequately replicated and statistical analysis adequate?

Yes, everything is adequate but just one subtle concern: check the significance of the number of digits in the entries listed in Table S3. Revise Table S3.

Minor comments:

• Specific experimental issues that are easily addressable.

not applicable (Table S3: see above)

• Are prior studies referenced appropriately?

No. Refs. 18- 32: The subjects of 'trehalose' and 'osmotic stress' have already been addressed in the Pseudomonas field and should be referenced. The authors cite work carried out on trehalose and osmotic stress on phylogenetically distant microorganisms, but do not cite related work from the Pseudomonas field which I consider to be inappropriate. Similarly, trehalose biosynthesis in Pseudomonas has not only been covered by refs. 47 and 48.

• Are the text and figures clear and accurate?

Extremely well written manuscript and prepared figures

• Do you have suggestions that would help the authors improve the presentation of their data and conclusions?

Revise the list of references and discuss more thoroughly your novel findings in the light of existing knowledge in the Pseudomonas field

Significance

2. Significance

• Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field.

Conceptual advance: The authors identified and characterized the enzymatic pathway of trehalose and alpha-glucan biosynthesis in Pseudomonas aeruginosa and its role to cope with osmotic and desiccation stress. The authors' conclusions do not correspond with recently published peers' work, hence they should discuss in more detail why they consider their data to be more accurate to discern the role of trehalose to contain desiccation and osmotic strass in P. aeruginosa.

• Place the work in the context of the existing literature (provide references, where appropriate).

Existing literature focusing on trehalose, osmotic stress, desiccation stress in the Pseudomonas field not cited by the authors

Pazos-Rojas LA, Muñoz-Arenas LC, Rodríguez-Andrade O, López-Cruz LE, López- Ortega O, Lopes-Olivares F, Luna-Suarez S, Baez A, Morales-García YE, Quintero- Hernández V, Villalobos-López MA, De la Torre J, Muñoz-Rojas J. Desiccation- induced viable but nonculturable state in Pseudomonas putida KT2440, a survival strategy. PLoS One. 2019 Jul 19;14(7):e0219554. doi:10.1371/journal.pone.0219554.

Wang T, Jia S, Dai K, Liu H, Wang R. Cloning and expression of a trehalose synthase from Pseudomonas putida KT2440 for the scale-up production of trehalose from maltose. Can J Microbiol. 2014 Sep;60(9):599-604. doi: 10.1139/cjm-2014-0330.

Harty CE, Martins D, Doing G, Mould DL, Clay ME, Occhipinti P, Nguyen D, Hogan DA. Ethanol Stimulates Trehalose Production through a SpoT-DksA-AlgU-Dependent Pathway in Pseudomonas aeruginosa. J Bacteriol. 2019 May 22;201(12):e00794-18. doi: 10.1128/JB.00794-18.

Cross M, Biberacher S, Park SY, Rajan S, Korhonen P, Gasser RB, Kim JS, Coster MJ, Hofmann A. Trehalose 6-phosphate phosphatases of Pseudomonas aeruginosa. FASEB J. 2018 Oct;32(10):5470-5482. doi: 10.1096/fj.201800500R.

Wang T, Jia S, Dai K, Liu H, Wang R. Cloning and expression of a trehalose synthase from Pseudomonas putida KT2440 for the scale-up production of trehalose from maltose. Can J Microbiol. 2014 Sep;60(9):599-604. doi: 10.1139/cjm-2014-0330.

Behrends V, Ryall B, Zlosnik JE, Speert DP, Bundy JG, Williams HD. Metabolic adaptations of Pseudomonas aeruginosa during cystic fibrosis chronic lung infections. Environ Microbiol. 2013 Feb;15(2):398-408. doi: 10.1111/j.1462-2920.2012.02840.x

Behrends V, Ryall B, Wang X, Bundy JG, Williams HD. Metabolic profiling of Pseudomonas aeruginosa demonstrates that the anti-sigma factor MucA modulates osmotic stress tolerance. Mol Biosyst. 2010 Mar;6(3):562-9. doi: 10.1039/b918710c.

Matthijs S, Koedam N, Cornelis P, De Greve H. The trehalose operon of Pseudomonas fluorescens ATCC 17400. Res Microbiol. 2000 Dec;151(10):845-51. doi: 10.1016/s0923-2508(00)01151-7.

van der Werf MJ, Overkamp KM, Muilwijk B, Koek MM, van der Werff-van der Vat BJ, Jellema RH, Coulier L, Hankemeier T. Comprehensive analysis of the metabolome of Pseudomonas putida S12 grown on different carbon sources. Mol Biosyst. 2008 Apr;4(4):315-27. doi: 10.1039/b717340g.

Hallsworth JE, Heim S, Timmis KN. Chaotropic solutes cause water stress in Pseudomonas putida. Environ Microbiol. 2003 Dec;5(12):1270-80. doi: 10.1111/j.1462-2920.2003.00478.x.

Ball P, Hallsworth JE. Water structure and chaotropicity: their uses, abuses and biological implications. Phys Chem Chem Phys. 2015 Apr 7;17(13):8297-305. doi: 10.1039/c4cp04564e

Cray JA, Russell JT, Timson DJ, Singhal RS, Hallsworth JE. A universal measure of chaotropicity and kosmotropicity. Environ Microbiol. 2013 Jan;15(1):287-96. doi: 10.1111/1462-2920.12018.

Chin JP, Megaw J, Magill CL, Nowotarski K, Williams JP, Bhaganna P, Linton M, Patterson MF, Underwood GJ, Mswaka AY, Hallsworth JE. Solutes determine the temperature windows for microbial survival and growth. Proc Natl Acad Sci U S A. 2010 Apr 27;107(17):7835-40. doi: 10.1073/pnas.1000557107.

These papers are of variable scientific quality, but the conceptual work by Hallsworth and the work by Behrens on the PA metabolome in CF lungs are worth discussing. All other work provides pieces of information on function and biosynthesis of trehalose up to now known by the Pseudomonas community. The authors resolved the function of the GlgA operon which will be definitely appreciated.

Strengths of the manuscript:

• Meticulously planned and carefully executed experiments, not a single experimental flaw • very high technical quality of experiments and primary data • comprehensive coverage of the research topic • excellent presentation in text and illustrations

only weakness: • insufficient consideration of peers' published work on trehalose and its role in stress response in P. aeruginosa

• State what audience might be interested in and influenced by the reported findings.

Scientists working in the fields of glycoconjugate and carbohydrate research, biochemists, microbiologists with interest in metabolic pathways, stress response and/or Pseudomonas

• Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate.

Reviewer's expertise: Pseudomonas genomics and physiology, respiratory tract infections, solid background in biochemistry and molecular biology

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reviewer #4

Evidence, reproducibility and clarity

We have reviewed "A specific regulator of neuronal V-ATPase in Drosophila melanogaster." by Dulac et al. The authors have identified VhaAC45L as a regulator of neuronal V-ATPase in Drosophila melanogaster. The authors have utilized multiple techniques to determine the localization of VhaAC45L in neurons and specifically in the synapse. The use of multiple approaches including determining RNA levels in different regions of the fly, and using CRISPR-Cas9 technique to insert V5 tag, makes a very convincing argument about the synapse-specific expression of VhaAC45L.

The combined use of co-immunoprecipitation technique and LC/MS to show that VhaAC45L co-precipitated with V-ATPase complex subunits is convincing that VhaAC45L is a subunit of V-ATPase. To determine the role of VhaAC45L in acidification of synaptic vesicles the authors have utilized pHluorins in combination with multiple RNAi lines. The authors have used a well-designed experiment to prove that VhaAC45L regulates acidification of the synaptic vesicles. Further, larval locomotion and quantal size determination using VhaAC45LRNAi which is known to be altered due to pH gradient of synaptic vesicles shows the functional role of VhaAC45L in synaptic vesicle acidification.

Minor comments:

1. For all graphs, please remove gridlines to make data points more visible.
2. Line 120-123: Authors indicate the VhaAC45LRNAi induced lethal phenotype when expressed in glutamatergic and cholinergic drivers but the figure is missing. Please indicate as "data not shown" if not included in Figure.
3. A diagram summarizing the role of VhaAC45L in V-ATPase enzymatic complex and specific role is recommended.

Significance

V-ATPase play a crucial role at the synapse by being responsible for acidification of the synaptic vesicles and identification of a synaptic vesicle specific regulator of V-APTase is important to understand the complex regulation of synapse function. The authors have used well-designed experiments to convince the localization and function of VhaAC45L in synaptic vesicle acidification.

Referees cross commenting

The summary of Reviewer#2 looks good!

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reviewer #3

Evidence, reproducibility and clarity

In this study, Dulac and colleagues investigated roles of VhaAC45-like gene, which codes one of the V-ATPase accessory proteins in Drosophila, in synaptic transmission. First, they demonstrated that VhaZC45L transcripts are expressed selectively in neurons and that the gene products are addressed to synaptic areas. Second, they showed that VhaAC45L is co-immunoprecipitated with some subunits of V-ATPases, which is consistent with bio-informatics predictions. They further demonstrated that VhaAC45L-knockdown (KD) resulted in defects in synaptic vesicle acidification as well as a reduction in quantal size of glutamate, indicating that VhaAC45L play a key role in regulating neurotransmitter release by modulating the driving force for transmitter uptake. Last, not least, they demonstrated that VhaAC45L-KD in motoneurons attenuated larvae locomotor performance, indicating its physiological relevance. Overall, this study is rigorously executed and nicely presented, and adds one more component of the V-ATPase that is responsible for neurotransmitter uptake into synaptic vesicles. However, since this study simply confirmed an established notion from other species such as yeast and mammals that AC45 is one of the accessory proteins of the V-ATPase complex, a conceptual novelty beyond the previous knowledge is relatively poor in its present form. Thus, this reviewer would suggest several issues as following to improve the comprehensiveness as well as novelty of the current manuscript.

1. The reason why the authors focused on VhaAC45-'like' is somewhat obscure, and therefore should be explained. How different VhaAC45 and VhaAC45L are in terms of amino acid sequences, tissue distributions, and KO phenotypes. It seems more comprehensive if the authors provide some experimental evidence on VhaAC45; e.g. whether it is also expressed in neurons or not (Fig. 1), and, if VhaAC45 is neuronal, whether it can rescue the phenotypes of VhaAC45L-KD to certain degree (Figs 4 & 5).
2. What is the mechanism of Ac45 in regulating V-ATPase activity? In mammals, it has been suggested that Ac45 is essential for proper sorting of the V-ATPase to the destined organelles (e.g. Jansen et al., Mol. Biol. Cell., 2010; Jansen et al., BBA, 2008). In this context, it should be examined whether VhaAC45L-KD would affect the synaptic localization of other V-ATPase subunits.
3. Given that a rodent brain SV contains a few copies of the V-ATPase on average (Takamori et al., 2006, and some newer papers by others), it is interesting that >80% reduction of Ac45 showed moderate effects on quantal size. If SVs under study also contains 1 or 2 V-ATPase per SV, there must be some SVs lacking VhAC45L upon KD. In this context, it is interesting to see how VhaAC-KD (RNAi1~3) affect the frequencies of minis.
4. In general, decrease in mini amplitudes is accounted for by changes in postsynaptic sensitivity for neurotransmitters. Although acidification deficits would support that decrease in quantal size is due to the decrease in the driving force for glutamate uptake, it should be examined whether the postsynaptic receptor fields are not affected by VhaAC45L-KD by recording postsynaptic response upon application of non-saturable concentrations of glutamate.
5. Related to 4, it is also interesting to see if evoked responses are also attenuated as a result of VhaAC45L-KD, which is more physiologically relevant for locomotor activity phenotype than minis.

Minor points

1. Quantal size of glutamate is not affected by reduced expression of DVGLUT (Daniels et al., Neuron, 2006), which highly contrasts with VhaAC45L, expression of which defines quantal size. Distinct regulation of quantal size by the transporter and the V-ATPase subunit should be discussed.
2. For electrophysiological experiments, respective sample traces should be shown in Figure 5.
3. Only RNAi1 and RNAi2 lines were examined for SV pH estimation and mini analysis. The results from RNAi3 should be presented, or at least mentioned in the text.

Significance

As mentioned above, as it stands, the authors merely confirmed the pre-existing bioinformatic knowledge on one of the AC45 homologues in Drosophila. The audience of The EMBO Journal might be interested in how different/similar VhaAC45 and VhaAC45-like are, and their functional relevance. In particular, is VhaAC45 also mandatory for the V-ATPase functioning in neurons? Adding some basic information of VhaAC45, e.g. tissue distribution, KO phenotypes, and ability to rescue the VhaAC45-like-KD phenotypes, will certainly improve the comprehensiveness of this study, and capture audience's attention.

Referees cross commenting

I am fine with the summary of Reviewer#2

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reviewer #2

Evidence, reproducibility and clarity

In this study and using Drosophila melanogaster as a model system, Dulac et al report the very interesting discovery of a previously characterized neuronal specific regulator of the V-ATPase called VhaAC45L. They combine genetics, IHC, Mass spec and ephys to unravel the expression pattern and function of this protein. They find that it is required to acidify synaptic vesicles in glutamatergic neurons of the Drosophila larval neuromuscular junction, for appropriate synaptic transmission and for larval locomotion. The experiments are very well performed, the data presented very convincing and the paper is well written. Nonetheless, a few additional pieces of evidence and some level of expanded analysis would strengthen the conclusions and increase the depth of the work.

Major comments:

1. Figure 1F: the while the localization to the presynaptic terminal is convincing, where exactly the protein is localized to is not studied. The imaging in these experiments could use increased resolution and concomitantly colocalization studies with more specific synaptic vesicle markers.
2. Figure 3B-G: these experiments should be complemented by a rescue experiment, ideally of the null mutant using a UAS construct and a pan neuronal driver, or - if such animals are viable to the third larval instar stage - a glutamatergic driver. If possible, it would also be good to study the NMJ phenotype of the null mutant rescued to viability using a neuronal driver that does not express in motor neurons (e.g. Chat-G4).
3. Figure 5: the authors focus on quantal size which measures the postsynaptic response to spontaneous release from the presynaptic terminal. However, it is unclear how this directly relates to the locomotor deficit beyond signaling potential deficits in vesicle loading or fusion. It would be more convincing to also study evoked release, and expand the analysis of presynaptic properties (number of events, amplitude, frequency).
4. General: showing some level of genetic interaction with V-ATPase subunits in at least some of the assays would be welcome.

Minor comments:

Some of the images, especially those in Figure 3, should be larger for ease of visualization.

Significance

The discovery of a neuronal specific regulator of the V-ATPase is very interesting. To my knowledge it is the first description of a neuronal specific V-ATPase related protein since the description of Vha100-1 by Hiesinger and colleagues in 2005. The work is therefore of great interest to researchers working on synaptic function in general and on synaptic vesicle biology in particular.

I note that I do not have in depth expertise in electrophysiology, although I am sufficiently familiar with basic NMJ physiology experiments to render the opinions stated above.

Referees cross commenting

There seems to be overall consensus among the reviewers on 3 issues:

1. A somewhat more precise understanding of the role of vhaAC45L in the synaptic vesicle cycle through better localization studies and some classic assays (like FM dye uptake).
2. A little more characterization of the transmission defects (e.g. studying evoked responses) would be welcome.
3. Ascertaining the validity of the alleles with rescue experiments, perhaps in the V5 mutant background to allow localization analysis in a rescued background.

I think further biochemical analysis is interesting but probably beyond the scope of this initial description and would take too much time.

The minor issues are easy to address

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reviewer #1

Evidence, reproducibility and clarity

Dulac et al. present a first in vivo characterization of the 'accessory' v-ATPase subunit vhaAC45L in Drosophila. The key findings are localization and association of the protein with v-ATPase complexes at synapses and a functional requirement based on lethality and reduced synaptic function. This is certainly a useful contribution to our understanding of neuronal v-ATPase functions in vivo. The main weakness of the study is a lack of depth. The study focuses on localization, co-IP of associated proteins, an analysis of acidification and reduced synaptic function in fly larvae, thus providing a baseline for mechanistic study. However, the mechanism of vhaAC45L is not addressed in this short report. How does is vhaAC45L function different from its homolog vhaAC45? Is it required for v-ATPase assembly? Is it required to localize the full v-ATPase complex (or just V0) to the synapse? Is the defect really due to partial loading of synaptic vesicles or does loss of vhaSC45L also affect endosomal and lysosomal function at synapses? The work as is certainly represents a publishable contribution without answering any of these questions - more as an invite for the community to study the role of vhaAC45L; however, I feel this is a bit of a missed opportunity to put the function of a new potential regulator of specific synaptic v-ATPase functions in the context of the most basic functions obvious in this field.

My main concerns are:

1. clearly, vhaAC45L is required for SOME function of v-ATPase in neurons - but it remains entirely unclear which one. It is not even clear what compartments are affected. Reduced quantal size of single vesicle exocytosis events can be a direct or indirect consequence of problems in SV biogenesis and recycling. Is exo- /endocytosis unaffected? (FM1-43 uptake!). What compartments are affects? (markers for synaptic vesicles versus lysosomal compartments!).
2. molecular function: is vhaAC45L required for v-ATPase assembly? (IP/Pull-downs of v-ATPase complexes in the presence or absence of vhaAC45L with other subunits!).
3. vha100 was proposed in Drosophila to function on synaptic vesicles and the lysosomal pathway, but, if I remember correctly, here quantal size was normal. I am missing a comparison between the two.
4. The V5 knock-in is used both as a mutant as well as a tool to analyze protein localization. This is likely okay, but a little concern of course has to be that by creating a mutant protein through stop codon deletion its subcellular localization, turnover, etc. are not normal. Similarly, anti-V5 co-IPs will isolate proteins bound to the mutant variant of vhaAC45L. Minimally, IPs or pull-downs using other members of the V0 complex should be done to understand the role of vhaAC45L in direct comparison with vhaAC45 on complex assembly and possibly targeting to the synapse (or ideally targeting to specific compartments).

Significance

There is significance to the reporting of an accessory v-ATPase subunit required for SOME function of the v-ATPase in neurons. There is some lack of significance in the absence of basic mechanistic insight as to what vhaAC45L does to the v-ATPase in neurons.

Referees cross commenting

I'm fine with this summary!

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

We would like to thank the editors and the four reviewers for their careful consideration of our manuscript. We are very grateful for their positive appreciation of our work and we believe that their suggestions, which have been included in the preliminary revised version of the manuscript whenever possible, have greatly improved the quality of the paper and have helped us deepen our understanding of the results.

We were happy to note that all the reviewers found value in our work, as stated in their general comments: “This is certainly a useful contribution to our understanding of neuronal V-ATPase functions in vivo” (…)” (Reviewer 1) – “Dulac et al report the very interesting discovery of a previously uncharacterized neuronal specific regulator of the V-ATPase. (…) The experiments are very well performed, the data presented very convincing and the paper is well written.” (Reviewer 2) – “The discovery of a neuronal specific regulator of the V-ATPase is very interesting (…) The work is therefore of great interest to researchers working on synaptic function in general and on synaptic vesicle biology in particular.” (Reviewer 3) – “The authors have used well-designed experiments to convince the localization and function of VhaAC45L in synaptic vesicle acidification.” (Reviewer 4).

In their remarks, the reviewers suggested additional experiments that could be done to improve our understanding of the role of this new V-ATPase regulator, as well as several minor issues. We have addressed all their comments in our answers below, in which the full text of the reviews is included in blue type, and the responses in black. The line numbers refer to the revised version of the manuscript.

Reviewer #1

Dulac et al. present a first in vivo characterization of the 'accessory' v-ATPase subunit vhaAC45L in Drosophila. The key findings are localization and association of the protein with v-ATPase complexes at synapses and a functional requirement based on lethality and reduced synaptic function. This is certainly a useful contribution to our understanding of neuronal v-ATPase functions in vivo. The main weakness of the study is a lack of depth. The study focuses on localization, co-IP of associated proteins, an analysis of acidification and reduced synaptic function in fly larvae, thus providing a baseline for mechanistic study. However, the mechanism of vhaAC45L is not addressed in this short report. How does is vhaAC45L function different from its homolog vhaAC45? Is it required for v-ATPase assembly? Is it required to localize the full v-ATPase complex (or just V0) to the synapse? Is the defect really due to partial loading of synaptic vesicles or does loss of vhaAC45L also affect endosomal and lysosomal function at synapses? The work as is certainly represents a publishable contribution without answering any of these questions - more as an invite for the community to study the role of vhaAC45L; however, I feel this is a bit of a missed opportunity to put the function of a new potential regulator of specific synaptic v-ATPase functions in the context of the most basic functions obvious in this field.

My main concerns are:

1. clearly, vhaAC45L is required for SOME function of v-ATPase in neurons - but it remains entirely unclear which one. It is not even clear what compartments are affected. Reduced quantal size of single vesicle exocytosis events can be a direct or indirect consequence of problems in SV biogenesis and recycling.

Is exo- /endocytosis unaffected? (FM1-43 uptake!).

We agree that alterations in the synaptic vesicle release/recycling cycle could indeed contribute to the locomotion defect, in addition to the acidification impairment observed in VhaAC45L knockdown larvae. As suggested by the reviewer, we plan to carry out FM-dye assays to measure endocytosis and exocytosis at the neuromuscular junction of control versus VhaAC45L-KD animals. If successful, a new figure will be added to the final version of the paper.

What compartments are affected? (markers for synaptic vesicles versus lysosomal compartments!).

Finding out whether VhaAC45L is specifically involved in the acidification of synaptic vesicles, or if it also plays a similar role in other synaptic organelles, in particular lysosomes, would be very interesting indeed. However, we found that it was technically difficult to address this issue in the Drosophila nervous system. A good way would be to check whether the lysosomal pH is affected by VhaAC45L knockdown, as it is the case for synaptic vesicles.

Unfortunately, because lysosomes are not abundant in neurons, lysosome-specific pH-sensitive probes such as Lysotracker do not yield detectable signals at Drosophila larval synapses. So, whether VhaAC45L is specific for synaptic vesicles or involved in the regulation of V-ATPase activity in all neuronal compartments reminas an open question for now.

1. molecular function: is vhaAC45L required for v-ATPase assembly? (IP/Pull-downs of v- ATPase complexes in the presence or absence of vhaAC45L with other subunits!).

In accordance with the reviewer, we are also very much eager to learn more about the precise molecular function of VhaAC45L, and in particular whether it is required or not for assembly of the V-ATPase complex. Pull-downs of V-ATPase proteins in controls versus VhaAC45L-KD could be used to address this question, but this would require a large quantity of antibodies directed against subunits of the V0 and V1 domains, respectively. Unfortunately, there are no such antibodies commercially available against Drosophila V-ATPase proteins. We have tried several antibodies that recognize V-ATPase subunits from other species and were predicted to react against Drosophila homologs, but with no success. The only V-ATPase antibodies currently at our disposal were samples generously sent to us by other laboratories in insufficient quantities for carrying out such experiments. To our regret, therefore, we were not able to answer this question until now because of the lack of appropriate tools.

1. vha100 was proposed in Drosophila to function on synaptic vesicles and the lysosomal pathway, but, if I remember correctly, here quantal size was normal. I am missing a comparison between the two.

We thank the reviewer for this comment. A comparison with previously published results on subunit Vha100-1 has now been added (lines 458-469) in the discussion related to this topic in the revised manuscript.

1. The V5 knock-in is used both as a mutant as well as a tool to analyze protein localization. This is likely okay, but a little concern of course has to be that by creating a mutant protein through stop codon deletion its subcellular localization, turnover, etc. are not normal. Similarly, anti-V5 co-IPs will isolate proteins bound to the mutant variant of vhaAC45L. Minimally, IPs or pull- downs using other members of the V0 complex should be done to understand the role of vhaAC45L in direct comparison with vhaAC45 on complex assembly and possibly targeting to the synapse (or ideally targeting to specific compartments).

It is indeed a legitimate concern to question the physiological relevance of results obtained by studying V5-tagged VhaAC45L. However, the V5 tag is very small (14 amino acids) and we fused it in place of the stop codon to keep intact the whole sequence of the protein. In addition, we found that the V5 knock-in flies are viable and fertile as homozygous. Given that the null mutants, as well as strong RNAi knockdowns, are lethal at early developmental stage, this suggests that the V5 knock-in has limited negative effects, if any, on VhaAC45L function. This led us to believe that at least a good portion of the V5-tagged protein might be targeted to the right subcellular compartment, and associate to its physiological partners.

Significance:

There is significance to the reporting of an accessory v-ATPase subunit required for SOME function of the v-ATPase in neurons. There is some lack of significance in the absence of basic mechanistic insight as to what vhaAC45L does to the v-ATPase in neurons.

We agree that we did not elucidate here the precise molecular mechanisms by which VhaAC45L contributes to synaptic vesicle acidification. It is rather an initial description of a novel neuronal protein that appears to be essential for proper synaptic functioning, and we provide consistent evidence that its function requires specific interaction with the V-ATPase complex, and in particular with three subunits that reproducibly co-immunoprecipitated with VhaAC45L (namely Vha1C39-1, Vha100-1 and ATP6AP2). Please note that it took many years and many papers before the molecular mechanisms of action of comparable accessory subunits, such as ATP6AP1/AC45 or ATP6AP2, was better understood, and it is still nowadays a matter of investigation. It is therefore very demanding to expect that we describe the exact function of the previously uncharacterized VhaAC45L at all levels in a single first paper.

Reviewer #2

In this study and using Drosophila melanogaster as a model system, Dulac et al report the very interesting discovery of a previously uncharacterized neuronal specific regulator of the V-ATPase called VhaAC45L. They combine genetics, IHC, Mass spec and ephys to unravel the expression pattern and function of this protein. They find that it is required to acidify synaptic vesicles in glutamatergic neurons of the Drosophila larval neuromuscular junction, for appropriate synaptic transmission and for larval locomotion. The experiments are very well performed, the data presented very convincing and the paper is well written. Nonetheless, a few additional pieces of evidence and some level of expanded analysis would strengthen the conclusions and increase the depth of the work.

Major comments:

1. Figure 1F: the while the localization to the presynaptic terminal is convincing, where exactly the protein is localized to is not studied. The imaging in these experiments could use increased resolution and concomitantly colocalization studies with more specific synaptic vesicle markers.

We agree that it would be very good to show this additional result. However, confocal microscopy does not provide sufficient resolution to localize the protein at the membrane of individual synaptic vesicles. Another way would be to see if VhaAC45L immunostaining co- localizes with domains enriched in synaptic vesicle markers, but these organelles are rather ubiquitously distributed in the synaptic boutons at the Drosophila neuromuscular junction. To correctly perform this experiment, we would have to do immuno-electron microscopy, a technique we do not master in our laboratory and that we did not plan to implement for the present work.

1. Figure 3B-G: these experiments should be complemented by a rescue experiment, ideally of the null mutant using a UAS construct and a pan neuronal driver, or - if such animals are viable to the third larval instar stage - a glutamatergic driver. If possible, it would also be good to study the NMJ phenotype of the null mutant rescued to viability using a neuronal driver that does not express in motor neurons (e.g. Chat-G4).

Although a rescue experiment could potentially add a further evidence that Vha45ACL deficiency is responsible for the synaptic vesicle acidification defect described in Figure 3, we don’t think that it is a requisite here because we obtained similar results by knocking down the gene using two different RNAis. As described in the manuscript, the pan-neuronal expression of Vha45ACL could rescue the embryonic lethality of the null mutant, so it would be theoretically possible to check the acidity level of synaptic vesicles at the neuromuscular junction of the recued larvae. However, this would involve making rather complex genetic constructions to express VMAT-pHluorin in motor neurons in rescued mutant background. In addition, the conclusions we could draw from such experiment would be limited by the lack of comparison. Indeed, in Figure 3 the defect was observed in knockdown context, and the same experiment could not be performed in knockout larvae due to the early lethality. If we could measure the acidity level of rescued null mutants, we would not have any comparison point besides the knockdown experiments. As knockout and knockdown are not likely to produce identical phenotype (especially in terms of magnitude of effect), the ideal would be to compare the rescued phenotype to the null mutant expressing VhaAC45L in all neurons except motoneurons, as suggested by the reviewer. However, such genotype would certainly not be viable, since we observed that expression of VhaAC45L RNAis with a stronger motoneurons driver (D42-Gal4) was sufficient to induce lethality at early developmental stage.

1. Figure 5: the authors focus on quantal size which measures the postsynaptic response to spontaneous release from the presynaptic terminal. However, it is unclear how this directly relates to the locomotor deficit beyond signaling potential deficits in vesicle loading or fusion. It would be more convincing to also study evoked release, and expand the analysis of presynaptic properties (number of events, amplitude, frequency).

We fully agree with this comment shared by Reviewers 2 and 3 related to the electrophysiology experiments. Note that these experiments have been carried out in collaboration with another laboratory located in another city. The Covid-19 situation during the past year has prevented, and is still complicating, movements between labs, preventing us from going further with the electrophysiology analyses of VhaAC45L KD. If the situation in the near future allows it, we would very much like to add a more extensive electrophysiological analysis, including in particular the study of evoked release. In the revised manuscript, we have nevertheless completed Figure 5 by adding representative distributions of spontaneous mEPSP amplitudes in control and VhaAC45L knockdown larvae, as well as the results of new analyses showing lack of effects the KD on the mean EPSP frequency.

1. General: showing some level of genetic interaction with V-ATPase subunits in at least some of the assays would be welcome.

We are definitely in accordance with the reviewer on that point, but we think that this would involve a lot of work and be beyond the scope of the present initial description. Here we show by proteomic analyses that at least 12 proteins co-precipitate and so potentially interact with VhaAC45L, three of them being previously identified constitutive or accessory V-ATPase subunits. In our opinion, studying the interactions between VhaAC45L and these proteins through genetic and molecular studies will be the subject of future works. As stated by Reviewer 2 in the Referees cross commenting below: “further biochemical analysis is interesting but probably beyond the scope of this initial description and would take too much time”. We fully agree with this statement.

Minor comments:

Some of the images, especially those in Figure 3, should be larger for ease of visualization.

As requested, the images of Figure 3 have been enlarged.

Significance

The discovery of a neuronal specific regulator of the V-ATPase is very interesting. To my knowledge it is the first description of a neuronal specific V-ATPase related protein since the description of Vha100-1 by Hiesinger and colleagues in 2005. The work is therefore of great interest to researchers working on synaptic function in general and on synaptic vesicle biology in particular.

We are grateful to the reviewer for his very positive assessment of our work.

I note that I do not have in depth expertise in electrophysiology, although I am sufficiently familiar with basic NMJ physiology experiments to render the opinions stated above.

Reviewer #3

In this study, Dulac and colleagues investigated roles of VhaAC45-like gene, which codes one of the V-ATPase accessory proteins in Drosophila, in synaptic transmission. First, they demonstrated that VhaZC45L transcripts are expressed selectively in neurons and that the gene products are addressed to synaptic areas. Second, they showed that VhaAC45L is co- immunoprecipitated with some subunits of V-ATPases, which is consistent with bio-informatics predictions. They further demonstrated that VhaAC45L-knockdown (KD) resulted in defects in synaptic vesicle acidification as well as a reduction in quantal size of glutamate, indicating that VhaAC45L play a key role in regulating neurotransmitter release by modulating the driving force for transmitter uptake. Last, not least, they demonstrated that VhaAC45L-KD in motoneurons attenuated larvae locomotor performance, indicating its physiological relevance. Overall, this study is rigorously executed and nicely presented, and adds one more component of the V- ATPase that is responsible for neurotransmitter uptake into synaptic vesicles. However, since this study simply confirmed an established notion from other species such as yeast and mammals that AC45 is one of the accessory proteins of the V-ATPase complex, a conceptual novelty beyond the previous knowledge is relatively poor in its present form. Thus, this reviewer would suggest several issues as following to improve the comprehensiveness as well as novelty of the current manuscript.

1. The reason why the authors focused on VhaAC45-'like' is somewhat obscure, and therefore should be explained. How different VhaAC45 and VhaAC45L are in terms of amino acid sequences, tissue distributions, and KO phenotypes. It seems more comprehensive if the authors provide some experimental evidence on VhaAC45; e.g. whether it is also expressed in neurons or not (Fig. 1), and, if VhaAC45 is neuronal, whether it can rescue the phenotypes of VhaAC45L- KD to certain degree (Figs 4 & 5).

Following the reviewer’s request, we have added a sequence alignment of VhaAC45 and VhaAC45L, as well as a graph showing tissue distributions of both genes in Supplementary Figure 1 of the revised manuscript. To our knowledge, there is no published functional study of VhaAC45 in Drosophila, so we can only make assumptions derived from studies on predicted homologs in evolutionarily distant species. For that reason, it is difficult to compare VhaAC45 to VhaAC45L, as it would first require an entire new study of VhaAC45 function in flies. Since our interest is to study neuronal physiology, we focused on VhaAC45L because compelling evidence indicates that this subunit is specific to the nervous system, as described in our manuscript, rather than on VhaAC45 which seems to be expressed in all tissues. In addition, homologs of VhaAC45L have never been functionally characterized to date in any species, making it very interesting to study this new protein in a genetically tractable organism.

1. What is the mechanism of Ac45 in regulating V-ATPase activity? In mammals, it has been suggested that Ac45 is essential for proper sorting of the V-ATPase to the destined organelles (e.g. Jansen et al., Mol. Biol. Cell., 2010; Jansen et al., BBA, 2008). In this context, it should be examined whether VhaAC45L-KD would affect the synaptic localization of other V-ATPase subunits.

We thank the reviewer for pointing out these very interesting references. We have indeed tried to determine the relative abundance of two other V-ATPase subunits at the larval neuromuscular junction in control and VhaAC45L knockdown contexts. However, because the tested subunits are not specific to neurons, and are expressed at relatively low levels in synapses, it was not possible for us to properly separate the synaptic signal from the background immunostaining in surrounding muscles. This unfortunately prevented us from performing an accurate and reliable quantification.

1. Given that a rodent brain SV contains a few copies of the V-ATPase on average (Takamori et al., 2006, and some newer papers by others), it is interesting that >80% reduction of Ac45 showed moderate effects on quantal size. If SVs under study also contains 1 or 2 V-ATPase per SV, there must be some SVs lacking VhAC45L upon KD. In this context, it is interesting to see how VhaAC-KD (RNAi1~3) affect the frequencies of minis.

The reviewer’s valuable comment prompted us to undertake new analyses on our electrophysiological recordings. We have now added in Figure 5E graphs showing the mean EPSP frequency for larvae expressing VhAC45L RNAi1 and RNAi2, which are the ones that were used in the quantal analysis. Both of these RNAi apparently decreased the frequency compared to controls, but this difference was not statistically significant. As detailed in the Discussion (line 458-469), this may suggest that VhaAC45L does not influence the abundance of the V-ATPase complex at nerve terminals, but rather its efficiency.

1. In general, decrease in mini amplitudes is accounted for by changes in postsynaptic sensitivity for neurotransmitters. Although acidification deficits would support that decrease in quantal size is due to the decrease in the driving force for glutamate uptake, it should be examined whether the postsynaptic receptor fields are not affected by VhaAC45L-KD by recording postsynaptic response upon application of non-saturable concentrations of glutamate.

Testing for potential postsynaptic receptor field alteration by glutamate application would be an interesting experiment indeed, but, as we believe, not a critical control for the present manuscript. Because we expressed RNAis presynaptically, any modification in the postsynaptic receptor field would have to be an indirect consequence of VhaAC45L downregulation in motoneurons, and so, likely to be related to the synaptic vesicle acidification defect. It would not change, therefore, our conclusion that VhaAC45L deficiency in motoneurons induces a decrease in quantal size. Because electrophysiology experiments were carried out in collaboration with another laboratory located in another city, the current sanitary context has so far prevented us from performing this test (please refer to our answer to comment 3 of Reviewer 2 for more details).

1. Related to 4, it is also interesting to see if evoked responses are also attenuated as a result of VhaAC45L-KD, which is more physiologically relevant for locomotor activity phenotype than minis.

We also agree with this comment, shared by Reviewer 2, to which we already responded above in our answer to comment 3 of Reviewer 2.

Minor points:

1. Quantal size of glutamate is not affected by reduced expression of DVGLUT (Daniels et al., Neuron, 2006), which highly contrasts with VhaAC45L, expression of which defines quantal size. Distinct regulation of quantal size by the transporter and the V-ATPase subunit should be discussed.

As suggested by the reviewer, a discussion of this point has been added (lines 458-469). and Daniels et al. 2006 is now cited in the revised manuscript.

1. For electrophysiological experiments, respective sample traces should be shown in Figure 5.

Quantal size is not directly visible in sample traces, so we added instead representative distributions of spontaneous mEPSP amplitudes in control and VhaAC45L knockdown larvae in the new Figure 5C.

1. <![endif]>Only RNAi1 and RNAi2 lines were examined for SV pH estimation and mini analysis. The results from RNAi3 should be presented, or at least mentioned in the text.

These experiments were performed using two different RNAi constructs to ensure that similar effects were observed and to exclude the possibility of potential off-targets. Knocking down VhaAC45L in neurons with RNAi1 and 2 was lethal at pupal stages, suggesting that they give similar levels of inactivation. RNAi3 systematically induced lighter phenotypes, producing viable adults, which led us to believe it had a lower efficiency. Because the results on synaptic vesicle acidification and electrophysiology were very consistent with RNAi1 and RNAi2, we considered that it was not necessary to repeat the experiment with RNAi3.

Significance

As mentioned above, as it stands, the authors merely confirmed the pre-existing bioinformatic knowledge on one of the AC45 homologues in Drosophila. The audience of The EMBO Journal might be interested in how different/similar VhaAC45 and VhaAC45-like are, and their functional relevance. In particular, is VhaAC45 also mandatory for the V-ATPase functioning in neurons? Adding some basic information of VhaAC45, e.g. tissue distribution, KO phenotypes, and ability to rescue the VhaAC45-like-KD phenotypes, will certainly improve the comprehensiveness of this study, and capture audience's attention.

As mentioned in our response to point 1 of the reviewer above, we have added more data comparing the structure and distribution of VhaAC45 and VhaAC45L in the revised manuscript. VhaAC45 appears to be ubiquitously expressed whereas VhaAC45L is neuron-specific.

Comparing VhaAC45 to VhaAC45L would require a completely new study of VhaAC45 function, because it has never been done before in Drosophila to our knowledge. This would require repeating all the experiments with this other gene, probably involving two more years of work, and would make for a much longer and very different manuscript. It is understandable that this cannot be envisaged. Because homologs of VhaAC45L have never been functionally characterized to date in any species, we considered that it was worth studying this new protein on its own.

Reviewer #4

We have reviewed "A specific regulator of neuronal V-ATPase in Drosophila melanogaster." by Dulac et al. The authors have identified VhaAC45L as a regulator of neuronal V-ATPase in Drosophila melanogaster. The authors have utilized multiple techniques to determine the localization of VhaAC45L in neurons and specifically in the synapse. The use of multiple approaches including determining RNA levels in different regions of the fly, and using CRISPR- Cas9 technique to insert V5 tag, makes a very convincing argument about the synapse-specific expression of VhaAC45L.

The combined use of co-immunoprecipitation technique and LC/MS to show that VhaAC45L co- precipitated with V-ATPase complex subunits is convincing that VhaAC45L is a subunit of V- ATPase. To determine the role of VhaAC45L in acidification of synaptic vesicles the authors have utilized pHluorins in combination with multiple RNAi lines. The authors have used a well- designed experiment to prove that VhaAC45L regulates acidification of the synaptic vesicles.

Further, larval locomotion and quantal size determination using VhaAC45LRNAi which is known to be altered due to pH gradient of synaptic vesicles shows the functional role of VhaAC45L in synaptic vesicle acidification.

Minor comments:

1. For all graphs, please remove gridlines to make data points more visible.

We found that gridlines can be helpful for the readers to assess approximate values on the graphs. So, we have not removed them but rather changed the colour to a light grey so it does not affect any more visibility. We have also placed the points over the error bars in all the graphs, so they become more apparent.

1. Line 120-123: Authors indicate the VhaAC45LRNAi induced lethal phenotype when expressed in glutamatergic and cholinergic drivers but the figure is missing. Please indicate as "data not shown" if not included in Figure.

This mention has been added in the manuscript (line 125).

1. A diagram summarizing the role of VhaAC45L in V-ATPase enzymatic complex and specific role is recommended.

We believe that it is too early in this first report to draw an accurate diagram summarizing the role of this new protein in the V-ATPase complex.

Significance

V-ATPase play a crucial role at the synapse by being responsible for acidification of the synaptic vesicles and identification of a synaptic vesicle specific regulator of V-ATPase is important to understand the complex regulation of synapse function. The authors have used well-designed experiments to convince the localization and function of VhaAC45L in synaptic vesicle acidification.

We thank the reviewer for his very positive appreciation of our work.

Referees cross commenting

(Written by Reviewer 2)

There seems to be overall consensus among the reviewers on 3 issues:

1. A somewhat more precise understanding of the role of vhaAC45L in the synaptic vesicle cycle through better localization studies and some classic assays (like FM dye uptake).

—See our answers to comments 1 of Reviewer 1 and Reviewer 2.

1. A little more characterization of the transmission defects (e.g. studying evoked responses) would be welcome.

—See our answers comment 3 of Reviewer 2.

1. Ascertaining the validity of the alleles with rescue experiments, perhaps in the V5 mutant background to allow localization analysis in a rescued background.

—See our answers to comment 2 of Reviewer 2.

I think further biochemical analysis is interesting but probably beyond the scope of this initial description and would take too much time.

We fully agree with this statement.

The minor issues are easy to address

We have addressed all of them in the preliminary revised version of the manuscript.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

Reviewer 1

We would like to thank you for the comments concerning our manuscript. We responded to each question, as described below. All the authors feel that our manuscript has been much improved by your comments.

Minor comments:

Q1. Fig 1 any male vs female mice differences in ATF6b expression?

Response. We performed qPCR using several tissues from male and felame WT mice, and confirmed no significant differences in Atf6b mRNA levels between male and female mice. We put this result in Figure S1 C.

Q2. Fig 2C. Please show molecular weight markers on blots

Response. We put molecular weight markers in Fig.2C, as you suggested.

Q3. Fig 2C. what are the doublet bands on calnexin?

Response. Calnexin is sometimes shown as double bands in tissues such as kidney, liver and heart by western blotting (Zeng et al., PLoS One. 2009 Aug 26;4(8):e6787). Although the mechanism is unknown, it could be due to the post-translational modification such as phosphorylation (Wong et al, J Biol Chem. 1998 Jul 3;273(27):17227-35) or partial degradation although proteinase inhibitors are added in the lysis buffer. To my knowledge, alternative splicing is not likely to be the case.

Q4. Fig 3. what are the ERSE sequences? several different binding sites are reported in literature.

Response. We put the ERSE sequence in Materials and Methods and in the Figure legends for Figure 3 as “CCAATN9CCACG (Yoshida et al., 1998)”.,

Q5. p8. What is meant by 5' Atf6b lacks 10 and 11?

Response We corrected to “Atf6b transcript, which lacks exon 10 and 11, in these mice”.

Discussion: Please clarify if anti-ATF6-beta antibodies were available for these studies.

Response. We tried different anti-ATF6β antibodies to detect endogenous ATF6β in culture neurons by western blot. We successfully observed both full-length and N-terminal fragment (the active form) using the one from Biolegend (#853202) (Figure 1E in the new version). We replaced the result with FLAG antibody in HEK293T cells in the old version.

Discussion: It is puzzling that ATF6a induces calreticulin more potently than ATF6b, but the calreticulin defect is selectively dependent on ATF6b. Could authors speculate on this paradox? It would be interesting to expand on differences between ATF6a and ATF6b function and phenotypes in Discussion in mouse and in people.

Response. In Discussion, we added sentences regarding a bit puzzling role for ATF6β in CRT expression in the CNS, as below. “All the data from RNA-sequence to the promoter analysis suggested that CRT expression was ATF6β-dependent in primary hippocampal neurons. However, overexpression of ATF6α and ATF6β both enhanced CRT promoter activity…”

And we proposed a new scenario as below,

“These results may raise a scenario that, in the CNS, expression of molecular chaperones in the ER is generally governed by ATF6α as previously described (Yamamoto et al., 2007) and that ATF6β functions as a booster if their levels are too low. However, expression of CRT is somewhat governed by ATF6β, and ATF6α functions as a booster. The underlying mechanism for this scenario is not clear yet, but neurons may require a high level of CRT expression even under normal condition, as described in Table S2, which may lead to the development of a unique biological system to constitutively produce CRT in neurons. Further studies are required to clarify the molecular basis how this unique system is constructed and regulated.”

Reviewer 2

We would like to thank you for the comments concerning our manuscript. We responded to each question, as described below. All the authors feel that our manuscript has been much improved by your comments.

Major comments:

Q1. The post-translational processing of ATF6beta must be demonstrated in hippocampal neurons and not in HEK293T cells in Figure 1E. The authors conclude on Page 6, line 18 that "these results suggest that ATF6beta functions in neurons" but it is not obvious how expression in HEK293T cells contributes to this conclusion in any way.

Response. We performed western blot with different anti-ATF6β antibodies to detect endogenous ATF6β in culture neurons. We successfully observed both full-length and N-terminal fragment (the active form) from the one from Biolegend (#853202). We therefore replaced the result in HEK293T cells with the one in the hippocampal neurons (Figure 1E in new version).

Q2. The hippocampal neurons are affected by the loss of ATF6β, even though the mice are not exposed to tunicamycin. Could the authors present evidence that there is physiological ER stress in hippocampal neurons? If not, why is ATF6beta required.

Response Evidence suggests that neuronal activities including excitatory signals can cause physiological ER stress and induce the UPR at the distal dendrites in the hippocampal neurons (Murakami et al., Neuroscience. 2007 Apr 25;146(1):1-8, Saito et al., J Neurochem. 2018 Jan;144(1):35-49). Among the UPR branches, Ire1-XBP1 pathway has been reported to play an important role in this dendritic UPR and expression of BDNF in cell soma (Saito et al., 2018). Although the present study focuses on the role of ATF6β in the pathological ER stress which causes neuronal death, we believe that it will be intriguing to analyze its role of ATF6β in the physiological ER stress and in the local UPR machinery in neurons.

Q3. In Figure 3, is there a specific reason why the authors do not mutate the ERSEs in the mouse CRT reporter, pCC1 and instead opt to analyze the huCRT reporter? Given that all the other observations in the manuscript are in mouse calreticulin, it is important to show that the ERSEs in the mouse calreticulin promoter are also regulated in an ATF6beta-dependent manner. Similar to the huCRT reporter, it is also crucial to examine if ATF6beta can regulate the mouse CRT promoter. This would provide an explanation for why calreticulin expression is not completely abolished in ATF6beta mutants.

Response We added the data of the deletion mutant of mouse CRT promoter, pCC3, which has only 415bp, but still keeps both ERSE1 and 2 in it. pCC3 showed similar promoter activity to pCC1 (Figure 3 B) and huCRT (wt) (Figure 3 C) in both of WT and Atf6b-/- neurons. Because pCC5, which has 260bp but does not have ERSEs in it, lost completely CRT promoter activity (Waser et al., 1997), it is most likely that mouse and human CRT promoters are regulated in a similar manner via ERSEs.

Q4. In Figure 5A and B, the density of Tubulin staining varies from panel to panel, and is much lower in ATF6beta mutants treated with Tg/Tm. Presumably this is because of cell death but this should be clarified in the main text. Additionally, it is unclear if the EthD-1 staining is nuclear localized. It would help if single channel images for Hoechst and EthD-1 were provided to visualize this.

Response In Figure 5A and B, we added the statement for the reduction of Calcein-AM (A) and βIII tubulin (B) in the main text. We also added single channel images for Hoechst and EthD-1 in Figure S4 to confirm the nuclear localization of EthD-1.

Q5. The literature reports that BAPTA-AM treatment itself could cause ER stress (e.g. PMID: 12531184). Here, the authors report the opposite effect. How could the authors reconcile the difference? The effects of BAPTA-AM and 2-APB must individually be examined in Figure 6C and not just in combination with Tm.

Response. We added the data that BAPTA-AM and 2-APB alone did not cause neuronal death at the concentrations used in this study in Figure S6 B and in the main text.

Q6. The authors allude to "impairment of Ca2+ homeostasis in ATF6beta mutants" in Page 13 Line 2, but do not show any direct evidence in support of it. While treatment with BAPTA-AM and 2-APB is a start in that direction, it certainly does not demonstrate that under homeostatic conditions in vivo or in vitro there is any change in calcium flux in ATF6beta hippocampal neurons. To make the case that there is indeed perturbation of Ca2+ in ATF6beta mutant hippocampal neurons, the authors need to examine calcium flux and measure calcium indicators and how they are affected when ER stress is induced in these mutant cells.

Response We added the data that the Ca2+ store in the ER was reduced and Ca2+ concentration in the cytosol increased in Atf6b-/- neurons both under normal and ER stress conditions in Figure 4C.

Q7. The effect of 2-APB and salubrinal alone on hippocampal neurons need to be examined in Figure 9B-D to eliminate the possibility that these drugs are not enhancing cell survival under normal conditions in a parallel manner.

Response We added the data that 2-APB and salubrinal alone did not cause neuronal death in the hippocampus in our model in Figure S8 C.

Q8. The rationale for the examination of Fos, Fosb and Bdnf is poorly described (page 14, line 13) and the conclusions from this line of experimentation are rather weak. The results from Figure 9 to some extent serve to confirm in vivo the data seen in Figure 6C but by no means provide a mechanism for why ATF6beta mutants have perturbed calcium homeostasis (page 14, line 22).

Response We agreed with your comments that the examination of Fos, Fosb and Bdnf is relatively weak. We, therefore, moved these data to supplementary information (Figure S8 A and B).

Minor comments:

Q1. Page 8, line 3: Their rationale for why ATF6beta 5'UTR sequences are seen in their RNA seq data is not clearly explained. This must be rewritten for clarity.

Response In Atf6b-/- mice, exon 10 and 11 were deleted by homologous recombination. Therefore, 5’ part of Atf6b gene including exon 1-9 can be transcribed. We added the statement in Results, as below.

“this may be due to the presence of the 5’ Atf6b transcript with exon 1-9 in these mice, in which exon 10 and 11 were deleted by homologous recombination.”

Q2. Page 8, line 5, the authors write that besides Atf6β , CRT was the only UPR-regulated gene downregulated in Atf6β mutant mice. The authors need to state how they defined "UPR-regulated genes". There must be a list, which the authors do not cite.**

Response. To avoid the possible confusion, we changed the term “UPR-regulated genes” to “ER stress-responsive genes”.

Q3. Page 9, line 10: A reference is required for ERSEs.

Response We added the reference for ERSEs, as you suggested.

Q4. Page 10, line 6: The authors say "ATF6beta specifically induces CRT promoter activity". This is a confusing statement because "induction" is in response to stress, but the context here is homeostatic regulation since there is ostensibly no stress being induced. This distinction should be made and corrected here and throughout the manuscript.

Response To avoid the confusion, we changed the sentence to “ATF6β specifically enhances CRT promoter activity”.

Q5. Page 10, line 16: The use of "latter" here is confusing and it would help to restructure this sentence for clarity.

Response To avoid the confusion, we changed the phrase to “under control condition and after stimulation with Tg (Figure 4A upper row) or Tm (Figure 4A lower row)

Q6. Figure 9A is missing Y-axis labels.

Response We changed Figure 9A (Figure S8 A in new version) and Figure Legends to clarify what each axis indicates.

Reviewer 3

We would like to thank you for the comments concerning our manuscript. We responded to each question, as described below. All the authors feel that our manuscript has been much improved by your comments.

Major comments

Comment #1. The authors show that overexpression of either Atf6a or Atf6b both increase Crt expression in Atf6b knockout cells. While it is clear that deletion of Atf6a does not basally reduce Crt levels, the overexpression experiment does lead to a question as to how Atf6b can specifically be involved in regulating Crt expression. In the discussion, the authors seem to propose that homo- and hetero-dimerization of ATf6a and Atf6b are required for the basal expression of Crt and that Atf6b serves as a 'booster' of ER chaperone expression. They explicitly state that "Atf6a and Atf6b are required to induce CRT expression". However, it remains unclear to me why in this case would Atf6a deletion not impair Crt expression? The authors address this by invoking a mechanism whereby hippocampal neurons are more reliant on Atf6b for Crt expression, but this doesn't really make sense to me. Ultimately, this point underscores the lack of clear mechanistic basis to explain how Atf6b selectively regulates Crt in the hippocampus. This needs to be better resolved through more experimentation. For example, a ChIP experiment monitoring the binding of ATF6b and ATF6a to the Crt promoter in hippocampal and control cells would go a long way towards addressing this issue.**

Response. In Discussion, we first made the point clearer that CRT expression is ATF6β-dependent, while those of other molecular chaperones in the ER are ATF6α-dependent. Then, we raised a scenario that, in the CNS, expression of molecular chaperones in the ER is generally governed by ATF6α as previously described (Yamamoto et al., 2007) and ATF6β functions as a booster if their levels are too low. However, expression of CRT is somewhat governed by ATF6β, and ATF6α functions as a booster. We also wrote the limitation of the current study and requirement of the further study to clarify the molecular basis of the unique system to ensure CRT expression in neurons.

Comment #2. The importance of ATF6b for protecting against insults needs to be better described. For example, the authors should show that overexpression of ATF6b protects against ER stress induced neuronal toxicity in cell culture and in vivo kainate induced neuronal toxicity. Similarly, the authors should evaluate how overexpression of ATF6a protects against these insults to better define the specific dependence of hippocampal neurons on ATF6b. The authors do show that overexpression of ATF6b can rescue the reduced Crt observed in Atf6b-deleted neurons, but the protection should similarly be demonstrated.**

Response. We performed rescuing experiments to see both of ATF6β and ATF6α overexpression improve cell viability of Atf6b-/- neurons under ER stress. Interesting. ATF6β, but not ATF6α, rescued Atf6b-/- neurons. In Discussion, we raised the possible reasons as below.

“The lack of rescuing effect of ATF6α may be due to the fact that this molecule enhances the expression of different genes including cell death-related molecule CHOP in addition to molecular chaperons in the ER (Yoshida et al., 2000).”

Comment #3. Similar to #2, the authors should show that the potential for ATF6b (and ATF6a) overexpression to protect against different insults is impaired in Crt+/- neurons. The authors demonstrate that Crt-depletion increases sensitivity to toxic insults. This would go a long way to demonstrate the importance of the proposed ATF6b-CRT signaling axis in regulating neuronal survival in response to pathologic insults.**

Response. Unfortunately, right now, the breeding of Calr+/- mice is not in good condition. Although we are increasing the number of mice used for breeding, we have to wait pregnancies to get embryos for isolating neurons from hippocampus. Once we get enough number of mice, we would try the rescuing experiment of Calr+/- hippocampal neurons with ATF6β and ATF6α. However, we also think rescuing experiments of Atf6b-/- neurons by ATF6β, ATF6α, and CRT may be enough in this paper.

Comment #4. When reporting the RNAseq data, the authors should use the q-value (i.e., FDR) instead of the p-value. This will likely affect the number of genes reported in Table 1, but it is the appropriate statistical test for this type of data.**

Response. As you suggested, we replace Table1 with a new list which was filtered with the q-value. However, some important and consistent information were obtained from the list filtered with the p-value, we keep it as Table S1 in the supplementary information.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

In this manuscript, the authors define the functional importance of ATF6b in the hippocampus. They show that ATF6b is highly expressed in the hippocampus relative to other tissues. They demonstrate that deletion or depletion of ATF6b in cultured hippocampal neurons enhances ER stress induced death. Similarly, Atf6b-/- mice show increased sensitivity to kainate induced neuronal death. These results reveal an important role for ATF6b in regulating hippocampal survival in response to pathologic insults. To define a molecular basis for this protection, the authors utilized RNAseq to identify the lectin chaperone calreticulin (Crt) as a gene whose expression is basally reduced in cultured hippocampal neurons where Atf6b is deleted. They show the re-overexpression of Atf6b (or Atf6a) both restore Crt levels in these neurons, underscoring the importance of Atf6 in regulating basal Crt levels. They go on to demonstrate that loss of Atf6b impairs ER stress-dependent increases in Crt, while minimally impacting other Atf6 target genes, again highlighting the importance of Atf6b for Crtregulation. Importantly, overexpression of Crt rescues the increased ER stress-induced toxicity observed in Atf6b knockout neurons, indicating that a primary mechanism by which Atf6b regulates neuronal survival in response to ER stress is through increased Crt expression. Consistent with this, mimicking the 50% reduction in Crt observed in Atf6b knockout neurons using Crt+/- mice showed similar sensitivity to kainate induced neuronal death. Collectively, these results describe an Atf6-Crt axis that is important for regulating neuronal survival in response to pathologic insults.

Overall the experiments are interesting and provide new insights into the importance of Atf6b in neuronal survival. Notably, the evidence showing that loss of Atf6b increases hippocampal neuron sensitivity to ER stress and kainate induced toxicity are compelling. Any results describing the biological function of Atf6b are interesting, considering how little we know about this ER stress sensing protein. That being said, I have some concerns about the work described that require addressing before publication. Notably, I think more work needs to be done to define the molecular basis for the specific dependence of Crt expression on ATF6b in hippocampal neurons. Further, the authors need to do more experiments to demonstrate the specific importance of ATF6b signaling in the context of ER stress and in vivo neuronal death. I outline these various concerns below:

Comment #1. The authors show that overexpression of either Atf6a or Atf6b both increase Crt expression in Atf6b knockout cells. While it is clear that deletion of Atf6a does not basally reduce Crt levels, the overexpression experiment does lead to a question as to how Atf6b can specifically be involved in regulating Crt expression. In the discussion, the authors seem to propose that homo- and hetero-dimerization of ATf6a and Atf6b are required for the basal expression of Crt and that Atf6b serves as a 'booster' of ER chaperone expression. They explicitly state that "Atf6a and Atf6b are required to induce CRT expression". However, it remains unclear to me why in this case would Atf6a deletion not impair Crt expression? The authors address this by invoking a mechanism whereby hippocampal neurons are more reliant on Atf6b for Crt expression, but this doesn't really make sense to me. Ultimately, this point underscores the lack of clear mechanistic basis to explain how Atf6b selectively regulates Crt in the hippocampus. This needs to be better resolved through more experimentation. For example, a ChIP experiment monitoring the binding of ATF6b and ATF6a to the Crt promoter in hippocampal and control cells would go a long way towards addressing this issue.

Comment #2. The importance of ATF6b for protecting against insults needs to be better described. For example, the authors should show that overexpression of ATF6b protects against ER stress induced neuronal toxicity in cell culture and in vivo kainate induced neuronal toxicity. Similarly, the authors should evaluate how overexpression of ATF6a protects against these insults to better define the specific dependence of hippocampal neurons on ATF6b. The authors do show that overexpression of ATF6b can rescue the reduced Crt observed in Atf6b-deleted neurons, but the protection should similarly be demonstrated.

Comment #3. Similar to #2, the authors should show that the potential for ATF6b (and ATF6a) overexpression to protect against different insults is impaired in Crt+/- neurons. The authors demonstrate that Crt-depletion increases sensitivity to toxic insults. This would go a long way to demonstrate the importance of the proposed ATF6b-CRT signaling axis in regulating neuronal survival in response to pathologic insults.

Comment #4. When reporting the RNAseq data, the authors should use the q-value (i.e., FDR) instead of the p-value. This will likely affect the number of genes reported in Table 1, but it is the appropriate statistical test for this type of data.

Significance

This manuscript provides new context for understanding the functional relationship between Atf6a and the less-studied Atf6b in regulating neuronal survival. As with other studies focused on the relationship between these two ATF6 isoforms, this study demonstrates that these transcriptional programs integrate to coordinate a tissue-specific response to ER stress. Intriguingly, these studies indicate that ATF6b has a specific role in regulating the ER lectin chaperone CRT and that this ATF6b-CRT axis uniquely regulates neuronal survival in response to ER stress. While additional experiments are required to support this claim, the work described herein is a nice addition to our evolving understanding of the importance of ATF6b in regulating ER and cellular physiology during pathologic insults.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #1

Evidence, reproducibility and clarity

Summary

Unfolded Protein Response (UPR) refers to homeostatic signaling pathways that play protective roles in various cell types. This work by Nguyen et al focuses on the UPR-mediator ATF6. In mammals, there are two isoforms of ATF6, alpha and beta. Nguyen et al show that the expression of the ATF6beta isoform is higher in hippocampal neurons whereas the ATF6alpha isoform is more evenly distributed across various neuronal subtypes. By performing gene expression profiling in mouse brain samples, they identify the ER chaperone calreticulin (CRT) as being significantly downregulated in ATF6beta null mutants. They further validate this observation by comparing hippocampi from ATF6alpha and ATF6beta null mice, where CRT is lowered in the latter but not the former. They identify and mutate putative ER stress response elements (ERSE) in the CRT promoter region to show that expression of CRT can be regulated by both ATF6alpha and beta. They demonstrate that treatment of hippocampal neurons with ER stress inducing chemicals leads to induction of CRT, which is suppressed in ATF6beta mutants. Such treatment also leads to cell death, which is exacerbated in ATF6beta mutants but rescued by ectopic expression of CRT. They also extend these observations to cell death induced by treatment with the glutamate receptor agonist, kainate, which was also exacerbated in ATF6beta mutants, but was rescued by counter treatment with ER stress inhibitors. Together, their data suggest a protective role for ATF6beta in hippocampal neurons in the context of ER stress.

Major comments

The primary advantage of this work is that much of it was done in vivo in mice, providing immediate context for the role of ATF6β under physiological conditions. They identify a specific region of the brain that requires ATF6beta. On the other hand, the ATF6-CRT signaling axis reported here had been established previously, and therefore, this study brings limited conceptual advances regarding the signaling mechanism itself (see Significance section below).

Overall, the authors' data support their primary claim that ATF6β has a neuroprotective role in the context of ER stress. The data presented are clear and convincing, and their methods appear rigorous. The manuscript could be further improved if the authors could provide sufficient rationale for some of their experiments, which are discussed below.

1. The post-translational processing of ATF6beta must be demonstrated in hippocampal neurons and not in HEK293T cells in Figure 1E. The authors conclude on Page 6, line 18 that "these results suggest that ATF6beta functions in neurons" but it is not obvious how expression in HEK293T cells contributes to this conclusion in any way.
2. The hippocampal neurons are affected by the loss of ATF6β, even though the mice are not exposed to tunicamycin. Could the authors present evidence that there is physiological ER stress in hippocampal neurons? If not, why is ATF6beta required.
3. In Figure 3, is there a specific reason why the authors do not mutate the ERSEs in the mouse CRT reporter, pCC1 and instead opt to analyze the huCRT reporter? Given that all the other observations in the manuscript are in mouse calreticulin, it is important to show that the ERSEs in the mouse calreticulin promoter are also regulated in an ATF6beta-dependent manner. Similar to the huCRT reporter, it is also crucial to examine if ATF6beta can regulate the mouse CRT promoter. This would provide an explanation for why calreticulin expression is not completely abolished in ATF6beta mutants.
4. In Figure 5A and B, the density of Tubulin staining varies from panel to panel, and is much lower in ATF6beta mutants treated with Tg/Tm. Presumably this is because of cell death but this should be clarified in the main text. Additionally, it is unclear if the EthD-1 staining is nuclear localized. It would help if single channel images for Hoechst and EthD-1 were provided to visualize this.
5. The literature reports that BAPTA-AM treatment itself could cause ER stress (e.g. PMID: 12531184). Here, the authors report the opposite effect. How could the authors reconcile the difference? The effects of BAPTA-AM and 2-APB must individually be examined in Figure 6C and not just in combination with Tm.
6. The authors allude to "impairment of Ca2+ homeostasis in ATF6beta mutants" in Page 13 Line 2, but do not show any direct evidence in support of it. While treatment with BAPTA-AM and 2-APB is a start in that direction, it certainly does not demonstrate that under homeostatic conditions in vivo or in vitro there is any change in calcium flux in ATF6beta hippocampal neurons. To make the case that there is indeed perturbation of Ca2+ in ATF6beta mutant hippocampal neurons, the authors need to examine calcium flux and measure calcium indicators and how they are affected when ER stress is induced in these mutant cells.
7. The effect of 2-APB and salubrinal alone on hippocampal neurons need to be examined in Figure 9B-D to eliminate the possibility that these drugs are not enhancing cell survival under normal conditions in a parallel manner.
8. The rationale for the examination of Fos, Fosb and Bdnf is poorly described (page 14, line 13) and the conclusions from this line of experimentation are rather weak. The results from Figure 9 to some extent serve to confirm in vivo the data seen in Figure 6C but by no means provide a mechanism for why ATF6beta mutants have perturbed calcium homeostasis (page 14, line 22).

Minor comments

1. Page 8, line 3: Their rationale for why ATF6beta 5'UTR sequences are seen in their RNA seq data is not clearly explained. This must be rewritten for clarity.
2. Page 8, line 5, the authors write that besides Atf6β , CRT was the only UPR-regulated gene downregulated in Atf6β mutant mice. The authors need to state how they defined "UPR-regulated genes". There must be a list, which the authors do not cite.
3. Page 9, line 10: A reference is required for ERSEs.
4. Page 10, line 6: The authors say "ATF6beta specifically induces CRT promoter activity". This is a confusing statement because "induction" is in response to stress, but the context here is homeostatic regulation since there is ostensibly no stress being induced. This distinction should be made and corrected here and throughout the manuscript.
5. Page 10, line 16: The use of "latter" here is confusing and it would help to restructure this sentence for clarity.
6. Figure 9A is missing Y-axis labels.

Significance

The authors summarize their major findings of the study (at the beginning of the Discussion) as ATF6β being required for CRT induction in the hippocampus, and that this ATF6β -CRT axis is important for the survival of hippocampal neurons. The idea that ATF6 induces CRT had been previously shown by others (PMID 9837962), and therefore, this is not the major new discovery of this study. In addition, the ATF6-calreticulin axis having a cell protective role had been reported in other biological contexts (e.g. PMID: 32905769), so that concept is also not a novel concept presented in this work. Similarly, the role of UPR in glutamate receptor agonist-induced neuronal cell death had been shown previously (the authors cite Kitao e tal., 2001; Sokka et al., 2007; Kezuka et al., 2016), so this link is not the major novel discovery revealed by this study. Instead, this study reports that ATF6β KO mice have specific phenotypes in hippocampal neurons, which had not been reported previously. Furthermore, this manuscript reports detailed information regarding Atf6β's downstream target genes in this tissue. In summary, this study's finding that ATF6 regulates CRT is confirmatory, rather than bringing new conceptual advances. The merit of this study is in the identification of the hippocampus as the organ that specifically requires ATF6beta. While the findings here may not appeal to a broader audience interested in UPR signaling mechanisms, it may draw interest from those who study hippocampal neuron physiology.

For the editor's reference, this reviewer's field of expertise is in UPR signaling mechanisms in animal models

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #2

Evidence, reproducibility and clarity

Nguyen and colleagues provide evidence that ATF6-beta selectively induces calreticulin expression in mouse hippocampal neurons to protect these neurons from ER stress-inducing toxins. This is a well-written and well-organized report that provides functional information about ATF6-beta, a poorly studied homolog of the ATF6-alpha Unfolded Protein Response regulator. The report suggests that ATF6-beta has a previously unknown and important function in helping brain neurons survive ER stress by regulating calreticulin.

The study shows that addition of BAPTA, 2-APB, or salubrinal significantly improves neuronal survival in ATF6-/- explants and mice brains in response to ER toxins. But, prior study (PMID: 15705855) used salubrinal at much higher concentration 75uM with little effect at the 5uM dose used in the current study. Evidence should be provided that these drugs are specifically inhibiting ER stress or off-target mechanisms should be discussed in their experimental models.

Minor comments:

Fig 1 any male vs female mice differences in ATF6b expression?

Fig 2C. Please show molecular weight markers on blots

Fig 2C. what are the doublet bands on calnexin?

Fig 3. what are the ERSE sequences? several different binding sites are reported in literature.

p8. What is meant by 5' Atf6b lacks 10 and 11?

Discussion: Please clarify if anti-ATF6-beta antibodies were available for these studies.

Discussion: It is puzzling that ATF6a induces calreticulin more potently than ATF6b, but the calreticulin defect is selectively dependent on ATF6b. Could authors speculate on this paradox? It would be interesting to expand on differences between ATF6a and ATF6b function and phenotypes in Discussion in mouse and in people.

Significance

ATF6-beta is homolog of ATF6-alpha and assumed to function like ATF6-alpha. This report describes a selective function of ATF6-beta in inducing calreticulin in mouse neurons during ER stress. This suggests ATF6-beta has some different functions than ATF6-alpha in the mouse hippocampal neurons.

Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Reply to the reviewers

RESPONSE TO REVIEWERS

We thank Review Commons and its three reviewers. Reviewers 2 and 3 provide detailed comments, which we address individually. Reviewer 1, however, gives a general critique of how we have approached asking how genome architecture affects the extent of evolution and the details of evolutionary trajectories. Our interpretation of their comments is that our approach and the one that they advocate represent two philosophically different, but complementary, views about how to study evolution in the laboratory. We begin by discussing this difference and then proceed to a point by point response to the three reviews.

Reviewer 1

Philosophical differences with Reviewer 1

We interpret Reviewer 1’s comments as endorsing a formal, quantitative study of evolution that aims to explain the factors that control the rate at which fitness increases during experimental evolution. This approach derives from classical population genetics and aims to use a mixture of theory and experiment to uncover general principles that would allow rates of evolution and evolutionary trajectories, expressed as population fitness over time, to be predicted from quantitative parameters, such population sizes, mutation rates, distributions of the fitness effects of mutations (including their degree of dominance in diploids), and global descriptions of either general (e.g. diminishing returns) or allele-specific epistasis.

This approach aims to predict how the average fitness trajectory should be affected by variations in these parameters and describe the variation, at the level of fitness, in the outcomes in a set of parallel experiments. This is an important approach and have previously used it to investigate how the strength of selection influences the advantage of mutators (Thompson, Desai, & Murray, 2006) and to produce and test theory that predicts how mutation rate and population size control the rate of evolution (Desai, Fisher, & Murray, 2007). Like every approach to evolution, this one has limitations: 1) if it doesn’t identify mutations or investigate phenotype other than fitness, it cannot reveal the biological and biochemical basis of adaptation or report on how variations in population genetic parameters (population size, haploids versus diploids, etc.) influence which genes acquire adaptive mutations, and 2) if the details of experiments (e.g. whether populations are clonal or contain standing variation, or which phenotypes are being selected for) have strong effects on the population genetic parameters, these must be measured before theoretical or empirical relationships could be used to predict the mean and variance of fitness trajectories produced by a given selection. A variety of evidence suggests that the second limitation is real. Examples include the absence of a universal finding that diploid populations evolve more slowly than haploids (discussed on Lines 437-442), even within the same experimental organism, and the finding that diminishing returns epistasis applies well to domesticated yeast evolving in a variety of laboratory environments (e. g. papers from the Desai lab, starting with (Kryazhimskiy, Rice, Jerison, & Desai, 2014) but not to the evolutionary repair experiments that we have conducted (Fumasoni & Murray, 2020; Hsieh, Makrantoni, Robertson, Marston, & Murray, 2020; Laan, Koschwanez, & Murray, 2015).

The second approach to experimental evolution, which we, as molecular geneticists and cell biologists, predominantly take, is to follow the molecular and cell biological details of how organisms adapt to selective pressure. We subject organisms to defined selective forces, identify candidate causative mutations, test them by reconstructing the evolved mutations, individually and in combination, and perform additional experiments to ask how these mutations are increasing fitness. Because these experiments are performed on model organisms and often address phenotypes that have been studied by classical and molecular genetics, we can often say a good deal about the cell biological and biochemical mechanisms that increase fitness and this work can complement and extend what we know from classical and molecular genetics.

The current manuscript and its predecessor are examples of finding causative mutations and asking how they improve fitness, with the first paper (Fumasoni & Murray, 2020) demonstrating how mutations in three functional modules could overcome most of the fitness cost of removing an important but non-essential protein and the current paper asking how alterations in genome architecture and dynamics (diploidy and eliminating double-strand break-dependent recombination) affect the extent to which populations increase in fitness and which genes and functional modules acquire mutations as they do so.

By definition, such experiments are anecdotal: they report on how particular genotypes and genome architectures respond to particular selection pressures. Any individual set of experiments can produce conclusions about the effects of variables, such a population size, mutation rate, and genome architecture, on the mutations that increased fitness in response to the specific selection, but they can do more than lead to speculation and inference about what would happen in other experiments: speculation from the results of a single project and inferences from the combined results of multiple projects. Our interpretation is that the evolutionary repair experiments that we have performed, which have perturbed budding, DNA replication, and the linkage between sister chromatids do indeed lead to a common set of inferences: most of the selected mutations reduce or eliminate the function of genes, the interactions between the selected mutations are primarily additive, and the mutations cluster in a few functional modules.

We believe that the population and molecular genetic approaches to experimental evolution are complementary and that a full understanding of evolution will require combining both of them. We think this will be especially true as we try to use the findings from laboratory studies to improve our understanding of evolution outside the lab, which takes place over longer periods, in more temporally and spatially variable environments, and is subject to variation in multiple population genetic and biological parameters.

Reviewer #1 (Evidence, reproducibility and clarity (Required)): In their previous work the authors examined adaptation in response to replication stress in haploid yeast, via experimental evolution of batch cultures followed by sequencing. Here they extend this approach to include diploid and recombination-deficient strains to explore the role of genome architecture in evolution under replication stress. On the whole, a common set of functional modules are found to evolve under all genetic architectures. The authors discuss the molecular details of adaptation and use their findings to speculate on the determinants of adaptation rate.

**SECTION A - Evidence, reproducibility and clarity** Experimental evolution can reveal adaptive pathways, but there are some challenges when applying this approach to compare genetic backgrounds or environments. They key challenge is that adaptation potentially depends on both the rate of mutation and the nature of selection. Distinct adaptation patterns between groups could therefore reflect differential mutation, selection, or both. The authors allude to this dichotomy but have very limited data to address it. The closest effort is engineering putatively-adaptive variants into all genetic background including those where they did not arise; the fact that such variants remain beneficial suggests they did not arise in certain backgrounds because of a lower mutation rate, but this is a difficult issue to tackle quantitatively.

We agree, wholeheartedly, that adaptation depends on the combination of mutation rates and the nature of selection and our goal was to ask how the molecular nature of adaptation depends on genome architecture when three different architectures are subjected to the same selection: constitutive replication stress caused by removing an important component of replisome. We used a haploid strain as a baseline and compared it to two other strains chosen to influence either the effect of mutations (a diploid, where fully recessive mutations that were beneficial in the haploid would become neutral) or the rate of mutations (a recombination-defective strain that would be unable to use ectopic recombination to amplify segments of the genome). In both cases, we expected to see effects that are closer to qualitative than quantitative: the absence of fully recessive mutations in evolved diploids and absence of segmental amplification in the recombination-deficient haploid. We see both effects and they then allow us to ask two other questions: 1) does influencing the effect of a class of mutation (diploids) or preventing a class of mutation (recombination defect) have a major effect on the rate of evolution, and 2) do these differences affect which modules adaptive mutations occur in. As far as we can tell, the answer is no to both questions. We use “as far as we can tell” because our experiments do have limitations. First, the recombination-defective strain has a higher point mutation rate making it impossible to tell how much this elevation, rather than any other factor, accounts for it showing a greater fitness increase than the recombination-proficient haploid. Unfortunately, to our knowledge, it’s impossible to abolish recombination without affecting mutation rates. Second, we only experimentally tested a subset of the inferred causative mutations meaning that for many genes, our assertion that they are adaptive is a statistical inference and their assignment to a particular functional module is based on prior literature rather than our own experiments. In response to this criticism, we have now rephrased some of our sentences (see below).

From mutation accumulation experiments, where the influence of selection is minimized, there is evidence that genetic architecture affects the rate and spectrum of spontaneous mutations. In this experiment, the allele used to eliminate recombination, rad52, will also increase the mutation rate generally. The diploid strain is also likely to have a distinct mutational profile--as a null expectation diploids should have twice the mutation rate of haploids. Recent evidence indicates the mutation rate difference between haploid and diploid yeast might be less than two-fold, but that there are additional differences in the mutation spectrum, including rates of structural change. The context for this study is therefore three genetic architectures likely to differ in multiple dimensions of their mutation profiles, but mutation rates are not measured directly.

The reviewer is correct that we did not explicitly measure mutation rates, although the frequency of synonymous mutations (Figure 3-S1B) is a proxy for the point mutation rate as long as the majority of these mutations are assumed to be neutral. By this measure, the mutation rates for ctf4∆ haploids and ct4∆/ctf4∆ diploids, expressed per haploid genome, are close to each other (1.94 for haploids and 1.37 for diploids) but different enough to return p = 0.044 by Welch’s test, whereas the mutation rate for the recombination-deficient, ctf4∆ rad52∆ haploid is 4 to 5-fold higher (7.03). In contrast, we can infer that the ctf4∆ rad52∆ strain has much lower rates of segmental aneuploidy produced by recombination: we see only one such event in this strain in contrast to 16 in the ctf4∆ haploid and 44 in the ctf4∆/ctf4∆ diploid (Supplementary table 4), even though the amplification of the cohesin loader gene, SCC2, confers similar benefits in all three strains.

The nature of selection on haploids and diploids is expected to differ because of dominance, but ploidy-specific selection is also possible. The authors discuss how recessive beneficial alleles may be less available to diploids, though this can be offset by relatively rapid loss of heterozygosity. However, diploids should also incur more mutations, all else being equal. The rate of beneficial mutation, as opposed to the rate of mutation generally, will depend on the mutational "target size" of fitness, and the authors findings recapitulate other literature (particularly regarding "compensatory" adaptation) that points to faster adaptation in genotypes with lower starting fitness.

We agree with the reviewer and tried to make the point that which mutations are fixed is primarily determined by the product of the rate at which they occur and the benefit which they confer (lines 193-196). Evidence in budding yeast suggests that in diploid cells, removing one copy of most genes fails to produce a measurable fitness benefit (Deutschbauer et al., 2005), suggesting that losing one copy of many genesis purely recessive. If this was always the case, it would be very hard for such heterozygous, loss-of-function mutations to contribute to evolution in diploids: a mutation that inactivates one copy of a gene would have to rise to high enough frequency by genetic drift that homozygosis of this mutation mitotic recombination would have a significant probability. Instead we find that heterozygous mutations in some genes (inactivation of RAD9, what are likely to be hypomorphic mutations in SLD5) but not others (inactivation of IXR1) confer benefits in diploids that allow their frequency to rise much more rapidly by selection than they would by drift, allowing them to reach frequencies at which mitotic recombination becomes probable.

There is ample literature on the above topics, particularly discussions of the evolutionary advantages of haploidy versus diploidy. While adaptation to replication stress provides a novel starting point for this investigation, much of the manuscript is devoted to long-standing questions that are not specific to replication stress. Unfortunately, the data the authors collected is not sufficient to shed light on these questions, because mutation and selection cannot be effectively distinguished. The Discussion states that "We find that the genes that acquire adaptive mutations, the frequency at which they are mutated, and the frequency at which these mutations are selected all differ between architectures but that mutations that confer strong benefits always lie in the same three modules" (line 379), but it is not clear that these statements are all supported by the data.

The reviewer makes two points: we fail to make a significant contribution to long-standing questions about the evolutionary genetics of adaptation and the we make statements that are not supported by our data. On the first we disagree: unlike much of the previous work which compares the effects of mutation rates and population sizes on the rates of evolution, we sequence genomes, identify putative causative mutations, verify that they increase fitness, and test, by reconstruction, how their contribution to fitness is affected by fully characterized genome architectures. We know of no comparable work and we believe that this is a useful contribution to understanding evolution. In addition, some of the literature, for example the discussion of haploidy versus diploidy, has failed to reach a universal conclusion. On the second point, we realized that the statement that the reviewer quotes is stronger than it should be since we do not show “that mutations that confer strong benefits always lie in the same three modules”. What we do show is that mutations in all three modules are found in all three genome architectures (Figure 5), and that combining one mutation from each module (using mutations in genes that are found in that architecture) can reproduce the observed fitness increase in each architecture (Figure 6 B), but the reviewer is correct that we have not demonstrated that every clone from every population has an adaptive mutation in all three modules. We have therefore modified the quoted sentence as follows (altered wording underlined)

"We find that the genes that acquire adaptive mutations, the frequency at which they are mutated, and the frequency at which these mutations are selected all differ between architectures but that mutations conferring strong benefits can occur in all three modules in each architecture" (Lines 405-408)

Focusing on the more novel aspect of their experiment-the presence of replication stress-would arguably be a better approach. On this topic the authors have some interesting observations and speculation, but clear predictions are lacking. The introduction section could be redesigned to explicitly state why genome architecture might affect adaptation in response to replication stress in particular, rather than (or in addition to) adaptation generally. If there were no differences in mutation, does the nature of Ctf4 lead to predictions that the molecular basis of compensatory adaptation should differ among genome architectures? Without such predictions it will be difficult for readers to know whether the observation that different genome architectures follow similar adaptive paths is surprising or not.

We believe that following this suggestion would diminish the paper. We set out to ask how genome architecture affected adaptation to the strong fitness defect produced by removing an important component of an essential process, DNA replication. We chose replication stress as an example of cell biological damage that cells would have to repair with the hope that the results would give general clues about evolutionary repair, rather than hoping that the experiment would inform us about how replication stress altered the types of mutation (e. g. point mutations versus segmental amplification) that were selected As we point out at the beginning of our response, we recognize that the result of any one such experiment must be anecdotal and any attempt to generalize must be described as speculation if it refers only to this one experiment, or inference if it refers to this experiment and other published work. In those cases where we discuss the effect of genome architecture on evolutionary trajectories, we can draw conclusions that apply to our own experiments, but can only speculate on adaptation to different selections. In others, where we see commonalities between our experiments and previous work on evolutionary repair (cite Review), we can make inferences about evolution to adapt to removing important proteins and speculate about other forms of selection. We have revised the discussion to make it clear where we conclude, where we speculate, and where we infer. We suspect that our finding that genome architecture has a larger effect on which genes acquire adaptive mutations than it does on which modules these mutations alter will generalize to other evolutionary repair experiments and may be true even more broadly.

We deliberately did not make predictions about the effect of genome architecture on the rate at which population fitness increased or the mechanism of adaptation to replication stress because we believed that our ignorance and the diverging results of previous experiments was sufficient to make both exercises worthless. After the fact, we interpret our results to suggest that mutations that reduce the activity of components, such as Sld5, that are stably associated with replication forks should be semi-dominant, but we were not nearly smart enough to make such a specific prediction before the experiment began!

**Minor comments:** Shifts in ploidy from diploid to haploid are less common than the reverse change, so the observation of such a shift (Fig. 1) should be discussed in more detail.

We now mention that haploids becoming diploids is more common than the reverse transformation and point out that genome sequencing reveals that these strains are true haploids rather than aneuploids.

“One diploid population (EVO14) gave rise to a population with a haploid genome content, suggesting a possible haploidization event during evolution. Sequencing revealed no aneuploidies as a potential explanation of this phenomenon. While diploidization has been recurrently observed during experimental evolution with budding yeast (Aleeza C. Gerstein & Otto, 2011; Aleeza C Gerstein, Chun, Grant, & Otto, 2006; Harari, Ram, Rappoport, Hadany, & Kupiec, 2018; Venkataram et al., 2016), reports of spontaneous haploidization events have been instead scarce. Given the difficulties introduced by the change of ploidy over the 1000 generations, we have excluded EVO14 from all our analyses.” (Lines 122-128)

We believe that the most likely mechanism is that the strain sporulated to produce haploids that were fitter than their diploid parent, but because this event occurred in only one out of eight populations and the proposed explanation is pure speculation we have not included in the revised manuscript.

Line 88 typo 'stains'.

Fixed. Thank you.

Reviewer #1 (Significance (Required)): **SECTION B - Significance** The novel aspect of this study is the combination of replication stress and genome architecture, but here the significance is limited by a lack of clear predictions on how these factors might interact. On the other hand, much of the manuscript is devoted to why adaptation might vary among genome architectures in general, but this long-standing and important question is not particularly well resolved by this experimental approach, which can't disentangle mutation and selection.

Our belief is that quantitatively predicting how selection will change fitness is nearly impossible because we lack the detailed knowledge of population genetic parameters that apply to our experiments. Prediction is even harder if the goal is to identify which genes will fix adaptive mutations and understand how these mutations alter cellular phenotypes to increase fitness. Thus our approach is almost entirely empirical: we do experiments that alter interesting variables, collect data, and do our best to interpret them and suggest how the conclusions of individual experiments might generalize.

The authors highlight the dichotomy when discussing the evolution of ploidy: "We suggest that... genome architecture affects two aspects of the mutations that produce adaptation: the frequency at which they occur and the selective advantage they confer" (line 399), but presenting this as a novel inference does not appropriately acknowledge prior research and discussion of these ideas; several relevant papers are cited by the authors in other contexts. It may be possible to recast these findings as a test of the role of genome architecture in adaptation generally, but the authors should clarify the limitations of experimental evolution and more fully consider the theory and data outlined in previous research. In particular, few studies can claim to directly compare mutation rates between genome architectures, and it is not obvious that the present study is an example of such.

We have the disadvantage that the reviewer doesn’t identify the literature we fail to cite. To us the argument the reviewer quotes is self-evident. As we mention above, our goal was not to test either general or detailed predictions and the level at which we analyzed our experiment, especially demonstrating that mutations were causal and reconstructing them individually and in combination, is missing from previous work. Finally measuring mutation rates is supremely difficult: you either need good ways of following all possible forms of mutation, quantitatively and without selection, or you resort to selecting mutations with a particular phenotype and molecularly characterizing them, knowing that these assays may well give different ratios of the rates of different types of mutation at different loci. We do make and report one measure of mutation rate, the rate of synonymous mutation in protein coding genes, which we discuss above.

Reviewer expertise: Evoutionary genetics; experimental evolution; mutation.

Reviewer #2 (Evidence, reproducibility and clarity (Required)): **Summary** This manuscript investigates the effect of an organism's genotype (or, as the authors call it, an organism's 'genome architecture') on evolutionary trajectories. For this, the authors use Saccharomyces cerevisiae strains that experience some form of replication stress due to specific gene deletions, and that further differ in ploidy and/or the type of gene(s) deleted. They find the same three functional modules (DNA replication, DNA damage checkpoint, sister chromatid cohesion) are affected across the 3 different genotypes tested; although the specific genes that are mutated varies. **Major comments** This is a solid and exceptionally eloquent paper, comprising a large body of work that is in general well presented. That said, I do have some suggestions and questions. At several points in the manuscript, the authors should perhaps be more careful in their wording and avoid to overgeneralize data without providing additional evidence for these claims.

We thank the reviewer for their constructive review and address their request for more careful wording below.

• Some key points of the study are not entirely clear to me; possibly because the study builds upon a previous study that was recently published in eLife. Anyhow, I think it would be useful to clarify the following points a bit more:

  • Why exactly was ctf4∆ chosen as a model for replication stress? What is the evidence that ctf4∆ is a good model for replication stress? Without including some evidence for this, it is unclear how well the findings in this study really can be generalized to replication stress (which is what the authors do now).

We described the reasons for choosing CTF4 deletion to mimic DNA replication stress in our previous eLife paper, to which we refer at. Nevertheless, the reviewer is right in asking us not to assume that the reader will have read our previous work. Briefly: DNA replication stress is a term that is loosely defined as the combination of the defects in DNA metabolism and the cellular response to these defects in cells whose replication has been substantially perturbed (Macheret & Halazonetis, 2015). Established methods in the field to induce DNA replication stress consist of either pharmacological treatments or genetic perturbation. Pharmacological treatments include hydroxyurea, which target the ribonucleotide reductase and hence stalls forks as a result of dNTP depletion (Crabbé et al., 2010), or aphidicolin, which directly inhibits polymerases α, ε and δ (Vesela, Chroma, Turi, & Mistrik, 2017b; Wilhelm et al., 2019). For genetic perturbation, the conditional depletion of replicative polymerases (Zheng, Zhang, Wu, Mieczkowski, & Petes, 2016) is frequently used. These methods are incompatible with experimental evolution, as cells can mutate the targets of replication inhibitors or alter the expression of genes that have been reduced in expression or activity. Removing an important but non-essential component of the replication machinery avoids these problems. We chose CTF4 deletion as a manipulation that affected the coordination of events at the replication fork: in the absence of Ctf4, the polα-primase complex is no longer physically bound to the replicative helicase, and thus the polymerase’s abundance at the replisome decreases (Tanaka et al., 2009). This manipulation achieves the same effects as polymerase depletion and replisome stalling, producing a constitutive DNA replication stress that can only be overcome by mutations in other genes. Multiple studies have shown that ctf4**D cells display replication intermediates commonly associated to DNA replication stress, such as the accumulation of ssDNA gaps and reversed forks (Abe et al., 2018; Fumasoni, Zwicky, Vanoli, Lopes, & Branzei, 2015), fork stalling (Fumasoni & Murray, 2020), checkpoint activation (Poli et al., 2012; Tanaka et al., 2009) and altered chromosome metabolism (Kouprina et al., 1992).

We now justify our choice of deleting CTF4 at line 74:

“DNA replication stress is often induced with drugs or by reducing the level of DNA polymerases (Crabbé et al., 2010; Vesela, Chroma, Turi, & Mistrik, 2017a; Wilhelm et al., 2019; Zheng et al., 2016). To avoid evolving drug resistance or increased polymerase expression, which would rapidly overcome DNA replication stress,** we deleted the CTF4 gene, which encodes a non-essential subunit of the DNA replication machinery (the replisome) (Kouprina NYu, Pashina, Nikolaishwili, Tsouladze, & Larionov, 1988). Ctf4 is a homo-trimer that functions as a structural hub within the replisome (Villa et al., 2016; Yuan et al., 2019) by binding to the replicative DNA helicase, primase (the enzyme that makes the RNA primers that initiate DNA replication), and other accessory factors (Gambus et al., 2009; Samora et al., 2016; Simon et al., 2014; Villa et al., 2016). In the absence of Ctf4, the Pol**a-primase and other lagging strand processing factors are poorly recruited to the replisome (Samora et al., 2016; Tanaka et al., 2009; Villa et al., 2016), causing several characteristic features of DNA replication stress, such as accumulation of single strand DNA (ssDNA) gaps (Abe et al., 2018; Fumasoni et al., 2015), reversed and stalled forks (Fumasoni & Murray, 2020; Fumasoni et al., 2015), cell cycle checkpoint activation (Poli et al., 2012; Tanaka et al., 2009) and altered chromosome metabolism (Hanna, Kroll, Lundblad, & Spencer, 2001; Kouprina et al., 1992). As a consequence of these defects, ctf4**D cells have substantially reduced reproductive fitness (Fumasoni & Murray, 2020).**”

Would the authors expect to see similar routes of adaptation if a 'genomic architecture' with a less severe/other replication defect would have been used? I realize the last question is perhaps difficult to address without actually doing the experiment (which I am not suggesting the authors should do); I just want to point out that perhaps some data should not be over-generalized.

We share the reviewer’s interest in asking whether different forms of DNA replication stress would lead to the same results described, and we plan to rigorously investigate this question in a separate paper. We note that the careful comparison between different forms of DNA replication stress has never been made and that authors studying this phenomenon often rely on a single perturbation to induce DNA replication stress (Crabbé et al., 2010; Wilhelm et al., 2019; Zheng et al., 2016). We agree that such a comparison will be useful, but we believe (as indicated by the reviewer) it will require an amount of work that goes beyond the scope of our study. To avoid over-generalization, we are using now using “a form of DNA replication stress” in lines 33, 244, 401, 414 and 461, to make it clear that our conclusions (as opposed to inferences and speculations) are restricted to the response to a single example of replication stress.

Likewise, why was RAD52 selected as the gene to delete to affect homologous recombination? I understand that it is a key gene, but on the flipside, absence of RAD52 affects multiple cellular pathways and (as the authors also observe in their populations) also results in increased mutation rates which might confound some of the results.

We aimed to observe the largest deficiency in DNA recombination possible and therefore chose to delete RAD52 because of its many roles in different forms of homologous recombination (Pâques & Haber, 1999) . The choice of other genes, such as RAD51, would have inhibited canonical double strand break (DSB) repair, but allowed other mechanisms that can rescue stalled replication forks (Ait Saada, Lambert, & Carr, 2018), such as break induced replication (BIR) or single strand annealing (SSA) (Ira & Haber, 2002).

Our position regarding the inevitable increase in mutations rates obtained while working with genome maintenance process has been instead elaborated in response to reviewer #1 above.

A sentence describing our choice to delete RAD52 has now been included at line 86:

“…as well as from haploids impaired in homologous recombination due to the deletion of RAD52 (Figure 1A), which encodes a conserved enzyme required for pairing homologous DNA sequences during recombination (Pâques & Haber, 1999). Because Rad52 is involved in different forms of homologous recombination, it’s absence produces the most severe recombination defects and thus allows us to achieve the largest recombination defect achievable with a single gene deletion (Symington, 2002)..”

Related to the first comment, it is also unclear to me how well the system chosen by the authors is representative of the replication stress experienced by tumor cells (as briefly touched upon in the final section of the discussion). Are some of the homologs key oncogenes that drive carcinogenesis?

We should have been clearer. Our goal was to argue that the lesions and responses produced by replication stress in tumor cells, such as stalled replication forks and checkpoint activation, were similar to those seen in yeast cells lacking Ctf4. We did not mean to imply removing Ctf4 from yeast cells had the same effects on cell proliferation and survival as inactivating tumor suppressors and activating proto-oncogenes have in mammalian cells. Despite the difference between direct (removing Ctf4) and indirect effects on DNA replication (tumor cells), the replication intermediates (ssDNA, stalled and reversed forks), the cell cycle defects (G2/M delay), the genetic instability (increased mutagenesis and chromosome loss) and chromosome dynamics (late replication zones and chromosome bridges) generated by the absence of Ctf4 are similar to those observed in oncogene-induced DNA replication stress in mammalian cells (Kotsantis, Petermann, & Boulton, 2018). We therefore believe our experiments reveal evolutionary responses to a constitutive DNA replication stress that resembles the replication stress seen in cancer cells. Nevertheless, we agree that the comparison with cancer evolution remains speculative and we therefore avoided mentioning cancer in the title our paper or our conclusions, and only discuss it in a speculative section of the discussion.

We have modified this section of the discussion as follows (line 554):

“While generated through a different mechanism (unrestrained proliferation, rather than replisome perturbation), oncogene induced DNA replication stress produces cellular consequences (Kotsantis et al., 2018) which are remarkably similar to those seen in the absence of Ctf4, such as the accumulation of ssDNA, stalled and reversed forks (Abe et al., 2018; Fumasoni & Murray, 2020; Fumasoni et al., 2015), genetic instability (Fumasoni et al., 2015; Hanna et al., 2001; Kouprina et al., 1992) and DNA damage response activation (Poli et al., 2012; Tanaka et al., 2009). Based on these similarities we speculate that evolutionary adaptation to DNA replication stress could reduce its negative effects on cellular fitness and thus assist tumor evolution.”

The authors should consider rephrasing some sentences regarding the occurrence of adaptive mutations. Sentences such as 'which genes are mutated depends on the selective advantage' (p1; lines 15-16); 'genome architecture controls the frequency at which mutations occur' (p15), "genome architecture controls which genes are mutated" (p1, line 20) makes it sound like the initial occurrence of mutations is not random, whereas in reality, the mutational landscape is the result of the combined effect of occurrence and fitness effect of the mutations, with the later rather than the former likely being the main driver behind the observed patterns.

We thank the reviewer for asking for more precision in the above sentences, whose proposed changes we now list:

“Mutations in individual genes are selected at different frequencies in different architectures, but the benefits these mutations confer are similar in all three architectures, and combinations of these mutations reproduce the fitness gains of evolved populations.” (Lines 13-15)

“Genome architecture influences the distribution of adaptive mutants” (Line 277)

"genome architecture influences the frequency at which mutations occur, the fitness benefit they confer, and the extent of overall adaptation." (Lines 462-463)

Some important methodological information is missing or unclear in the manuscript:

The authors should provide more details on how they decided which clones to select for sequencing. Did they select the biggest colonies; were colonies picked randomly, ...

This following sentence is now reported in the materials and methods section (Line 603)

“To capture the within-population genetic variability we selected the clones displaying the largest divergence of phenotypes in terms of resistance to genotoxic agents (methyl-methanesulfonate, hydroxyurea and camptothecin).”

What is the population size during the evolution experiment?

We now added the following sentence at line 599:

“In this regime, the effective population size is calculated as N0 x g where N0 is the size of the population bottleneck at transfer and g is the number of generations achieved during a batch growth cycle and corresponds to approximately to 107 cells.”

Sequencing of populations and clones: coverage should be mentioned

The following sentence has now been added at line 616:

“Clones and populations were sequenced at approximately the following depths: 25-30X for haploid clones, 50-60X for diploid clones, 50-60X for haploid populations and 120-130X for diploid populations.”

Identification of mutations (p19, line 573): Is this really how the authors defined whether a variant is a mutation? Based on the definition given here, DNA mutations that lead to a synonymous mutation in the protein are not considered as mutations?

We apologize for this typo. We do identify and consider synonymous mutations as evidenced by Figure 3-S1B. Now the sentence at line 626 correctly reports:

“A variant that occurs between the ancestor and an evolved strain is labeled as a mutation if it either (1) causes a substitution in a coding sequence or (2) occurs in a regulatory region, defined as the 500 bp upstream and downstream of the coding sequence.”

Perhaps the information can be found elsewhere, but the source data excel files for mutations is incomplete and should at the very least contain information on the type of mutation (eg. T->A), as well as the location of this mutation in the respective gene.

Perhaps the reviewer is referring to Supplementary table 2, where we list the number of times a gene has been mutated in different populations (and thus summaries different types of mutations affecting the same gene). The information they request is reported in Supplementary table 1 for all the variants detected in populations and clones sequencing.

**Minor comments** • While the author already cite several significant papers relevant for their manuscript, some other studies could also be included:

• On how genome architecture influences evolution: https://doi.org/10.1093/molbev/msaa172
• Ploidy and fitness/adaptation: DOI:https://doi.org/10.1016/j.cub.2018.01.062

We thank the reviewer for highlighting these references, which are now cited at line 28

From the text in the abstract, it is unclear what the three genomic architectures (line 13) exactly are, the authors should consider spelling this out.

In repose o the reviewer request for clarity we now propose the following change in line 13:

“We asked how these trajectories depend on a population’s genome architecture by comparing the adaptation of haploids to that diploids and recombination deficient haploids.” (Lines 9-11)

Can the authors speculate on why a homozygous ctf4D/ctf4D rad52D/rad52D would be lethal, and a haploid not?

See below

The authors note that a diploid ctf4D/ctf4D strain is less fit than its haploid counterpart. Why do the authors think this is the case?

In response to the two previous questions, we now propose the following speculations that we include in the text (Line 97):

“Diploid cells require twice as many forks as haploids and Ctf4-deficient diploids are thus more likely to have forks that cause severe cell-cycle delays or cell lethality. We speculate that this increased probability explains the more prominent fitness defect displayed by diploid cells. Interestingly, homologs of Ctf4 are absent in prokaryotes, where the primase is physically linked to the replicative helicase (Lu, Ratnakar, Mohanty, & Bastia, 1996) and Ctf4 is essential in the cells of eukaryotes with larger genomes such as chickens (Abe et al., 2018) and humans (Yoshizawa-Sugata & Masai, 2009). Rad52 is likely involved in rescuing stalled replication forks by recombination-dependent mechanisms (Fumasoni et al., 2015; Yeeles, Poli, Marians, & Pasero, 2013). We speculate that the absence of Rad52 increases the duration of these stalls and leads some of them to become double-stranded breaks resulting in cell lethality and explaining the decreased fitness of ctf4D rad52D haploid double mutants. In diploids ctf4D rad52D cells, which have twice as many chromosomes, the number of irreparably stalled fork may be sufficient to kill most of the cells in a population, thus explaining the unviability of the strain.”

The authors passage their cells for 100 cycles and assume that this corresponds to around 1000 generations for each population. However, the fitness differences between the different starting strains (see also Figure 1B) are likely to cause considerable differences in number of generations between the different strains. Do the authors have more precise measurements of number of generations per population? If not, perhaps it should be noted that some lineages may have undergone more doublings than others, and perhaps also discuss if and how this could influence the results?

In a batch culture regime, where populations are allowed to reach saturation after each dilution, the number of generations at each passage are dictated by the dilution factor (Van den Bergh, Swings, Fauvart, & Michiels, 2018). A dilution of 1:1000 from a saturated culture will allow for approximately 10 generations before populations reach a new saturated phase. As long as saturation is allowed to occur, this number is independent of the fitness of the cultured strains: Slower-dividing strains will simply employ more time to reach saturation after each dilution. At the beginning of the experiment, we had to dilute the ctf4D rad52**D strains being passaged every 48hrs instead of 24hrs. After generation 50, ctf4D rad52**D strains reached saturation within 24hrs and were then diluted daily. The total count considers the number of passages a culture has undergone, and not the number of days of culture, and thus should guarantee approximately the same number of generations in all three genome architectures.

Panel A of figure 1A is somewhat confusing; as this seems to indicate that the ctf4∆ was introduced after strains were made, for example, haploid recombination deficient (which is not how these strains were constructed). Perhaps a better way of representing would be to have the indication of DNA replication stress pictured inside the yeast cells.

We have modified Figure 1A to better represent the way the strains were constructed. For space reasons we have not represented a perturbed fork within each cell, but rather above all of them.

Legend to Figure 1: is fitness expressed relative to haploid or diploid WT cells for the diploid strains?

We apologize for having missed this detail in the figure legends. Throughout the figures, haploid and diploid cells were competed against reference strains with the same ploidy. We now add this sentence in Figure 1 and in the materials and methods (line 686).

Figure 3: to improve readability of this figure, the authors could consider placing the legend of the different symbols (#, *,..) in the figure as well and not just in the figure legend.

We now include the symbols legend in Figure 3.

Figure 5 shows Indels, but if I am correct, these mutations are not discussed in the text; nor is it mentioned what the authors used as a cut-off to determine indels (the authors use the term 'small indels' without defining it)? For example, the data shown in Figure 3 and Figure 4 only includes SNPs and not indels (correct?) - but the indels should also be taken into account when investigating which modules are hit.

Gapped alignments of the relatively long 150 paired-end reads in our data set permits the identification of small indels ranging in size from 1–55 bp using VarScan pileup2indels tool (Koboldt et al., 2012). All small indels (and the respective sequence affected) are listed together with SNPs in Supplementary table 1. Figure 3A, Figure 4 and Figure 5B are representation of ‘gene mutations’ which include both SNPs and small InDels. Large chromosomal Insertion and deletions, not detectable by short read gap alignment are instead identified using the VarScan pileup2copynumber tool (Koboldt et al., 2012), and are represented as amplifications or deletions in Figure 3B and 5C.

The following sentence has been added to the material and methods at line 629:

“Gapped alignments of the 150 paired-end reads in our data set permits the identification of small indels ranging in size from 1–55 bp using VarScan pileup2indels tool (Koboldt et al., 2012). All small indels (and the respective sequence affected) are listed together with SNPs in Supplementary table 1.”

The following definition has been added in Figure legends 3A, 4 and 5A and B.

“Gene mutations (SNPs and small InDels 1-55bp)”

Figure 5 mentions: # gene mutations. So these are only the mutations in genes, and not in their up- or downstream regulatory regions?

We use a broader definition of a gene, not restricted to the open reading frame, and including its regulatory regions. The following definition has been added to figure 5’s legend.

“Frequency of SNPs and small InDels (1-55bp) affecting genes (Open reading frames and associated regulatory regions).”

Figure 3-S1: labels of C panels are missing.

Labels are now included in Figure 3-S1

Figure 3-S1, panel B: why did the authors focus on synonymous mutations?

The panel B is commented upon in line 186 and contrasted with panel A to argue that the increased number of mutations detected in ctf4∆ rad52∆ strains is due to a higher mutation rate(which is expected to increase synonymous mutations) instead of an higher number of adaptive mutations (which are less likely to be synonymous) being selected.

Reviewer #2 (Significance (Required)): This is a solid and clearly written study, comprising a large body of work that is generally well presented and that will be of interest to scientists active in the field of (experimental) evolution and replication. However, many aspects studied in this manuscript have already been studied and reported before; including the recent eLife paper by the same group, as well as studies by other labs that have investigated how genome architecture / genotype affects evolutionary trajectories, the effect of ploidy on evolution, .... Because of this, I do feel that the authors should put their findings more in the context of existing literature context, including a general description of which results are truly novel, which confirm previous findings and which results seem to go against previous reports. This is already so at some points in the text, but I feel this could be done even more.

We now rephrase the following paragraphs in our discussion to better highlight the main conclusions in contrast to the existing literature:

“Engineering one mutation in each module into an ancestral strain lacking Ctf4 is enough to produce the evolved fitness increase in all three genomic architectures. Furthermore, engineering mutations in individual genes confer benefits in all three architectures (Fig. 6A) ,even in those where the mutations in these genes was rare, and combining these mutations recapitulated the evolved fitness increase in all three architectures (Fig. 6B). Altogether our results demonstrate the existence of a common pathway for yeast cells to adapt to a form of constitutive DNA replication stress.” (Lines 409-414)

“Our results thus go against the trend of slower adaptation in diploids as compared to haploids reported by the majority other studies (A. C. Gerstein, Cleathero, Mandegar, & Otto, 2011; Marad, Buskirk, & Lang, 2018; Zeyl, Vanderford, & Carter, 2003). This effect is not limited to populations experiencing DNA replication stress (Figure 2A) but is also present in control wild-type populations (Figure 2B). Our results support the idea that the details of genotypes, selections, and experimental protocols can determine the effect of ploidy on adaptation.” (Lines 437-442)

“Our results therefore agree with previous reports observing declining adaptability across strains with different initial fitness but largely fail to observe diminishing return epistasis as a potential justification of this phenomenon. Our experiments and two previous evolutionary repair experiments (Hsieh et al., 2020; Laan et al., 2015) both show interactions that are approximately additive between different selected mutations. The reasons for this difference are currently unknown.” (Lines 450-455)

Additionally, I think the authors should be more careful not to over-generalize their findings, which come from only a few specific genetic manipulations that might not be representative for general replication stress. For example (p15), can the authors really claim that they have unraveled general principles of adaptation to constitutive DNA replication stress? Perhaps a better motivation of the choice of ctf4 as a model mutation for DNA replication stress could also help (see also my earlier comments). A similar comment applies to the molecular mechanisms affecting adaptation in diploid cells - what evidence do the authors have that their findings are not specific to the one specific type of diploid strain they used in their study? Adding a bit more background information or nuance for some of the claims would help tackle this issue.

We now followed the suggestions made previously by the reviewer to justify our experimental choices better and to use a language that avoids over-generalizations.

Field of expertise of this reviewer: genetics, evolution, genomics

Reviewer #3 (Evidence, reproducibility and clarity (Required)): **Summary:** Here the authors carry out an evolution experiment, propagating replicate populations of the budding yeast with the CTF 4 gene deleted in three different genetic backgrounds: haploid , diploid and recombination deficient (RAD52 deletion). The authors find that the rate of evolution depends on the initial fitness of the different genetic backgrounds which is consistent with a repeated finding of evolution experiments: that beneficial mutations tend to have a smaller fitness effect in high fitness genetic backgrounds. Curiously even though the targets of selection tended to be specific to each of the three different genetic backgrounds, genetic reconstruction experiments showed beneficial mutations convert a fitness increase in all genetics backgrounds. The authors go on to provide a plausible explanation for why each of the three genetic backgrounds are predisposed to certain types of beneficial mutations. Overall, these results provide important context and caveats for an emerging consensus that genetic background determines the rate of evolution, a comprehensive molecular breakdown of adaptation to DNA replication stress and a mechanistic explanation for why different beneficial mutations are favoured in diploids, haploids and recombination deficient strains. This is a well-executed study that is beautifully presented and easy to follow. This will be of great interest to those in the experimental evolution community and the data an excellent resource.

We thank reviewer #3 for emphasizing that reconstructed mutations are beneficial even in architectures where they were not ultimately detected at the end of the experiment. We have now highlighted this point in our conclusions as a response to the reviewer’s #1 and #2 request for more clarity regarding our novel findings.

“We find that the genes that acquire adaptive mutations, the frequency at which they are mutated, and the frequency at which these mutations are selected all differ between architectures but that mutations that confer strong benefits can occur in all three modules in each architecture. Engineering one mutation in each module into an ancestral strain lacking Ctf4 is enough to produce the evolved fitness increase in all three genomic architectures. Furthermore, reconstruction of a panel of mutations into all three architectures proved they are adaptive even in architectures where the affected genes were not found significantly mutated by the end of the experiment. Altogether our results demonstrate the existence of a common pathway for yeast cells to adapt to a form of constitutive DNA replication stress.” (Lines 405-414)

**Major comments:**

• Are the key conclusions convincing? Yes, the convergent evolution analysis, fitness assays, and genetic reconstructions are sufficient to characterise the genetic causes of adaptation in this experiment, and are of the highest standard. The authors do particularly well to fully recover the fitness increases that evolved with their genetic reconstructions, which imparts a completeness to their understanding of what happened in their evolution experiment.
• Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? No, in nearly all cases the authors make reasonable claims. One exception is on L419 in the discussion, where the authors speculate why some mutations do not follow diminishing returns epistasis, but this idea does not really have any basis (no citation or reasons to suggest that DNA repair genes are less connected with other genes in the genome). If the authors cannot support this statement, it should be removed, and instead write that is currently unknown why some individual mutations do not follow the pattern of diminishing returns.

On reflection, we agree with the reviewer and now state,

“Our results confirm previous reports observing declining adaptability across strains with different initial fitness but largely fail to observe diminishing return epistasis as a potential justification of this phenomenon. Our experiments and two previous evolutionary repair experiments (Hsieh et al., 2020; Laan et al., 2015) both show interactions that are approximately additive between different selected mutations. The reasons for this difference are currently unknown.”

A hypothesis, which would need experimental validation, could be that the different mutations have different degrees of epistatic interactions with the rest of the genome. Ixr1, whose mutation follows diminishing return epistasis, is a transcription factor that could in principle affect the expression of many other genes implicated in different cellular modules. Sld5, Scc2 and Rad9 instead, whose mutations have the same effect across different genome architectures, having more mechanistic roles in genome maintenance may have strong epistatic interactions only with a restricted number of cellular modules implicated with DNA metabolism.

• Would additional experiments be essential to support the claims of the paper? No.
  • Are the data and the methods presented in such a way that they can be reproduced? Yes, but some more details are needed for the convergent evolution analysis, see minor comments.
  • Are the experiments adequately replicated and statistical analysis adequate? Yes, but some more statistic reporting in the main text or figure legends would be helpful, for example. L159: Please report the statistical test, test statistic and p value in the text or in the figure legend. Currently significance is indicated, but the methods do not specify the test.

We apologize for the lack of clarity in the main text. The test used for all fitness analysis was only reported in the materials and methods as follow:

“The P-values reported in figures are the result of t-tests assuming unequal variances (Welch’s test)”

We now include the test and the associated p-value in line 184, and write the above sentence in all the relevant figures.

This should also be done for the GO analysis shown in figure 3A.

We thank reviewer #3 pointing out this omission. We now include the following section:

“Gene ontology (GO) enrichment analysis:

The list of genes with putatively selected mutations (Figure 3A) or homozygous mutations in diploids (Figure 4) were input as ‘multiple proteins’ in the STRING database, which reports on the network of interactions between the input genes (https://string-db.org). The GO term enrichment analysis provided by STRING are reported in Supplementary Table 3 and Supplementary Table 6 respectively. Briefly, the strength of the enrichment is calculated as Log10(O/E), where O is the number of ‘observed’ genes in the provided list (of length N) which belong to the GO-term, and E is the number of ‘expected’ genes we would expect to find matching the GO-term providing a list of the same length N made of randomly picked genes. P-values are computed using a Hypergeometric test and corrected for multiple testing using the Benjamini-Hochberg procedure. The resulting P-values are represented as ‘False discovery rate’ in the supplementary tables and describe the significance of the GO terms enrichment (Franceschini et al., 2013).”

**Minor comments:**

• Specific experimental issues that are easily addressable. Not a new experiment, but extra details are required. The authors carried out both clone and whole population sequencing. For their convergent evolution analysis, what is the criteria for a mutation to be included- ie, does it need to be fixed, have attained a certain frequency? This is important- if the criteria were low (say 5%), it would be important to know whether gene A had fixed in 4 populations, while gene B had attained a frequency of 10% in 5 populations. As it stands both would be described as examples of convergent evolution. This can be handled by providing these details in the methods.

For the population sequencing we disregarded variants found at less than 25% and 35% of the reads in haploid and diploid populations respectively as we observed they were largely the product of alignment errors. All the variants found at frequencies higher than the thresholds indicated were used for the parallel evolution analysis. The frequency at which each individual variant was detected in each population is reported in Supplementary table 1, while the average frequency at which a gene has been found mutated across different populations is reported in Supplementary table 2. The reason why we didn’t solely focus on fixed mutations for our convergent evolution analysis was that from previous work we knew of the existence of clonal interference which kept the frequency of verified adaptive mutations that coexisted in the same population (e.g. ixr1 and sld5) well below 90% (Fumasoni & Murray, 2020).

For clarity we now add the following sentence in the material and methods:

“Variants found in less than 25% and 35% of the reads in haploid and diploid populations respectively were discarded, since many of these corresponded to misalignment of repeated regions. For clone sequencing, only variants found in more than 75% of the reads in haploids and 35% of the reads in diploids (to account for heterozygosity) were considered mutations. The frequency of the reads associated with all the variants detected are reported in Supplementary table 1”

• Are prior studies referenced appropriately? I note that the authors use the term declining adaptability where as other papers use the term diminishing returns epistasis- I am sure the authors have good reasons for their choice of nomenclature but I think it would be helpful for their readers to connect this work to other work by mentioning that declining adaptability is also referred to as diminishing returns.

We use both terms (for instance in line 446 and line 448) with a different meaning : By ‘declining adaptability’ we refer the phenomenon where more fit strains display lower adaptation rates than less fit ones. By ‘diminishing returns epistasis’ we refer to a possible explanation of such a phenomenon, where adaptive mutations have different fitness effects due to their ‘global’ epistatic interactions with other alleles. It has to be noted that ‘diminishing returns epistasis’ is not the only proposed explanation of the phenomenon of declining adaptability (Couce & Tenaillon, 2015). In our case, we do find evidence of declining adaptability but very limited evidence for diminishing return epistasis (only 1 mutation in 5 has a different fitness effect in different architectures).

A reference the authors have missed: L419, as well as citing the Desai Lab bioxive paper, they should cite another theory paper that obtained similar conclusions. Lyons, D.M., et al. https://doi.org/10.1038/s41559-020-01286-y.

We thank the reviewer for the suggested reference, which is now cited at line 450.

• Are the text and figures clear and accurate? This paper is beautifully written and easy to follow, a lot of thought has gone into the figures which are aesthetically pleasing and easy to navigate.

  • Do you have suggestions that would help the authors improve the presentation of their data and conclusions? No.

  **Typos**

  L32 "do" should be "to" L95 analyzed L219 are the authors referring to ref 15 here? I think so, but please specify

We thank the reviewer for carefully finding the typos, which are now all corrected.

Reviewer #3 (Significance (Required)):

• Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field. This paper is an important conceptual result and an immediate advance for basic research. The authors have done an outstanding job of showing the potential for the clinical translation of this research, especially regarding cancer biology.
• Place the work in the context of the existing literature (provide references, where appropriate). This study follows up on and builds upon an earlier paper by these same authors published in E-life in 2020. Conceptually this work is most closely related to work in Michael Desai's, Sergey Kryazhimskiy's, Tim Coopers and Chris Marx's labs work looking at diminishing returns epistasis in yeast, and studies contrasting evolution of haploids and diploids led by Greg Lang's and Sarah Otto's labs.
• State what audience might be interested in and influenced by the reported findings. This work will be of great interest to the Experimental evolution and molecular evolution communities and also of interest to those who study cancer genomics and DNA replication and repair.
• Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate. Microbial experimental evolution.

REFERENCES CITED IN THE REVIEW RESPONSE

Abe, T., Kawasumi, R., Giannattasio, M., Dusi, S., Yoshimoto, Y., Miyata, K., … Branzei, D. (2018). AND-1 fork protection function prevents fork resection and is essential for proliferation. Nature Communications, 9(1), 3091. https://doi.org/10.1038/s41467-018-05586-7

Ait Saada, A., Lambert, S. A. E., & Carr, A. M. (2018, November 1). Preserving replication fork integrity and competence via the homologous recombination pathway. DNA Repair. Elsevier B.V. https://doi.org/10.1016/j.dnarep.2018.08.017

Couce, A., & Tenaillon, O. A. (2015). The rule of declining adaptability in microbial evolution experiments. Frontiers in Genetics, 6(MAR), 99. https://doi.org/10.3389/fgene.2015.00099

Crabbé, L., Thomas, A., Pantesco, V., De Vos, J., Pasero, P., & Lengronne, A. (2010). Analysis of replication profiles reveals key role of RFC-Ctf18 in yeast replication stress response. Nature Structural & Molecular Biology, 17(11), 1391–1397. https://doi.org/10.1038/nsmb.1932

Desai, M. M., Fisher, D. S., & Murray, A. W. (2007). The speed of evolution and maintenance of variation in asexual populations. Current Biology : CB, 17(5), 385–394. https://doi.org/10.1016/j.cub.2007.01.072

Deutschbauer, A. M., Jaramillo, D. F., Proctor, M., Kumm, J., Hillenmeyer, M. E., Davis, R. W., … Giaever, G. (2005). Mechanisms of Haploinsufficiency Revealed by Genome-Wide Profiling in Yeast. Genetics, 169(4), 1915–1925. https://doi.org/10.1534/GENETICS.104.036871

Franceschini, A., Szklarczyk, D., Frankild, S., Kuhn, M., Simonovic, M., Roth, A., … Jensen, L. J. (2013). STRING v9.1: Protein-protein interaction networks, with increased coverage and integration. Nucleic Acids Research, 41(D1), D808. https://doi.org/10.1093/nar/gks1094

Fumasoni, M., & Murray, A. W. (2020). The evolutionary plasticity of chromosome metabolism allows adaptation to constitutive DNA replication stress. ELife, 9. https://doi.org/10.7554/eLife.51963

Fumasoni, M., Zwicky, K., Vanoli, F., Lopes, M., & Branzei, D. (2015). Error-Free DNA Damage Tolerance and Sister Chromatid Proximity during DNA Replication Rely on the Polα/Primase/Ctf4 Complex. Molecular Cell, 57(5), 812–823. https://doi.org/10.1016/j.molcel.2014.12.038

Gambus, A., van Deursen, F., Polychronopoulos, D., Foltman, M., Jones, R. C., Edmondson, R. D., … Labib, K. (2009). A key role for Ctf4 in coupling the MCM2-7 helicase to DNA polymerase alpha within the eukaryotic replisome. EMBO J., 28(19), 2992–3004. https://doi.org/10.1038/emboj.2009.226

Gerstein, A. C., Cleathero, L. A., Mandegar, M. A., & Otto, S. P. (2011). Haploids adapt faster than diploids across a range of environments. Journal of Evolutionary Biology, 24(3), 531–540. https://doi.org/10.1111/j.1420-9101.2010.02188.x

Gerstein, Aleeza C., & Otto, S. P. (2011). Cryptic fitness advantage: Diploids invade haploid populations despite lacking any apparent advantage as measured by standard fitness assays. PLoS ONE, 6(12), 26599. https://doi.org/10.1371/journal.pone.0026599

Gerstein, Aleeza C, Chun, H.-J. E., Grant, A., & Otto, S. P. (2006). Genomic Convergence toward Diploidy in Saccharomyces cerevisiae. PLoS Genetics, 2(9), e145. https://doi.org/10.1371/journal.pgen.0020145

Hanna, J. S., Kroll, E. S., Lundblad, V., & Spencer, F. a. (2001). Saccharomyces cerevisiae CTF18 and CTF4 are required for sister chromatid cohesion. Mol Cell Biol., 21(9), 3144–3158. https://doi.org/10.1128/MCB.21.9.3144-3158.2001

Harari, Y., Ram, Y., Rappoport, N., Hadany, L., & Kupiec, M. (2018). Spontaneous Changes in Ploidy Are Common in Yeast. Current Biology, 28(6), 825-835.e4. https://doi.org/10.1016/j.cub.2018.01.062

Hsieh, Y. Y. P., Makrantoni, V., Robertson, D., Marston, A. L., & Murray, A. W. (2020). Evolutionary repair: Changes in multiple functional modules allow meiotic cohesin to support mitosis. PLoS Biology, 18(3), e3000635. https://doi.org/10.1371/journal.pbio.3000635

Ira, G., & Haber, J. E. (2002). Characterization of RAD51-independent break-induced replication that acts preferentially with short homologous sequences. Molecular and Cellular Biology, 22(18), 6384–6392. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/12192038

Koboldt, D. C., Zhang, Q., Larson, D. E., Shen, D., McLellan, M. D., Lin, L., … Wilson, R. K. (2012). VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Research, 22(3), 568–576. https://doi.org/10.1101/gr.129684.111

Kotsantis, P., Petermann, E., & Boulton, S. J. (2018, May 1). Mechanisms of oncogene-induced replication stress: Jigsaw falling into place. Cancer Discovery. American Association for Cancer Research Inc. https://doi.org/10.1158/2159-8290.CD-17-1461

Kouprina, N., Kroll, E., Bannikov, V., Bliskovsky, V., Gizatullin, R., Kirillov, A., … et al. (1992). CTF4 (CHL15) mutants exhibit defective DNA metabolism in the yeast Saccharomyces cerevisiae. Mol Cell Biol., 12(12), 5736–5747. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/1341195

Kouprina NYu, Pashina, O. B., Nikolaishwili, N. T., Tsouladze, A. M., & Larionov, V. L. (1988). Genetic control of chromosome stability in the yeast Saccharomyces cerevisiae. Yeast (Chichester, England), 4(4), 257–269. https://doi.org/10.1002/yea.320040404

Kryazhimskiy, S., Rice, D. P., Jerison, E. R., & Desai, M. M. (2014). Microbial evolution. Global epistasis makes adaptation predictable despite sequence-level stochasticity. Science (New York, N.Y.), 344(6191), 1519–1522. https://doi.org/10.1126/science.1250939

Laan, L., Koschwanez, J. H., & Murray, A. W. (2015). Evolutionary adaptation after crippling cell polarization follows reproducible trajectories. ELife, 4, e09638. https://doi.org/10.7554/eLife.09638

Lu, Y. B., Ratnakar, P. V, Mohanty, B. K., & Bastia, D. (1996). Direct physical interaction between DnaG primase and DnaB helicase of Escherichia coli is necessary for optimal synthesis of primer RNA. Proc. Natl. Acad. Sci. USA, 93(23), 12902–12907. https://doi.org/10.1073/pnas.93.23.12902

Macheret, M., & Halazonetis, T. D. (2015). DNA replication stress as a hallmark of cancer. Annu. Rev. Pathol. Mech. Dis., 10, 425–448. https://doi.org/10.1146/annurev-pathol-012414-040424

Marad, D. A., Buskirk, S. W., & Lang, G. I. (2018). Altered access to beneficial mutations slows adaptation and biases fixed mutations in diploids. Nature Ecology & Evolution, 2(5), 882–889. https://doi.org/10.1038/s41559-018-0503-9

Pâques, F., & Haber, J. E. (1999). Multiple pathways of recombination induced by double-strand breaks in Saccharomyces cerevisiae. Microbiology and Molecular Biology Reviews : MMBR, 63(2), 349–404. https://doi.org/

Poli, J., Tsaponina, O., Crabbé, L., Keszthelyi, A., Pantesco, V., Chabes, A., … Rothstein, R. (2012). dNTP pools determine fork progression and origin usage under replication stress. EMBO J., 31(4), 883–894. https://doi.org/10.1038/emboj.2011.470

Samora, C. P., Saksouk, J., Goswami, P., Wade, B. O., Singleton, M. R., Bates, P. A., … Masai, H. (2016). Ctf4 Links DNA Replication with Sister Chromatid Cohesion Establishment by Recruiting the Chl1 Helicase to the Replisome. Molecular Cell, 63(3), 371–384. https://doi.org/10.1016/j.molcel.2016.05.036

Simon, A. C., Zhou, J. C., Perera, R. L., van Deursen, F., Evrin, C., Ivanova, M. E., … Pellegrini, L. (2014). A Ctf4 trimer couples the CMG helicase to DNA polymerase alpha in the eukaryotic replisome. Nature, 510(7504), 293–297. https://doi.org/10.1038/nature13234

Symington, L. S. (2002). Role of RAD52 Epistasis Group Genes in Homologous Recombination and Double-Strand Break Repair. Microbiology and Molecular Biology Reviews, 66(4), 630–670. https://doi.org/10.1128/mmbr.66.4.630-670.2002

Tanaka, H., Katou, Y., Yagura, M., Saitoh, K., Itoh, T., Araki, H., … Shirahige, K. (2009). Ctf4 coordinates the progression of helicase and DNA polymerase alpha. Genes to Cells, 14(7), 807–820. https://doi.org/10.1111/j.1365-2443.2009.01310.x

Thompson, D. A., Desai, M. M., & Murray, A. W. (2006). Ploidy controls the success of mutators and nature of mutations during budding yeast evolution. Current Biology : CB, 16(16), 1581–1590. https://doi.org/10.1016/j.cub.2006.06.070

Van den Bergh, B., Swings, T., Fauvart, M., & Michiels, J. (2018). Experimental Design, Population Dynamics, and Diversity in Microbial Experimental Evolution. Microbiol Mol Biol Rev., 82(3), e00008-18. https://doi.org/10.1128/MMBR.00008-18

Venkataram, S., Dunn, B., Li, Y., Agarwala, A., Chang, J., Ebel, E. R., … Petrov, D. A. (2016). Development of a Comprehensive Genotype-to-Fitness Map of Adaptation-Driving Mutations in Yeast. Cell, 166(6), 1585-1596.e22. https://doi.org/10.1016/J.CELL.2016.08.002

Vesela, E., Chroma, K., Turi, Z., & Mistrik, M. (2017a). Common Chemical Inductors of Replication Stress: Focus on Cell‐Based Studies. Biomolecules, 7(1), 19. https://doi.org/10.3390/biom7010019

Vesela, E., Chroma, K., Turi, Z., & Mistrik, M. (2017b, February 21). Common chemical inductors of replication stress: Focus on cell-based studies. Biomolecules. MDPI AG. https://doi.org/10.3390/biom7010019

Villa, F., Simon, A. C., Ortiz Bazan, M. A., Kilkenny, M. L., Wirthensohn, D., Wightman, M., … Dutta, A. (2016). Ctf4 Is a Hub in the Eukaryotic Replisome that Links Multiple CIP-Box Proteins to the CMG Helicase. Molecular Cell, 63(3), 385–396. https://doi.org/10.1016/j.molcel.2016.06.009

Wilhelm, T., Olziersky, A. M., Harry, D., De Sousa, F., Vassal, H., Eskat, A., & Meraldi, P. (2019). Mild replication stress causes chromosome mis-segregation via premature centriole disengagement. Nature Communications, 10(1), 1–14. https://doi.org/10.1038/s41467-019-11584-0

Yeeles, J. T., Poli, J., Marians, K. J., & Pasero, P. (2013). Rescuing stalled or damaged replication forks. Cold Spring Harbor Perspectives in Biology, 5(5), a012815. https://doi.org/10.1101/cshperspect.a012815

Yoshizawa-Sugata, N., & Masai, H. (2009). Roles of human AND-1 in chromosome transactions in S phase. J Biol Chem., 284(31), 20718–20728. https://doi.org/10.1074/jbc.M806711200

Yuan, Z., Georgescu, R., Santos, R. de L. A., Zhang, D., Bai, L., Yao, N. Y., … Li, H. (2019). Ctf4 organizes sister replisomes and Pol α into a replication factory. ELife, 8, e47405. https://doi.org/10.7554/eLife.47405

Zeyl, C., Vanderford, T., & Carter, M. (2003). An Evolutionary Advantage of Haploidy in Large Yeast Populations. Science, 299(5606), 555–558. https://doi.org/10.1126/SCIENCE.1078417

Zheng, D.-Q., Zhang, K., Wu, X.-C., Mieczkowski, P. A., & Petes, T. D. (2016). Global analysis of genomic instability caused by DNA replication stress in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A., 113(50), E8114–E8121. https://doi.org/10.1073/pnas.1618129113

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #4

This reviewer did not leave any comments

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

---

Referee #3

Evidence, reproducibility and clarity

Summary:

Here the authors carry out an evolution experiment, propagating replicate populations of the budding yeast with the CTF 4 gene deleted in three different genetic backgrounds: haploid , diploid and recombination deficient (RAD52 deletion). The authors find that the rate of evolution depends on the initial fitness of the different genetic backgrounds which is consistent with a repeated finding of evolution experiments: that beneficial mutations tend to have a smaller fitness effect in high fitness genetic backgrounds. Curiously even though the targets of selection tended to be specific to each of the three different genetic backgrounds, genetic reconstruction experiments showed beneficial mutations convert a fitness increase in all genetics backgrounds. The authors go on to provide a plausible explanation for why each of the three genetic backgrounds are predisposed to certain types of beneficial mutations. Overall, these results provide important context and caveats for an emerging consensus that genetic background determines the rate of evolution, a comprehensive molecular breakdown of adaptation to DNA replication stress and a mechanistic explanation for why different beneficial mutations are favoured in diploids, haploids and recombination deficient strains. This is a well-executed study that is beautifully presented and easy to follow. This will be of great interest to those in the experimental evolution community and the data an excellent resource.

Major comments:

• Are the key conclusions convincing? Yes, the convergent evolution analysis, fitness assays, and genetic reconstructions are sufficient to characterise the genetic causes of adaptation in this experiment, and are of the highest standard. The authors do particularly well to fully recover the fitness increases that evolved with their genetic reconstructions, which imparts a completeness to their understanding of what happened in their evolution experiment.
• Should the authors qualify some of their claims as preliminary or speculative, or remove them altogether? No, in nearly all cases the authors make reasonable claims. One exception is on L419 in the discussion, where the authors speculate why some mutations do not follow diminishing returns epistasis, but this idea does not really have any basis (no citation or reasons to suggest that DNA repair genes are less connected with other genes in the genome). If the authors cannot support this statement, it should be removed, and instead write that is currently unknown why some individual mutations do not follow the pattern of diminishing returns.
• Would additional experiments be essential to support the claims of the paper? No.
• Are the data and the methods presented in such a way that they can be reproduced? Yes, but some more details are needed for the convergent evolution analysis, see minor comments.
• Are the experiments adequately replicated and statistical analysis adequate? Yes, but some more statistic reporting in the main text or figure legends would be helpful, for example. L159: Please report the statistical test, test statistic and p value in the text or in the figure legend. Currently significance is indicated, but the methods do not specify the test. This should also be done for the GO analysis shown in figure 3A.

Minor comments:

• Specific experimental issues that are easily addressable. Not a new experiment, but extra details are required. The authors carried out both clone and whole population sequencing. For their convergent evolution analysis, what is the criteria for a mutation to be included- ie, does it need to be fixed, have attained a certain frequency? This is important- if the criteria were low (say 5%), it would be important to know whether gene A had fixed in 4 populations, while gene B had attained a frequency of 10% in 5 populations. As it stands both would be described as examples of convergent evolution. This can be handled by providing these details in the methods.
• Are prior studies referenced appropriately? I note that the authors use the term declining adaptability where as other papers use the term diminishing returns epistasis- I am sure the authors have good reasons for their choice of nomenclature but I think it would be helpful for their readers to connect this work to other work by mentioning that declining adaptability is also referred to as diminishing returns.

A reference the authors have missed: L419, as well as citing the Desai Lab bioxive paper, they should cite another theory paper that obtained similar conclusions. Lyons, D.M., et al. https://doi.org/10.1038/s41559-020-01286-y. .

• Are the text and figures clear and accurate? This paper is beautifully written and easy to follow, a lot of thought has gone into the figures which are aesthetically pleasing and easy to navigate.
• Do you have suggestions that would help the authors improve the presentation of their data and conclusions? No.

Typos

L32 "do" should be "to" L95 analyzed<br> L219 are the authors referring to ref 15 here? I think so, but please specify

Significance

• Describe the nature and significance of the advance (e.g. conceptual, technical, clinical) for the field. This paper is an important conceptual result and an immediate advance for basic research. The authors have done an outstanding job of showing the potential for the clinical translation of this research, especially regarding cancer biology.
  • Place the work in the context of the existing literature (provide references, where appropriate). This study follows up on and builds upon an earlier paper by these same authors published in E-life in 2020. Conceptually this work is most closely related to work in Michael Desai's, Sergey Kryazhimskiy's, Tim Coopers and Chris Marx's labs work looking at diminishing returns epistasis in yeast, and studies contrasting evolution of haploids and diploids led by Greg Lang's and Sarah Otto's labs.
  • State what audience might be interested in and influenced by the reported findings. This work will be of great interest to the Experimental evolution and molecular evolution communities and also of interest to those who study cancer genomics and DNA replication and repair.
  • Define your field of expertise with a few keywords to help the authors contextualize your point of view. Indicate if there are any parts of the paper that you do not have sufficient expertise to evaluate. Microbial experimental evolution.