On 2021-11-08 16:42:36, user Nicolas wrote:

The current Preprint has been accepted for publication and a revised version can be accessed here: https://doi.org/10.1111/nph...

On 2021-11-06 20:07:56, user Binks Wattenberg wrote:

We find this to be a very exciting and compelling study that establishes that the turnover of the ORMDL proteins is regulated by sphingosine-1-phosphate signaling in vascular endothelial cells. We do, however, have a different model as to the role of this system as a homeostatic mechanism controlling sphingolipid biosynthesis.

We consider the ORMDLs to be regulatory subunits of SPT which, like many regulatory subunits, are not intrinsically inhibitory until they are triggered by a ligand. Our evidence strongly indicates that the ligand for the SPT/ORMDL complex is ceramide. With this in mind, we envision that the S1P regulation of ORMDL stability overlays an acute and direct ORMDL-dependent regulation of SPT by ceramide. In our view, the S1P-dependent stabilization of the ORMDLs maintains them as ceramide-sensitive regulators of SPT. In the absence of S1P signaling, degradation of the ORMDLs renders the SPT complex insensitive to ceramide and therefore strongly increases SPT activity.

Below we outline evidence which brings us to this view. But before doing so, we would like to emphasize one of the exciting and important aspects of the work outlined in this pre-print. Considering that S1P signaling is mediated by the G-protein linked S1P receptors (in this case S1PR1), it is an interesting possibility that other cell types with different requirements for control of sphingolipid biosynthesis will utilize the same downstream signaling, perhaps mediated by other G-protein receptors, to control ORMDL levels. A hint of this is found in the regulation of ORMDL turnover by cholesterol loading reported by Gulshan and colleagues (Autophagy. 2015;11(7):1207-8).

The experiments that underlie our view of that the enzymatic activity of the SPT/ORMDL complex is directly responsive to ceramide levels is as follows. It is important to emphasize that the bulk of these studies were performed in Hela cells. The biochemistry of the SPT/ORMDL complex itself is likely independent of cell type, but additional regulatory mechanisms, such as those presented in this pre-print, are doubtless cell-type dependent:

1. Sphingoid bases do not mediate an acute ORMDL-dependent regulation of SPT. We tested the identity of the sphingolipid that triggers ORMDL inhibition of SPT by feeding Hela cells sphingosine. This results in an almost complete inhibition of SPT. This is consistent with the sphingosine inhibition of SPT activity in intact cells originally reported by Kondrad Sandhoff’s group (Mandon EC, van Echten G, Birk R, Schmidt RR, Sandhoff K.Eur J Biochem. 1991 Jun 15;198(3):667-74). This inhibition is entirely ORMDL-dependent. Importantly, we demonstrated that sphingosine inhibition of SPT was completely blocked by preventing ceramide generation with the ceramide synthase inhibitor Fumonisin B1 (Figure 2 of J Biol Chem. 2012 Nov 23;287(48):40198-204 and Figure 2 of J Biol Chem. 2019 Mar 29;294(13):5146-5156. ). Thus, inhibition by sphingosine requires its conversion to ceramide. S1P generation, enhanced by sphingosine feeding, is not blocked by Fumonisin B1, yet SPT inhibition is ablated. Therefore, in this system, S1P does not appear to have a role in regulating SPT activity in Hela cells under the short-term conditions that we used. We concluded from these experiments that the triggering sphingolipid is ceramide or a complex sphingolipid such as sphingomyelin or a glycosphingolipid, but not a sphingoid base.

2. Elevation of ORMDL levels alone does not lead to SPT inhibition. We have shown that inhibition of SPT by ceramide In Hela cells and human bronchial epithelial cells (HBEC) cannot be explained by increased ORMDL protein expression. We do not observe an increase in ORMDL protein expression in response to C6-ceramide treatment of Hela Cells (Figure 5, Siow et al 2015), under conditions in which SPT activity is strongly inhibited. And we demonstrate that ectopically increasing ORMDL protein expression in either Hela cells or HBEC does not result in inhibition of SPT activity (J Lipid Res. 2015 Apr;56(4):898-908). In both of these cells there is sufficient ORMDL at steady state to serve the needs of SPT regulation, yet ceramide is strongly inhibitory. We make the point in this paper that the stoichiometry of ORMDL to SPT expression is important and make clear that it is likely that in some cell types changes in ORMDL expression will impact on SPT regulation. This is consistent with the response reported in this pre-print in response to S1P signaling.

3. Biochemical reconstitution demonstrates a direct inhibitory effect of ceramide. We have reconstituted ceramide-triggered, ORMDL-dependent inhibition of SPT in isolated membranes in a biochemical assay in which neither protein synthesis, turnover, nor post-translational modifications can occur. We are confident that this biochemical reconstitution reflects a physiological response of the SPT/ORMDL complex to ceramide. We demonstrated that the response to ceramide is strictly stereospecific with respect to ceramide stereoisomers. Only the native, D-erythro ceramide stereoisomer triggers ORMDL-dependent inhibition of SPT. Moreover, we addressed the possibility that the short-chain ceramides that we routinely use (for their solubility properties) might not reflect physiological inhibition. We generated native chain-length ceramide in the isolated membranes using the endogenous ceramide synthases. This ceramide was strongly inhibitory (Figure 2 of J Biol Chem. 2019 Mar 29;294(13):5146-5156).

4. The recently published structures of SPT/ORMDL complexes reinforces the view of the ORMDLs as regulatory subunits. The ORMDLs are firmly embedded in the structure. Moreover, comparison of the structures in the substrate-free and substrate-bound state indicates that the ORMDLs inhibit SPT via an amino-terminal sequence that reversibly inhabits the substrate binding site of SPT. These structures suggest that the inhibitory sequence must be stabilized in the active site of the protein to achieve inhibition. We propose that ceramide binding to the complex accomplishes this stabilization.

Taken together, our data and that presented in this pre-print form the picture that the ORMDLs are involved in multiple levels of regulation of SPT. A direct and rapid inhibition by ceramide, and the loss of that regulation when the ORMDLs are degraded as a result of interrupted S1P signaling. There is doubtless more to come and we look forward to further discoveries illuminating regulation of this essential system and the physiological impact of that regulation.

On 2021-10-28 14:56:26, user USHA MAHAWAR wrote:

Really really interesting paper about the ORMDLs and SPT regulation. I am curious to know that whether half-life of over expressed ORMDL3 is similar to that of endogenous ORMDL3. And does all the three ORMDL isoforms have same or similar half-lives . I know that ORMDL3 is emerging as a center of attraction in the field of SPT regulation studies but question is why do cells have three ORMDL isoforms ( when all the three ORMDL isoforms have more that 90% amino acid sequence similarity).

On 2021-11-06 02:27:49, user Quan Long wrote:

This paper has been accepted for publication by Genetics

On 2021-11-06 01:04:21, user Campbell Lab wrote:

Congratulations on the great work! Inspired by your previous contributions, we attempted something similar a few years ago but never got it to work as well as you did. In our design, the two strands were linked in tandem rather than appended to the ends of the GFP. If you are interested, you can access the published thesis here: https://era.library.ualbert...

On 2021-11-05 20:01:25, user Morgane Boone wrote:

The final peer-reviewed version of this paper was published in Nature Communications (https://doi.org/10.1038/s41....

On 2021-11-05 16:17:02, user Alizée Malnoë wrote:

The manuscript by Seydoux et al. investigates the role of proton potassium antiporter KEA3 in diatoms. The authors first demonstrated the pH dependence on photoprotection, specifically non photochemical quenching (NPQ) and showed that NPQ can be induced in the dark by acidic pH. They found that KEA3 modulates NPQ by impacting the proton motive force (PMF); indeed generated kea3 mutants showed increased partitioning into deltapH. Importantly they showed that diatom KEA3 in contrast to plant KEA3 possesses an EF hand motif which can bind Ca2+ and proposed that it controls KEA3 activity. The role of KEA3 and pH in affecting the NPQ response has been previously shown in other photosynthetic organisms however the novelty of this study lies in the demonstration that NPQ can be induced in the dark by acidic pH and the proposed role of Ca2+ in regulating KEA3 function.

Major comments<br /> - Page 5, you state that pH-induced quenching in the dark was accompanied by the conversion of DD into DT. Please provide de-epoxidation state (DES) at t15 time (Fig. 1B) to substantiate this statement. Starting DES would also be informative to ensure there was no retention of DT/zeaxanthin in the dark. <br /> - Also to ensure there is no sustained NPQ (and/or damage or disconnected antenna) at t0, please provide Fo and Fm levels for all NPQ kinetics experiments. Assessing PSII accumulation by D1 immunoblot could be done to ensure PSII damage does not occur.<br /> - In Fig. 2F, it is not clear which data points represent HL or ML treatment as well as which ones come from light or dark period. Please indicate them in different colors or symbols. Also clarify whether you have averaged data from the kea3 mutant alleles.<br /> - To confirm that lack of complementation by deltaEF is not due to mislocalization, please show whether deltaEF accumulates at the thylakoid membrane.

Minor comments<br /> - Page 3, Introduction, specify qE after NPQ response; PSBS should be written PsbS<br /> - Page 4, DD-dependent NPQ should be DT-dependent<br /> - Page 4, we suggest changing “crucial” to “Given the unknown role” if pH-dependence of NPQ in diatoms hasn’t been fully established before<br /> - Page 8, KEA3 most likely homolog, were there other homologs than the two shown in Fig. S5? also discuss conservation of other ion channels (is Phatr J11843 thylakoid-localised?) and if they could compensate for the absence of KEA3 in KO mutant (by being upregulated for instance).<br /> - Fig2B, comment on the band at ~80kDa in OE, is that from cleavage of GFP?<br /> - Fig2G, shouldn’t you expect a lower dpH in the OE? Please comment.<br /> - Page 13, for the statement that only dpH can modulate NPQ, we would suggest to tone down or specify that this is the assumption made here as it could be that dpsi modulates NPQ but has yet to be shown!<br /> - Most of the protein analyses were performed loading samples based on protein content, when possible please provide proof that chlorophyll levels are comparable between the genotypes (at least for the native gels)<br /> - Abstract, extra ‘of’ between capacity and via; page 23, extra ‘being’ between likely and less important<br /> - Define acronyms when used for the first time<br /> - There is a lot of ‘peculiar’ in the text ;-)<br /> - Fig. 2D, star symbol instead of square symbol, check consistency of symbols

Pushan Bag, Pierrick Bru (Umeå University) - not prompted by a journal; this review was written within a preprint journal club with input from group discussion including Alizée Malnoë, Maria Paola Puggioni, Jingfang Hao, Jack Forsman, Wolfgang Schröder, Emma Cocco, Jianli Duan.

On 2021-11-05 11:28:22, user Ben Dubin-Thaler wrote:

Great work NYC virus hunters!

On 2021-11-04 19:54:56, user Rahul Patharkar wrote:

Is the thesis of this abstract valid? It appears to be a misinterpretation of the point of the Patharkar and Walker 2016 paper and is quite sensationalistic. Everyone in the abscission fields knows that ethylene plays a role in abscission (this has been known for decades). The point of the 2016 paper is that it shows that drought can trigger leaf abscission in Arabidopsis and it defines some of the molecular basis for that. When we first observed that Arabidopsis cauline leaves abscised in our growth chamber, we wanted to know why this was happening because it was not described in the literature. We did not find evidence of ethylene build up in our chamber and we found that exogenous application of ACC did not trigger abscission, rather we found water-deficit triggered the abscission. We make no claim that ethylene is "not involved" in abscission rather it is not the external stimuli that triggers abscission. Regardless of how our words are interpreted, our firm stance is that ethylene "is involved" in abscission as a hormone that positively regulates abscission. We have published multiple review articles that have diagrams that show this point. Here is the actual sentence that from the 2016 paper that uses the word "triggered" rather than "involved": "After sequentially ruling out differences in genotypes, pesticide treatment, and ethylene accumulation, we found leaf abscission can be triggered by withholding water until the plants began to wilt followed by rewatering (Fig. 1)."<br /> Again, our view is that ethylene is a positive regulator of abscission but external stimuli like water-deficit or pathogen attack initiate the cauline leaf abscission pathway.<br /> Sincerely,<br /> Rahul Patharkar

On 2021-11-04 18:51:39, user Run Zhou Ye wrote:

Please cite this article as: R.Z. Ye, C. Noll, G. Richard et al., DeepImageTranslator: A free, user-friendly graphical interface for image translation using deep-learning and its applications in 3D CT image analysis, SLAS Technology, https://doi.org/10.1016/j.s...

On 2021-11-04 18:06:08, user Donald R. Forsdyke wrote:

Macroevolution versus Microevolution

Presumably this paper (1) has been released in preprint form to obtain feedback before formal publication. Coauthored by a consortium of current leaders in the field of population genetics, it states that "the ability to fit the parameters of one's preferred model to data does not alone represent proof of biological reality." They hope fellow practitioners, having been alerted by this "simple truism," will avoid various pitfalls. Apart from concerns on synonymous site neutrality (2), calls to reconsider evolutionary fundamentals (3, 4) are not mentioned.

The historical authority of William Provine is referred to (5). He described the early 20th century dispute between geneticist William Bateson and the "Biometricians" (Pearson and Weldon). While disputing Mendelism, the latter made outstanding contributions to statistics. However, Provine concluded The Origins of Theoretical Population Genetics diffidently: "With the gap between theoretical models and available observational data so large, population genetics began and continues with a theoretical structure containing obvious internal consistencies."

Despite these 1971 words and "the wealth of data" now available, that gap remains. To bridge, the authors appeal to "interdisciplinarity … in order to connect genotype to phenotype" (1). This should remind us that in the 1920s Bateson foresaw (3) "that before any solution is attained, our knowledge of unorganized matter must first be increased." So sadly, regarding his topic, genetics: "For a long time we may have to halt." It was only following great progress in molecular biology, that in the 1970s WWII bomber pilot Richard Grantham, at the Université de Lyon, was able to ask the very question the authors pose (1): "Whether, and if so how, accurate evolutionary inferences can be extracted from DNA sequences sampled from a population?" In the authors' words (1), Grantham was able to use "molecular variation and divergence data to infer evolutionary processes." What Grantham called his "genome hypothesis" (6) was later related to the earlier ideas of Darwin's research associate, George Romanes, and Bateson (3).

While readily adopting Bateson's coinages – homozygote, heterozygote, allelomorph, epistasis, homeotic, meristic – the modern-day biometricians (1) have overlooked the most fundamental of his ideas, the "residue" (3), as they did Grantham's "genome hypothesis" and Romanes' "collective variation." Likewise, to make the mathematics easier, they embraced the neutral ideas of Kimura (2), instead of the "homostability" ideas of his compatriot, Akiyoshi Wada (7), who pressed unsuccessfully for a Japanese "genome project," which would have anticipated by many years that of the USA (8).

The works of Grantham, Romanes and Bateson, together with those of various Russian evolutionists and Richard Goldschmidt, focus on the fundamental distinction between inter-species "macroevolution" and intra-species "microevolution" (7). This crystallized historically in 1990 in the lectures and writings of the Russian specialist, Mark Adams. He stressed that the understanding of macroevolution would demand "a radically new interpretation of the history of Darwinism, population genetics and the evolutionary synthesis." For "if intra- and inter-specific variation differ not in kind, but only in degree, then it is possible, by extension, to envision selection as the creator of a new species. But if varieties are fundamentally different from species – if the fundamental character of intraspecific and interspecific variation is essentially different – then the effect of selection on a population cannot explain evolution." Initially published in French in 1990, Adams' work is now available in English (4).

1. Johri et al. (2021) Statistical inference in population genomics. bioRxiv: doi.org/10.1101/2021.10.27.... Nov 2.
2. Kern AD, Hahn MW (2018) The neutral theory in light of natural selection. Mol Biol Evol 35:1366–1371.
3. Cock AG, Forsdyke DR (2008) Treasure Your Exceptions. The Science and Life of William Bateson. Springer, New York.
4. Adams MB (2021) Little evolution, big evolution. Rethinking the evolution of population genetics. In: Delisle RG (ed), Natural Selection. Revisiting its Explanatory Role in Evolutionary Biology. Springer Nature, Switzerland, pp. 195-230.
5. Provine WB (1971) The Origins of Theoretical Population Genetics. University of Chicago Press.
6. Grantham R, Perrin P, Mouchiroud D (1986) Patterns in codon usage of different kinds of species. Oxford Surveys in Evolutionary Biology 3:4 8-81.
7. Forsdyke DR (2016) Evolutionary Bioinformatics, 3rd edn. Springer, New York.
8. Cyranoski D (2009) Reading, writing and nanofabrication. Nature 460:171-2.

On 2021-11-03 19:53:58, user Raphael Kopan wrote:

The paper is now in print on Nature Communication web site: https://www.nature.com/arti...

Note the review history is included and will explain the evolution of the submitted form in BioRxiv to the published product. We wish to publicly thank the reviewers for greatly improving our manuscript.

On 2021-11-03 14:33:03, user antibodies-online wrote:

ABIN101961 used for CUT&Tag/MulTI-Tag in this preprint

On 2021-11-03 14:11:34, user Raghu Parthasarathy wrote:

Fascinating study, but I think there are several issues of data presentation and analysis that need to be addressed. Figure 2 *really* needs to have a plot of effect sizes, not p-values. Figure 1 needs corrections for multiple comparisons. (I elaborate on these, especially the first, here: https://eighteenthelephant.... ). I'd also like to see a better discussion of robustness to cluster number -- the kinks in Figure S3 are pretty weak.

On 2021-11-03 13:53:58, user Prof. T. K. Wood wrote:

L 59: Both “there is little evidence for a clear genetic basis or the molecular mechanisms involved for the persistence trait” and “…the genetics and mechanisms of persistence are yet unknown” are patently false statements. Persistence is an elegant response to myriad stresses, as shown already through single-cell studies, though not reported here. See doi 10.1016/j.bioflm.2019.100018.

Persistence non-heritable; that is the whole point. There are no genetic changes in persister cells and they don’t require 19 years to form. Resistance arises from mutation and often the mutations are cumulative without noticeable changes in MIC. So authors must use genome sequencing and show no genetic change before calling cells “persisters” as they do in this manuscript.

L 190: how long were the cells treated with Abs prior to plating to measure CFU? Should be at least 3 hr and kill curves should be added to demonstrate these are persister cells; i.e., that prolonged exposure does not lead to a reduction in CFU.

Persistence is no more “a stepping stone to resistance” than any other cell type: persisters wake and when they wake, these non-persister cells mutate like all microorganisms. But dormant cells don’t mutate.

On 2021-11-02 17:55:46, user Nicole Reynolds wrote:

Lichtheimia is not in the Zoopagomycota, it is in the Mucoromycota. Please fix this before publication.

On 2021-11-02 09:56:52, user David Bhella wrote:

To help readers understand the path to publication, I am adding an account of the peer review process to each preprint.

This article was initially rejected without peer-review by PLOS Pathogens. We then submitted to Scientific Reports, where the paper was accepted following review:

Reviewer comments:

Reviewer #1 (Technical Comments to the Author):

In this manuscript, Ho et al. reported a 7-Å resolution cryoEM reconstruction model of MrNV VLP expressed in insect cells. MrNV could cause white tail disease in the giant freshwater prawn with high mortality rate, therefore is a serious threat to aquaculture. Together with PvNV infecting marine shrimp, MrNV may represent a new genus in the Nodaviridae family. The structure presented here shows a different arrangement of protruding spikes on the icosahedral capsid surface, compared to other nodaviruses, supporting this classification. The most significant difference is that the protrusions are dimeric, instead of trimeric as in other nodaviruses.

This manuscript is well written. The methodology from VLP expression, purification, to imaging and 3D reconstruction is standard and clearly explained. The conclusions are logical based on the results. Some discussions could be better elaborated:

1.The authors devoted a lot of space (especially figures) to the homology modeling which did not provide much information besides that the P domain of MrNV capsid protein is different from the input homologous models. It would be more helpful to instead show figures of the models fitted in the MrNV map, to directly show the discrepancies and suggest possible location of the MrNV P domain.

2.Given the current information, there is not sufficient evidence to say whether the fuzzy density beneath 5-fold symmetry axis is RNA. The authors could discuss the possibility of it being protein, such as the N-terminal region of capsid, which is usually disordered in other nodaviral structures.

3.Literature (ref. 14 &15) has shown two different assembly states of MrNV VLP expressed in E. coli and sf9 cells respectively. Could the structural information reported here help to explain the differences?

4.Structural characterization of MrNV is in need due to the threat from white tail disease. Now with the 7-Å resolution available, the authors could discuss more about followup studies and/or downstream applications leading to potential intervention against white tail disease.

Some minor points:

1.Has the final map been deposited to the EMDataBank?

2.With the current figures, the comparison between AB and CC dimers is a little hard to follow. It would help to label the A, B, C subunits. It is fine to label the dimers with colored arrows, but it would be more clear if the coloring is consistent between Figures 2 and 3. Please also consider including the measurements of angles and lengths in the figures, and labeling the supporting legs of CC dimer with an arrow or asterisk.

Reviewer #2 (Technical Comments to the Author):

The authors present work showing a cryo-EM 3D reconstruction of MrNV virus-like particles with the finding that “pronounced dimeric blade-shaped spikes" protruding above the surface of the particle are arranged differently than canonical structures of Alphanodaviruses. Thus the authors believe the new structure supports the prior assertion that MrNV belongs to a new genus of Nodaviridae designated Gammanadovirus.

The authors use a generally accepted approach during the reconstruction process although the use of a crystal structure as an initial model rather than using an initial model generated from their experimental 2D class averages could possibly confound the interpretation. Whenever a known structure is used it can lead to potential model bias. It is this reviewer’s assumption that the authors used FHV for the initial model since FHV doesn’t have significant spikes on the surface. The authors also used a low-pass filter of 60 angstroms to the FHV initial model to partially mitigate model bias. In both of these cases this is typically an ok approach if significant homology exists. However the authors force icosahedral symmetry during reconstruction and they themselves highlight the fact that MrNV and FHV share only 20% homology. The manuscript could therefore be greatly strengthened by a reference-free 3D reconstruction where the initial model is created from the experimental 2D class averages rather than the FHV crystal structure. If the final reconstruction for the reference-free approach remains similar/identical to the current reconstruction, then the authors will have demonstrated conclusively that the interpretation is sound. Therefore it is suggested that the authors incorporate the results of a reference-free reconstruction into the manuscript (a supplemental figure will be fine). As this requires a rerun of only the 3D refinement image processing step and not new data acquisition, this should not be considered a major modification and if this is successfully implemented then this reviewer recommends publication.

A few other minor comments to be addressed:

According to Reference #9 (NaveenKumar et al. 2013) the capsid protein of MrNV and PvNV only share 44.6% homology but that drops to 22% for the last 115 amino acids at the C-terminus which is the region the author attribute to forming the protruding spikes. Thus, it seems possible that the structure of PvNV may be different. It is this reviewer’s suggestion that the authors refrain from extending their interpretation towards PvNV and simply focus on MrNV throughout the manuscript.

Please define “VLPs” as “virus-like particles” in the abstract rather than just using the acronym.

There appears to be a 6xHis-tag on the capsid protein but it is not used for purification scheme. A sentence should be added to describe why it is included and whether the additional amino acids are anticipated to be present within the dimeric spikes or otherwise impact the interpretation.

During the post-processing steps, a b-factor of -890 square angstroms was applied. Was this calculated automatically using Relion or was it manually chosen?

Figure 1, it would be helpful to see a sampling of the refined 2D class averages in addition to the central slice of the reconstruction.

On line 120, suggest deleting “sharply resolved” to leave sentence as “Inspection of figure 1(b) reveals a capsid shell measuring between 2 and…” since “sharply resolved” is a qualitative term that others may feel is only appropriate for truly atomic resolution structures.

Finally, the homology modelling is an interesting addition to the paper. However, since no conclusive results can really be drawn from the models at this time, it seems more appropriate for figure 4 to move to a supplemental figure.

On 2021-11-02 07:58:51, user David Bhella wrote:

To help readers understand the process of peer-review, I am adding the peer-reviewer comments and article submission history for all of my preprints.

This article describes one of the most significant discoveries of my career and I started out sending a presubmission enquiry to Nature, where it was taken forward for review.

Review round 1<br /> Referee #1 (Remarks to the Author):

This work by Conley et al. addresses a long-time enigma in the calicivirus field: elucidating the unknown function of the small basic protein, VP2, encoded near the 3’-end of every genome. This protein had no detectable sequence homology with other known viral or eukaryotic proteins and was highly variable in sequence and mass among members of the Caliciviridae genera. In a stunning piece of cryo-EM structural detective work enabled by knowledge of the feline calicivirus receptor (fJAM-A) and capsid architecture, the authors pursued the identity of an unexpected density that appeared only when the FCV capsid was bound with its cognate receptor ectodomain. This density was shown to be a dodecameric arrangement of VP2 to create a unique structure at one 3-fold symmetry axis of each virus particle. Evidence was presented that this structure was involved in the creation of a portal assembly at the unique 3-fold axis for release of the RNA genome from the virion into the cell (likely from within an endosomal compartment). The paper is well-written and the figures and movies are of excellent quality. This original work should create interesting new avenues of research, not only in the calicivirus field, but potentially for other viruses as well.

1. This discovery is exciting, and deserves further discussion in context of the published literature. Perhaps there is a stringent word limit, but, if possible, please elaborate briefly on a couple points. Direct VP1 and VP2 interactions had been reported in the literature (e.g., Kaiser et al., JGV, 2006 and Vongpunsawad et al., JVI, 2013), and the VP2 was proposed to play a role in viral assembly and stability. Are the interaction sites mapped in previous studies consistent with the new model and how might VP2 enhance stability of the particle? In addition, it is noteworthy that the virion-incorporated VP2 (now 12 copies) is more abundant in the particle than previous stoichiometric studies suggested (such as the approx. 1.5 copies per norovirus VLP of Glass et al., JVI 2000 or the 1-2 copies per FCV virion of Sosnovtsev and Green, Virology 2000). One early study of FCV could not detect the ORF3 product in virions at all (Tohya et al., JVM, 1999). Does the new model provide a possible explanation for this discrepancy?

The putative metal binding site in the VP1 is new and of interest. Will this be explored further with crystallography and metal soaking experiments? Is mutagenesis of the putative potassium binding site lethal for F9?

1. The time of incubation at each of the conditions shown in Figures S13 and S15 would be helpful to include in the figure legend. Please use an arrow or two in Figure S15 to show a representative ball of density said to likely reflect the viral RNA.

2. Are the six tyrosine residues that line the capsid pore conserved in all FCV strains? Other caliciviruses? Can VP2 proteins among the other calicivirus genera be modeled to interact in a similar dodecameric arrangement?

3. The VPg is not mentioned in the model of viral entry. Was anything learned about the orientation of the viral genome as it exits the portal, and can it be predicted? There has been no reported evidence that VP2 binds RNA. How might interaction with the genome occur?

4. The first demonstration of VP2 as a minor calicivirus structural protein was shown with RHDV by Wirblich, Thiel and Meyers (JVI, 1996). Perhaps this work should be cited along with Glass et al. (reference 9).

Referee #2 (Remarks to the Author):<br /> This is a potentially interesting paper that provides structural evidence for the existence of a 12-fold symmetric tube, similar to a portal protein, in caliciviruses. My fundamental problem with this work is that the identification of a tube that ‘looks’ like a portal protein is solely based on the structural similarity with the barrel domain of P22 portal protein (by the way this work is not even referenced!).

There are no ‘hard’ data supporting the various speculative hypotheses that VP2 functions in DNA ejection(like a portal protein) and that VP2 hydrophobic N- termini insert into the endosomal membrane forming a channel (like protein H in phiX174).

In essence, this paper is speculative, preliminary and stuffed with hypotheses that could be tested experimentally. At the current stage, this work is of medium impact and should find publication in a specialized journal like the Journal of Virology, Virology, etc.

Specific comments

• Figure 1. This figure could go to the Supplementary Information or, at least, be deemphasized.

• Figure 2. Panel b and c are redundant and provide the same information. Instead this figure lacks a clear description of the putative portal protein. All I see is a 12-fold symmetric tube of density that looks like the barrel domain of P22 portal protein (e.g. again, the authors don’t even reference the various papers that describe P22 portal protein and its barrel domain). B

• Figure 2a. The tube of density in panel b is after applying averaging. It's not clear if the authors used 3-fold or local 12-fold averaging. In either case, how does the unaverage density look like?

• What percentage of particles shows the portal-like structure?

• Figure 3C. What metal ion is shown in this panel? What is the evidence to assign a metal ion rather than a water molecule? Dedicating two panels to a ‘yet to be determined ‘ metal ion seems inappropriate

• Figure 5 is overcrowded and somewhat incremental.

Referee #3 (Remarks to the Author):<br /> Using asymmetric reconstruction techniques on cryo electron microscopy data, Colney et al show that one vertex of icosahedral feline calicivirus recognizes the cellular receptor, namely junctional adhesion molecule A. The vertex is formed by six copies of the VP2 minor capsid protein. Presumably the virus assembles with six VP2 molecules decorating each 3-fold vertex, but association with receptor at one vertex modifies the icosahedral structure of this vertex to create a portal and channel wide enough to allow the +RNA genome to escape into the potential host cell.

The paper is mostly well written although the figures and figure captions need improvement and clarification. The frequent use of the word “reveal” (“show” would be better to avoid spiritual implications) and “interest” (the reader should judge for him/her self to determine what is interesting) should be avoided. The asymmetric cryoEM analysis was well executed. Reference to the earlier publication on calicivirus structure (Prasad et al Science, 286, 287-288, 1999) would have been helpful for an explanation of “AB” and “CC” dimers and a description of the protruding P domain.

It is not clear whether VP2 is present at all 20 3-fold vertices in the assembled, mature virus or whether the virus assembles with VP2 present only at one unique vertex. In the former case the act of binding the receptor would create the unique portal vertex. In the latter case the uniqueness of one vertex would have been established during the assembly process. This needs to be clarified. However, the presence of a unique vertex in a calicivirus, no matter by what mechanism, suggests that similar situations might exist in many other apparently icosahedral viruses. Thus this opens the door to a vast new appreciation of “icosahedral” virus assembly processes and cell entry mechanisms. This is therefore an extremely important paper

Round 2

Referees' comments:<br /> Referee #1 (Remarks to the Author):<br /> This is a revised version of a manuscript by Conley et al. that utilizes cryo-EM and structural modeling to define a proposed function for the minor calicivirus capsid protein, VP2. The authors have added helpful new references, modified or added figures, and clarified a number of points suggested by the referees. The authors have not provided additional experimentation that would directly support their proposed function of the VP2, citing an expected prolonged delay in the publication of this structure. Although compelling evidence for interaction of the VP2 portal with an endosomal membrane bilayer would indeed be a crowning achievement of this work, the new insight into the conformational changes in the structure of a calicivirus particle following binding to its cognate receptor remains an important advance. Please clarify the following:

1. In the response to Referee 2, the authors note the following: “This was addressed in the legend to supplemental figure S7 (now S8), 81.6% of particles showed one or more portals.” Were there indeed certain particles that appeared to contain more than one portal? How commonly were these observed? Please clarify what this observation implies because it is somewhat inconsistent with the model of a single exit point for the genome.

2. The citation of work that shows the importance of Leucine 7 in the FCV VP2 is appropriate, but this new text should be re-written and clarified. For example, without context that this is mutagenesis of the FCV genome by reverse genetics and virus recovery studies, the concept of “strongly selected for” and the meaning of the other amino acid substitutions is unclear.

On 2021-11-01 16:46:16, user Edward Emmott wrote:

Please note, this preprint has now been published in Nature Protocols as: https://www.nature.com/arti...

On 2021-11-01 14:59:09, user David Bhella wrote:

To help readers understand the process from preprint to published articles, I have decided to share details of peer review and reviewers comments for my articles.

This paper was submitted to PLOS Biology, where after peer review it was eventually accepted.<br /> Reviewers comments:

Reviewer 1

The portal vertex is the aperture located within many virus capsids that allows the nucleic acid to enter and exit. It is an essential feature but in herpesviruses has escaped high resolution observation because of its similarity with the 11 other pentonic vertices located in herpesvirus capsids. This paper describes a structure of the portal vertex of herpes simplex at the highest resolution to date. New features are revealed including a novel 5-fold symmetric bridge between the external surface of the capsid and the inner portal comprising copies of UL6, and a large structure on the surface that borders the portal vertex, presumably helping to anchor it in place. Density consistent with viral DNA is located within the pore located within the portal, which is similar to some tailed bacteriophages, confirming a longstanding hypotheses that the herpesvirus capsid and some phage capsids work in similar ways. The work is particularly timely given recent publications in Science magazine (Volume 360, April issue) that show high resolution structures of the capsid; but those structures lack the portal shown here.

This will be of interest to virologists in general and herpesvirologists in particular. The interest to scientists outside the field includes the novel methods used with the Titan cryoelectron microscope to reconstuct this asymmetrical feature at high resolution in the face of similar but functionally, and as it turns out, structurally distinct pentonic vertices. The novel approach in which focusing on each facet was used to classify capsids into 10 groups, one of which led to the appearance of a novel pentonic vertex in the reconstruction was clever and could be used more broadly for other asymmetric structures thath are less than obvious at first glance. Thus, structural biologists, nanotechnologists and materials scientists may also find this paper intriguing.

This paper is outstanding in the discipline and has presented data that answers a longstanding problem in the field, with appropriate posing of new questions. The authors have reached reasonable conclusions based on the data, with reasonable speculation of the identity of the protein comprising the 5-fold symmetric bridging structure based on what is currently known. There will be great interest to identify what this protein is.

The paper is well witten with only a few Typos:

Line 140 does not read correctly.

Lines 78, 155, 256. The word comprise is misused. “The apartment comprises 4 rooms” is correct. “An apartment is composed of 4 rooms” is also correct. “The apartment is comprised of 4 rooms” is incorrect.

Reviewer 2

Several major articles in recent years have revealed the high resolution structures of herpesviruses icosahedral capsids. However, information on the portal vertex that is essential for both viral genome encapsidation and decapsidation is still missing. In this article, McElwee et al. describe a subnanometer structure of the portal vertex and reveal several previously unknown structural features surrounding it. Such data should be useful for any researcher aiming at understanding the essential processes involving this vertex. The resolution is still insufficient for assigning unambiguously some of these different features to HSV1 proteins limiting its mechanistic implications.

Major points:

• The PVAT unique structural organization in the icosahedral capsid is the novelty of this work. The five-fold symmetric features of the portal vertex are enhanced by C5 symmetry reconstruction. Clear densities of the purple and magenta regions of the vertex in Figure 2e,f support that they follow a five-fold symmetric organization. This is less obvious for the region exposed to the capsid outside where 2x5copies of the UL25 Cter structure were docked (film 2) which appears somehow blurred in cut-open structures (Figure 2a,c,e and film 3). The 12-fold symmetric portal protein UL6 is smeared-out, as expected, by the C5 symmetrisation. This raises a number of structural and biological questions that the authors shall address. How is the portal protein position maintained in the portal vertex (very few contacts are observed) to establish a continuous channel with the portal vertex pentameric protein for DNA passage in spite of their symmetry mismatch? How are the two UL25 Cter rings maintained in place and where do their Nter extensions fit in the portal vertex densities? The assignment as show in figure 2e (portal vertex in mauve, CATC in cyan) does not seem compatible with the observation by Pasdeloup et al. (2009; J Virol 83:6610) that UL6 and UL25 are directly interacting. How do the authors reconcile this discrepancy?

• Figure 2 does not optimally illustrate the different novel features highlighted in the text. This information is present in the supplementary movies but it would be probably useful to have these points illustrated in the main figures. Probably the most important would be to show a comparison between the portal vertex and the other vertices, this would show the extension of the specificities of the portal vertex. A closer view of the newly described pentameric portal vertex protein would also be useful. This could even grow into an additional figure to better document differences between the portal and other penton vertices. Across these lines Figure 3, which deals with DNA packing inside the capsid, might be of less interest and eventually used as a Supplementary figure.

• Line 355 and throughout the manuscript, including figure legends and movies: the authors use the expression "a sharpened map" to designate high threshold rendering of the reconstructions and use different thresholds in the figures. They shall define what they mean by ‘sharpened map” and specify the threshold used in each figure panel, as this varies, and probably also in the movie parts.

Minor points :

1- Line 56 : "dense proteinaceous layer". "dense" is unclear and should either be removed or precised.

2- Line 66: "preformed procapsids" is more correct than "nascent capsids"

3- Lines 77-95: references to more recent high resolution structures of herpesvirus capsids are needed (CMV: Yu et al. 2017, Science 356:1350KHSV: Dai et al 2018, Nature 553:521). They provide also useful up-to-date information for interpretation of the pentons molecular organization. After the authors submitted their manuscript there were also two publications on the 6 April issue of Science that report atomic models of the HSV-1 and HSV-2 icosahedral capsids. These structures do not reveal the portal vertex structure but could be useful to refine the quality of the other capsid pentons structure shown in the present manuscript and their description (e.g. in the paragraph starting in line 97; in the refs given in lines 131, 134…).

4 - the authors shall define or give a reference for “focused-classification” (line 98 and thereafter) for the non-initiated.

5 – line 140: “a metadata file was created with expanded the icosahedral symmetry…” ?!

6 - The authors should describe in greater detail how the CATC could close the portal pore. In particular, they could compare the diameter of the gap through the UL25 rings with the diameter of a double stranded DNA and discuss whether this is sufficient for blocking DNA or not. This comparison could also be performed with the PVAT.

7- the CATC seems different in figure 2d and 2e, is it due to a different segmentation, a different density threshold or to another reason ?

8- Line 167-176: the authors should explain how they docked UL25 in the CATC density shown in movie 2, especially in the distal tier that seems to be less resolved.

9- Lines 208 to 231 : this discussion seems particularly long with regard to its interest for the work presented (the main issue is to propose that pUL33 could be the unidentified portal-vertex protein), it should probably be significantly shortened.

10 – line 235: the authors shall explain in Methods how they determine the handedness of DNA organization inside the capsid based on their cryoEM reconstruction.

11 – line 272: the packing of DNA is very tight inside viral capsids reaching a high concentration but its density does not change.

Reviewer 3

The MS “Structure of the herpes-simplex 1 virus portal-vertex” by McElwee, Vijayakrishnan, Rixon and Bhella reports structures of a pathogenic herpes simplex virus (HSV-1) focusing on the organization of its portal-vertex where is a portal protein complex located. The portal complex is a nano-motor accomplishing the packaging of the viral genome into the viral capsid during maturation process and later participating in the genome release into a host cell during infection. In this MS the authors shared their method of how to move from highly symmetrical models to the less symmetrical complexes. This important approach will definitely be used by other labs. The authors obtained two structures of the virus one with the icosahedral symmetry and another with symmetry C5. The last one revealed the portal vertex position. The novelty of the MS is in a detail description of how the orientations of particle images where analysed and modified to be applied to the structure with C5 symmetry. Usage of the lower symmetry during the reconstruction process has enabled the authors to see the handiness of the packaged DNA.

The MS is sufficiently well written and explains the methodology used to obtain the C5 structure. However there are some questions related to the figures. While the overall view of the HSV-1 is consistent with well-established facts, the details that are discussed in the MS are not seen on the presented pictures of the capsids and not shown or indicated on enlarged figures. The authors are discussing such fine details of the huge virus as bundles of four helices, but they (in spite a huge size of the figure attached) are not seen in the figures, and in the printed version of them, they are too small and they will be small in the final published version of the manuscript. All figures are poorly labelled.

Lines 130-131 -> “a clear four-helix bundle that has been attributed to pUL17, pUL25 and pUL36” One can wonder why the authors did not provide any fitting of atomic models. The authors have to label where are these proteins are located. So these parts have to enlarged and clearly indicated, showing the differences between the icosahedral symmetry and C5. Figures 1a, 1b can be removed, surroundings in 1e -1h could be reduced allowing to make 5-fold views bigger and give room for labels.

Figures are repetitive to the high extend. The authors have to show firstly the overall structures and then go to the details indicating the helices and position of proteins.

The movies were unloadable. So it is unknown what do they show.

Minor comments:<br /> Line 99. “replaced by a unique five-fold symmetric assembly” - >The authors have to be cautious, if the C5 symmetry has been imposed during the reconstruction process they will be not able to see if there are symmetry mismatches in the portal vertex. Overall symmetry will still be C5, but it is has been suggested while ago that the portal complex itself highly possible has symmetry C12, while the proteins that provide connections with the viral envelope and a host cell outer membranes may have a number of different symmetries varying from asymmetrical ring complexes via C2 to C6 rotational symmetric and may be other symmetries as well. Please rephrase the sentence,

Line 144. Please explain what is that: “grouping the data into self-similar classes”. References would be helpful.

Lines 155,161 and 164. How positions of the pUL17, pUL25 and pUL36 proteins were identified? How the authors assessed where and which part of densities were assigned to pUL36?

Lines 172-173, “The distal (outermost) tier being rotated ~36o relative to the proximal one (movie S2)”. Possibly will be good to have a figure, the movie was unloadable.

Line 180. Some confusion. What the authors mean: “we can see noisy density that we attribute to the porta protein pUL6”. How noise can be attributed to a protein? Possibly it was some indications of presence of the signal related to the Portal complex? Please explain and rephrase lines 180-185.

Line 195. Please rephrase “Lying between the portal and the pUL25 PVAT density, a novel five-fold symmetric assembly replaces the usual pUL19 penton.” It is not a new assembly that replaces the penton, this is a complex that has been eventually found. It is highly possible that it does not have the 5-fold symmetry; otherwise it would be resolved better.

Helices have to be shown and indicated in the figure.

Lines 208-223 possibly should be moved into introduction, they are bit out of place.

Lines 233-249 possibly fit better to the conclusions.

Lines 243-245 “Our data suggest a reason for this, showing that pUL25 forms a double-layered cap on the outer face of the portal-vertex (the PVAT).” Figure would be helpful. One cannot see that cap in any of the figures. The authors have to illustrate their text and possibly provide a better explanation.

Line 247 “pentameric portal-vertex protein complex” the authors do not have any proves, that the complex is pentameric. The symmetry has been imposed. Please rephrase the sentence in a more careful way.

The authors repeat that again in the conclusions, but the emphasis on the 5-fold symmetry should to be avoided.

Line 303. The dose has to be not per linear angstrom, but per square angstrom.

Reviewer 4<br /> The manuscript “Structure of the herpes-simplex 1 virus portal-vertex” by McElwee et al describes their structural analysis of the herpesvirus capsid portal vertex using cryo-electron microscopy to image capsids inside intact virions. This is a significant step forward for the field after a number of conflicting and low-resolution reports on the portal structure, and the present work answers several questions about this very important part of the herpesvirus capsid. The manuscript is well written and generally focused and rational. A significant limitation is that the 12-fold symmetry of the portal itself was not resolved and so the main conclusions are about the organization of the 5-fold symmetric capsid features around the vertex where they don’t interact directly with the symmetry-breaking portal itself. This is not to diminish the value of this work, and the title accurately reflects this situation, but I feel the authors may have made some inferences that are on less sound ground and which they may wish to reconsider. That aside, I believe the relevance of their work and its potential impact on the field well justifies publication by PLoS Biology.

Of several points that would benefit from additional thought, the first is perhaps trivial but bears on the concept of capsid and tegument. The capsid proteins co-assemble and remain associated in the virion. Terminase subunits are clearly not capsid proteins because of their transitory association, and neither are tegument proteins because they form a compact but largely disordered layer outside the capsid and serve no known structural role in the capsid. Proteins that are intimately, specifically and symmetrically associated with the capsid would seem to include the variously named CCSC, CVSC and now CATC where the T stands for tegument. The subunits include pUL17 that is implicated in capsid assembly, and pUL25 that maintains the genome inside the capsid amongst other functions, and its not clear why the authors describe them as “tegument”. Certainly, adding another name to the list seems unnecessary. Similarly, is the PVAT that is composed of 10 copies of pUL25 really tegument since it is integral to the capsid portal vertex?

The imposition of 5-fold symmetry naturally obscures interpretation of the 12-fold symmetric portal, but it also affects any interface where the symmetry of either side may be affected locally. This is evident, for example, in the asymmetric map of phi29 (ref [32]) where neither symmetry is appropriate in this region. Interpreting weaker density is fraught with the difficulty of knowing whether the symmetry mismatch or flexibility is responsible for fuzziness, and this bears on the identification of the density at the Ta position as another protein(s), the identification of pUL25 with the PVAT despite the poor quality of the fits (Supplemental Movie 2), and the claim of highly-ordered DNA packing in a left-handed spool. Indeed, imposition of 5-fold symmetry may be enhancing a spurious signal for the DNA – spherical sections might demonstrate the soundness of this density as surfaces such as those in Figure 3 may be misleading. In general, I feel the authors may want to moderate text about the weaker density in case their interpretations don’t hold up.

A technical question – the so-called “gold-standard” Fourier shell correlation of 0.143 is used as a cut-off to give the most optimistic measure of resolution. However, the Methods do not state that the gold-standard method was applied during analysis, and this is essential to justify such a low correlation limit. Was the gold-standard method followed from start to finish? Alternatively, at 6.3 Å for the icosahedrally symmetrized map, helices should be evident as tubes, possibly with chirality and even possibly with density corresponding to large side-chains. Are such features observed in the density map? Further, with the reduction from 60-fold to 5-fold symmetry, the resolution was changed only to 7.7Å – wouldn’t a greater loss have been expected? How do the density maps compare to X-ray structures that have been filtered to these resolutions?

A minor point on lines 91-95 – another significant reason why the herpesvirus portal has been hard to find is that it resembles a penton in size and mass, unlike in the dsDNA tailed phages where the portal is generally more massive relative to the pentamers of major capsid protein.

Reference is made to the pUL6 portals being decameric (eg, line 183) and I believe the structure of the isolated portal by Trus et al, 2004, is relevant here, and could be compared with the density maps in general terms (recognizing that the Trus structure has 12-fold symmetry and the density map has 5-fold) – ie, would it fit, and can it be placed in the density? Can its directionality be assigned?

Wording. There are a number of semantic and grammatical errors, including:

Line 19: suggests virions may contain several capsids, which is possible but rare.

Line 72: the colon should be a comma.

Line 132: “one on top of the pUL19” means the pUL19 of a penton, as opposed to a hexon.

Line 134: Reference [16] is not to PRV but instead to KSHV.

Line 140: “…with expanded the…” needs correction.

On 2021-11-01 11:10:19, user David Bhella wrote:

Some papers live a seemingly charmed life, while others encounter considerable resistance under peer review. To help readers understand the path to publication I have decided to share the reviews for all of my papers on BiorXiv.

We first submitted this article to Nature Structural Molecular Biology, where it was declined without review. We then submitted it to PLOS Biology, where it received favourable reviews:

Reviewer #1: (Signed review - reviewer name is redacted)<br /> Ho et al provide a detailed structural description of Macrobrachium rosenbergii nodavirus (MrNV), a pathogen of fresh water prawns based on Virus-like-particles made with a baculovirus system programmed with the capsid protein gene of the virus. The close similarity between the VLP structure determined at 3.3Å and the authentic virus particles determined at 6.6Å resolution provide convincing evidence that the VLPs are good representations of the authentic virus. The highlight of the paper is the close similarity of the subunit organization and the capsid morphology to that of tomato bushy student virus (TBSV). Key features are the Arginine Rich Motifs (ARMs) of the C subunits that interact with RNA and are not visible, but sequentially precede an intricate series of subunit interactions mediated by residues following this region; the N-proximal shell (S) domain and the C-proximal protruding (P) domain. All of these features, as well as the dimer organization of the protruding domains and the metal ion stabilized shell domain interaction are far more reminiscent of the TBSV structure than the structure of other reported insect nodavirus structures.

The technical quality of the paper is high and it is well written and illustrated. The arguments are convincing that the family nodaviridae, comprised of the alpha (primarily insect infecting viruses) and beta (primarily infecting aquatic species) genera, result from convergent evolution of these genera toward bipartite RNA particles that clearly display subunits and capsid organization that are distinctly different from each other, with the beta particles showing much closer similarity to TBSV-like viruses than the alpha nodavirus structure.

The paper will be improved if the authors address the following points.

1. They should reference Tang, L., Lin, C., Krishna, N., Yeager, M., Schneemann, A., and Johnson, J. 2002. Virus-like-particles of a fish nodavirus display a capsid subunit domain organization different from insect nodaviruses. J. of Virology 76:6370-6375. While these subunits formed trimeric interacting surface domains, the organization of the subunit domain structure was described as TBSV-like.

2. The authors should provide more comparison with the beta nodavirus structure previously reported that they reference but do not discuss (Chen N-C, Yoshimura M, Guan H-H, Wang T-Y, Misumi Y, Lin C-C, et al. (2015) Crystal Structures of a Piscine Betanodavirus: Mechanisms of Capsid Assembly and Viral Infection. PLoS Pathog 11(10): e1005203).

3. Inter S-domain stabilization by divalent cations is not unique to TBSV and was described in some detail for Flock House Virus (Banerjee, M., Speir, J. A., Kwan, M. H., Huang, R., Aryanpur, P. P., Bothner, B., and Johnson, J. E. 2010. Structure and function of a genetically engineered mimic of a nonenveloped virus entry intermediate. J Virol 84:4737-46.). The similarity between MrNV and TBSV is convincing, but the discussion could be expanded a bit.

4. A striking difference between alpha and beta nodaviruses is that the former undergo a post assembly maturation cleavage that is required for infectivity. This activity does not appear to be present in beta nodaviruses and this should be pointed out and possibly commented on in terms of any potential homologous residues in the two genera now that there are two high resolution structures of beta nodavirses.

Reviewer #2:

The authors report the structure of a nodavirus infecting fresh water prawns, MrNV. The structure of virus-like particles (VLPs) reached 3.3 A resolution and the biological significance of this structure was validated by repeating the structure determination on the virus itself, purified from prawn post-larvae, at 6.6 A resolution. The VLP structure allowed modelling the atomic structure of the viral capsid.

The manuscript discusses evolutionary considerations that are interesting to those working in the field of structural virology and viral evolution: Unlike other nodavirus capsids that are built from trimeric capsomers, then MrNV capsid is built from dimeric capsomers. This structural difference reported by the authors earlier has raised the question whether the MrNV should be allocated to a new genus within Nodaviridae. This manuscript highlights further differences and intriguing parallels to the members of the Tombusviridae family.

The work is technically of excellent quality and the manuscript is clearly written. However, I would move most of the detailed methods from the Results section to the Methods section (especially cryoEM).

It would be beneficial if the Discussion could be developed further as now it somewhat just summarizes the results and has several unrelated minor points. For instance it could be focused more around the evolutionary differences and similarities. Where are the Ca2+ sites in tombusviruses and what is the similarity in other parts of the capsid protein? The strand-swapping seems unique to MrNV. It would be interesting to discuss the possible evolutionary origins of this – for instance are the interactions that this strand has with the neighbouring CP capsomer in MrNV similar to those that the equivalent strand has within the same capsomer in tombusviruses (see Bennett, et al. 1995 Protein Sci. 4, 2455–2468).

Finally, it is not clear from the figures if the topology of the MrNV P-domain can be clearly seen in the EM map. As comparison to tombusviruses and similarity of the fold is a key part of the paper, it is essential to provide stronger evidence for this.

Minor comments:<br /> Line 229: Please clarify what is meant by “unsupervised 3D refinement”?

Please add FSC curve for the 6.6 A map and validation curves (phase randomization) for all FSC curves.

Add supplementary Table for all EM statistics, including the B-factors used for sharpening of the maps.

Reviewer #3:

The authors present an atomic-resolution model of the Macrobrachium rosenbergii nodavirus calculated by cryo EM of virus-like particles (VLPs). MrNV is a pathogen of freshwater prawns that poses a threat to food-security and causes significant economic losses in the aquaculture industries of many developing nations. VLP were produced in insect cells, and data were reconstructed to 3.3 Å resolution map. CryoEM of MrNV virions purified from infected freshwater prawn post-larvae yielded a 6.6 Å resolution structure confirming the biological relevance of the VLP structure. The manuscript is well written with especially clear descriptions of the structural organization and comparisons to other structures. Figure 3 in particular is nicely presented. This is an important and impressively reported effort.

Minor:

“S” and “P” domains are used first in Results section at about line 277 without definition. Fig 1 b (central section) is called to illustrate fuzzier density of P domain, but there is no label/arrow in Fig 1 b to show P domain location.

For surface rendered maps in all figures, please add some type of label or axis indicator to designate symmetry axes to orient the reader. A description of the orientation might be sufficient, if it is preferred by authors and editor in order to preserve the visual integrity of the maps.

There is a clear and accurate statement in Results: “At this resolution, the (6.6Å) map appears identical to that of the MrNV VLP in all respects.” However, elsewhere, the same accuracy is required. Please adjust the Fig 7 legend where it states that “The authentic MrNV virion has an identical capsid structure to that of the VLP,” and “…. was indistinguishable from the VLP reconstruction” since both statements need to include some mention of the qualification of achieved resolution or other softening term to acknowledge that your conclusion that these are identical structures is highly probable, but cannot be stated conclusively due to the 6.6Å map.

On 2021-11-01 09:15:53, user Marius L wrote:

Congrats to releasing MARGARET. Please consider citing CellRank (see cellrank.org or the preprint), which has many conceptual similarities with MARGARET, e.g. CellRank automatically detects initial and terminal states, computes absorption probabilities on the Markov chain and charts gene expression trends using GAMs. Interesting to you might also be CellRank's efficient computation of absorption probabilities, which uses iterative linear solvers to exploit sparsity, circumventing the need to sample waypoint cells while being much more efficient than Palantirs implementation both in terms of time & memory (see preprint benchmarks). Recent releases generalize CellRank beyond RNA velocity, including e.g. a PseudotimeKernel to assign directionality based on any pseudotime (Palantir inspired) or a Real-time kernel to link cells across experimental time-points (Waddington OT inspired).

On 2021-11-01 07:26:38, user Prof. T. K. Wood wrote:

Authors also make common mistake of thinking lack of PI staining indicates viability (see doi:10.1111/1462-2920.14075).

On 2021-10-29 20:26:02, user Prof. T. K. Wood wrote:

Just exceedingly-poor title as the persisters are not "actively growing" but instead originate from formerly actively-growing cells. Persisters form, as shown repeatedly, from stress.

Also, it is already known stress induces persisters = main cause (see doi:10.1128/AAC.02135-12).

Key problem: reliance on the 2004 report that indicates there are different kinds of persisters and that they arise spontaneously.

Persistence is an elegant response to myriad stresses, as shown already through single-cell studies, though not reported here. See https://doi.org/10.1016/j.b... .

Already shown through single cell studies that the E. coli cell morphology becomes round (see doi:10.1111/1462-2920.14093).

On 2021-10-31 05:03:02, user Sandhya S wrote:

Supplemental Information is not complete. Only movies are uploaded.

On 2021-10-29 18:46:31, user Kevin Tyler wrote:

This is lovely work and credibly resolves a lot of discussion about what might be going on in a robust and well evidenced piece of work.

On 2021-10-29 13:45:32, user Angelino Carta wrote:

our preprint has been accepted for publication and a link will be forthcoming

On 2021-10-29 12:04:21, user FranziG wrote:

Impressive work confirming some important concepts in the transcription and GR field. I think the single cell part is very relevant. I am just wondering if changes in enhancer-promoter interactions profiled by H3K27ac HiChIP can be interpreted when the 'chipped' mark itself is changing. I would expect loosing contacts whenever H3K27ac is lost after Dex and gain contacts where H3K27ac is gained. Meaning that the observed gain/loss is not necessarily due to changes in contact frequencies. Somehow what you see in Fig. 3B and C. I am not a HiCHIP expert so ;) Do you normalize for epitope changes upon treatment during the HiChIP analysis?

On 2021-10-28 14:02:42, user Boyi wrote:

Is there a supplementary file? Thanks.

On 2021-10-28 10:08:07, user Thomas van Gurp wrote:

Hi Patrick, I would love to try this protocol, I'm curious to hear what you have learned and what you struggled with.

On 2021-10-28 09:34:38, user Peter Ellis wrote:

What an ABSOLUTELY fascinating system! This paper blew my mind clean out my ears. Excellent work :-)

I have only one quibble, relating to lines 329-333, i.e. the potential for conditional Y-linked drive.

You show that it is possible for a Y-borne gene to favour transmission of the paternal X (and oppose transmission of the paternal Y) in matings between XY males and X*Y females. I think it would be worth pointing out that the paternal Y cannot be selected to drive against itself. Rather, in this case the maternal Y is being selected to drive against the paternal Y.

In the case of the two-step pathway (b2'+3), a Y-borne drive modifier can only invade the population if it acts in X*Y females, not if it acts in XY males, because it is the maternal copy of the Y that is favoured by the drive in these matings - the paternal copy is disfavoured.

The same applies to the one-step pathway b2. Even if a single Y-linked gene is responsible for both directions of conditional drive, if its only mode of action is by perturbing sperm function, then it will be rapidly selected to become an unconditional driver. It must therefore act in X*Y females as well.

This means that conditional drive almost certainly has two separate mechanisms of action: one acting paternally, and the other acting maternally. This makes the two-step pathway much more likely than the one-step pathway, and may give some clue towards tracking down the mechanism of action - the proposed mandarin vole system in ref 11 (maternal Y acts via imprinting to inactivate an essential gene on the X*, so only embryos that inherit a paternal X can survive) is a beautifully elegant solution, and blew my mind for a second time in one evening.

I personally think the most likely course of events is:

1) Acquisition of unconditional Y-drive, acting paternally. <br /> We know that there is a paternally-acting sex ratio drive system in mus musculus, and some of the interacting partners (Sstx and Ssty) are also present in rat. So this is likely quite ancient. We also recently showed that the proximate mechanism for this is probably differential motility of X and Y-bearing sperm.<br /> https://pubmed.ncbi.nlm.nih...

2) Appearance of a feminising X*, facilitated by the presence of Y drive

3) Development or enhancement of compensation in X*Y females to improve fertility via polyovulation.<br /> In a transgenic system that eliminates male embryos in the peri-implantation, we show that there is some inherent compensation of litter size in mus musculus. So it seems some element of poly-ovulation may be common in rodents, allowing for a certain amount of pre-/peri-implantation attrition without reducing litter size. This seems like the sort of phenotype that could relatively easily be increased to allow greater levels of compensation.<br /> https://www.biorxiv.org/con...

4) Development of conditional drive in which X*Y females drive against the paternal Y<br /> Once compensation is well established in step 3, the X*Y mothers have more scope to eliminate even more embryos prior to implantation and thus select only the ones they want.

Mechanistically, all this can be most readily tested by IVF and/or embryo transplantation experiments - are these techniques established for mus minutoides yet?

Once again, thanks for one of the most enjoyable papers I've read in a long time!

On 2021-10-28 08:34:48, user Patrick Deelen wrote:

Very nice manuscript, congratulations. Can you elaborate on how this method compares to: https://pubmed.ncbi.nlm.nih...

On 2021-10-27 21:18:16, user Jennifer Wenger wrote:

Were the antibody lineages used for the study from unvaccinated people?

On 2021-10-27 13:53:05, user antibodies-online wrote:

Proteins used for validating In Silico Driven Prediction of MAPK14 Off-Targets (among others): NMNAT1, FABP4, MEN1

On 2021-10-27 11:11:24, user Sebastian Lobentanzer wrote:

Hi there,<br /> cool paper! I have a minor comment on Natalizumab: it targets integrin alpha 4, which is a receptor for VCAM1. I am not aware of an association to ICAM1, could you elaborate on that?

On 2021-10-27 09:05:18, user Dong wrote:

The R package Miso also allows for user-defined isotopologue pattern search. Maybe you could consider citing this package: https://academic.oup.com/bi...

On 2021-10-26 23:22:30, user Xin Chen wrote:

We appreciate that the authors tested our previous results using new reagents and methods. However, we have to point out that there is a big misunderstanding of our published work. First of all, asymmetric histones do NOT imply the existence of “immortal histones” as the authors hypothesized and used to make predictions in their experimental design. In fact, distinguishing old versus new canonical histone must be in the context of cell cycle progression: Old refers to the pre-existing histones before S phase and new refers to newly incorporated ones during S phase. These two populations can be distinguished by the tag-switch or photoconversion methods only after the switched or converted cell goes through one complete S phase and enters the subsequent M phase. Moreover, the new histones with switched or converted labels will mature over time during cell cycle and gain old histone features, and thus there are no “immortal” histones. However, we are not seeing any labels in this work that indicate active cell cycle progression, which is very concerning given these tissues are ex vivo for more than 40 hours.<br /> Second, it would be highly appreciated if the authors include germline versus somatic cell markers in their figures. As of now, it is impossible to tell whether the weak H3 signals in Figure 1C and 1E come from germ cells or somatic gonadal cells. The bright spot in Figure 3E was interpreted as hub cells, which are quiescent somatic cells. If this is the case, it would be very strange that such a quick old to new H3 turn-over occurs in these cells, as indicated in Figure 3E legend.<br /> Finally, we have to point out that our previous results were entirely misinterpreted in the “Alternative Hypothesis 2” in Figure 2, because we are not assigning random stem cells (GSC) and progenitor cells (SG) together as pairs — all GSC-GB pairs we analyzed are still connected by the spectrosome structure (Tran et al., 2012; Xie et al., 2015; Wooten et al., 2019), indicating that they are daughter cells derived from one GSC division. Furthermore, our previous conclusions were not solely based on the post-mitotic GSC-GB pairs, but also on stem cells undergoing asymmetric cell divisions, based on fixed and live cell imaging.<br /> In summary, this work is based on both misunderstanding and misinterpretation of our work, leading to an incorrect hypothesis. Additionally, there is no single dividing stem cell or a pair of daughter cells derived from stem cell division shown in this work that can lead to the conclusion of “Symmetric Inheritance of Histones H3 in Drosophila Male Germline Stem Cell Divisions”. We hope these comments clarify several critical points for both the authors and the readers of this preprint. Thank you for your attention!<br /> Xin Chen<br /> Johns Hopkins University

On 2021-10-26 21:56:12, user Swastik Phulera wrote:

Great publication, the reference numbering seem to be a bit off. Eg ref 73 (in text) should be 70

On 2021-09-27 18:29:26, user anna moroni wrote:

Dear Authors, very interesting results. I noticed that in C-type inactivated Shaker channels, the selectivity filter is impressively similar to that of HCN4 channels in their non-conductive form (Saponaro et al, Mol Cell 2021,DOI:10.1016/j.molcel.2021.05.033). The comparison between WT and W434F mutant in Shaker highlights the large movement of Y445 and D447 sidechains, similar to those of Y482 and R484 observed by comparing conductive and non-conductive HCN4 SF. Further, C-type inactivated Shaker channels show two ion binding sites only and low conductance, two typical features of HCN, as well as reduced selectivity for K over Na (Kiss et al, 1999, DOI:10.1016/S0006-3495(99)77194-8). So, really striking similarities!

On 2021-10-26 03:30:22, user Manasi Datar wrote:

Dear authors,

Thank you so much for sharing your valuable <br /> research. It was indeed an insightful paper which has a remarkable application in future bio monitoring of terrestrials. It would have been nice to know the reasons behind using the specific COI forward and reverse primers in your experiment and some more information regarding the target genes. Also, hedgehogs that are present in the vicinity don't visit the zoo as they may be hibernating during winters, but will still keep physically shedding DNA in the air which may have been detected in your samples. Overall, I enjoyed reading your paper and am intrigued with the efforts taken to conserve biodiversity.

On 2021-10-26 02:19:49, user Amelia wrote:

This is a fascinating paper! I'm writing as a social scientist, and not a biologist, but I was curious about the potential impacts of this approach for the collection of human DNA. I read an interview in LiveScience where you describe the contamination of samples with human DNA as a "hurdle" and a source of contamination. Yet, I was wondering if there are any ethical concerns about this kind of passive DNA collection without consent?

On 2021-10-25 08:35:03, user CDSL JHSPH wrote:

Dear Dr. Clare et. al.,

It is my pleasure to review your paper! Thank you for contributing to global terrestrial biomonitoring and ecological analysis. Using airDNA as a biomonitoring tool under natural settings show great potential. The decline in biodiversity throughout the world urges the development of non-invasive techniques that could offer rapid and accurate results. Your study successfully reveals the power of airDNA sampling at distance, and from my perspective, the new technique would truly revolutionize terrestrial biodiversity surveys.

My question would be, do you anticipate any difference of conducting airDNA sampling between zoo setting and real natural environment? As in the zoo, animals are gathering together and each kind of them have their own enclosures. While in the natural environment, animals will move more freely as there will be no space limit. Also, we know that some endangered species actually live under extreme conditions. For example, the weather might be extreme. Will the study results be influenced by extreme weathers?

Moreover, I am wondering if we need more intermediate steps to shift from zoo setting to actual field as zoo setting can evoke less complex behaviors and is still involved by human. Since we are aiming at non-invasive technique, what future steps could be done?

All in all, I think the paper makes significant contribution to the biomonitoring field, the methodology is quite convincing. I am just curious about the actual application of airDNA in the wild and the future plans regarding this study.

Thank you for your work and I am looking forward to future outcomes on this topic!

On 2021-10-24 23:25:38, user CDSL JHSPH wrote:

This paper has so much potential in innovating the yield of ecology with a new biomonitoring technique. However, there are some parts of the paper that need further clarity. For instance, in figure 1, if there was more clarity on the enclosures and their distances from one another that can give a better understanding of the relationships between the animals and the number of reads given. Also, if there was a supplementary portion of what each of these target animals consume and the typical amount, it can perhaps clarify the large amount of recovered sequences coming from the farm animals. Additionally, depending on the season of when sampling is done, do you believe that played an impact on the sequence levels?

On 2021-10-24 15:04:18, user Kayla Hess wrote:

This investigation will surely revolutionize the field of terrestrial biomonitoring as it establishes the successful use of airDNA outside of a controlled laboratory setting. Collecting samples at multiple locations inside and outside of the park provided a large and diverse pool of data, thus supporting the validity of the investigation. The read count variability of airDNA inside the enclosures in comparison to outside seemed unexpected. Especially since there was so much drift between the closures, I expected there to be a similar amount of drift to outside the park as well. What might account for this difference? Wouldn’t the same weather and climate patterns that caused airDNA to drift between the enclosures also cause it to drift outside? I figured that maybe most of the drift could have been due to human activity rather than weather events but I could be mistaken. In addition, how might human activity affect the accuracy and precision of airDNA terrestrial biomonitoring in the wild?

On 2021-10-24 14:07:57, user SC wrote:

Such an interesting preprint, it really has set the stage for further developments in measuring terrestrial biodiversity! Out of curiosity though, I was wondering if you had any additional data on the wind speeds or wind patterns at the time of testing, especially given the range of eDNA findings presented in the tables – as your abstract notes on airDNA dispersal away from sources, the "ecology" of this airDNA traffic in air could possibly benefit from anemometric readings, wind–mapping, or general information relevant to climate and seasonality when having tested. A satellite map of wind patterns could possibly elucidate dispersal of this airDNA to places far from its origin!

On 2021-10-24 02:45:04, user Ashlyn Blevins wrote:

This is an exciting study!!! I am very much looking forward to reading more of your work and about airDNA. This could be a great new way of biomonitoring. I am not very familiar with ecology field work and would like to know more about how you took your samples. The orange rings in figure 1 look quite large. Did you just walk around with your sampling apparatus, use a drone, etc? I also noticed that two samples of mole rat DNA were excluded due to cross contamination of tubing from prior use. Was the tubing in this experiment cleaned or changed between each sampling area or would the filters you used catch everything? Thanks for any insight - I am really looking forward to learning more! :-)

On 2021-10-24 01:08:38, user CDSL JHSPH wrote:

The method of this study is creative and unique, and the experiments are well designed and set a precedent in detecting eDNA directly from the air to achieve terrestrial biomonitoring. The claims and conclusions are mostly reliable, except for the "versatility" of this biomonitoring method. But more improvements and research need to be done to refine the method to get more accurate data, before this biomonitoring method can be used in absolute natural conditions where animals are not spatially confined and certain. The writing of the introduction is great. The context of this field, as well as the need and importance of their study are well stated. Regarding figures and tables, I think the best way to present data like the tables in the paper is to summarize and simplify numbers into straightforward diagrams like figure 1.

On 2021-10-18 19:24:43, user CDSL JHSPH wrote:

This is a great potential application of airDNA being used for bio monitoring. The tables certainly show the specificity and range of this technique being used on various types of animal species in a natural setting. I’m just curious if you think the time of sampling has any impact on what DNA samples are detected? I ask mainly because it was noted how the hedgehog DNA was detected in certain enclosures despite it being more absent during the time of sampling. Do you think there is a certain range of time that these samples are representative of? Are they detecting animals that are present the day of sampling or possibly animals that were there a few days or even weeks prior to sampling?

On 2021-10-26 02:55:59, user CDSL JHSPH wrote:

This is a good introductory body of work regarding how mitochondria are acting as a signaling source to promote cellular proliferation instead of working to promote favorable bioenergetics. When analyzing what you are commenting upon I am curious as to why you chose to include the methods and materials section within the supplementary materials portion of you manuscript? I also am curious about the organoid model utilized to test this concept. Was there any other differences observed to make sense of the bone tissue migration compared to the breast cancer tissue migration? Does it overall seem like this process could very well happen in all types of cancer outside of the breast cancer cells tested? This specific data, while very important to understanding mechanistically what is going on with this mitochondrial transfer, I think needs more clarity quantitatively or qualitatively. If we were too look at this phenomenon in vivo are there going to be confounding factors that perhaps change this transfer process? This could potentially start to explain why there are differences in transference patterns observed between tissue types. I believe that overall this body of work is a great stepping stone for further investigation and understanding of what this process could mean for both "healthy" cells as well as "diseased" cells such as cancer cells as well as tissue health.

On 2021-10-25 06:55:40, user Sneha Pramod wrote:

This experimental study was performed to understand the mechanism by which laterally transferred macrophage mitochondria promote cancer cell proliferation. The results of the research study have been documented meticulously and clearly. One of the key findings of the study is that the transferred mitochondria lose their membrane potential and become depolarized within the cancer cell. They also report that the transferred mitochondria act as a signalling source and induce ROS generation. I would like the authors to elaborate on whether there is a link between loss of membrane potential and increased ROS production. Additionally, I was wondering if the authors explored the dysregulation/mutation of mtDNA in transferred mitochondria and whether it contributes to the signalling capacity in any way.

On 2021-10-25 06:21:23, user Suchitra Magesh wrote:

The study was novel and used different methodologies to test the transfer and function of mitochondria in tumor cells. The figures that depicted the methodology especially were detailed and neatly presented. I had a query regarding the correlation of the Warburg effect with the results of this study. Warburg effect, one of the hallmark of cancer, explains the use of aerobic respiration in the mitochondria of tumor cells resulting in the generation of ROS. Does the transferred mitochondria undergo the Warburg effect? In addition does the increase in ROS, as well as proximity of mitochondria to ROS, result in increase of risk of mtDNA damage?

On 2021-10-22 19:08:45, user Anissa Cervantez wrote:

Hi there!

I agree with the comments below, if you could further explain how you were able to connect that ROS lead to increased cell proliferation. I understand that due to the increased ERK-KTR translocation that this means theres an increase in proliferation but I am confused about how from your data (Figure 4D) how you are then concluding that ROS is the reason behind the increased proliferation.

Overall, I think this paper has great clinical relevance and am wondering what future work you could see being done. I know you had mentioned that these macrophage mitochondria are not being degraded and am wondering if you have any indication as to why this is? Also do you foresee this having implication in many cancer cell lines? It looks like this could have implications in bone and breast cancer but what other types of cancer do you think could be affected by this process?

On 2021-10-21 14:07:25, user Iris Chen wrote:

Great analysis and synthesis of ideas pertaining to the mechanism by which macrophages stimulate cancer cell proliferation. The findings are interesting and innovative, and the arrange of figures and conclusions is also great! And I am just a little bit confused with Figure2A, so why you chose to observe the recipient cancer cells for 15 hours? Because I find in some experiments, the data was collected after cells were co-cultured for 24 hours, so maybe you can also lengthen the observation time in 2A to 24 hours, in this way I think your conclusions will be better supported. And another possible advice is for Figure4D, I think here you can also add a figure to show the number of mito fragments with in the artifact macrophages, just like Figure4B. And I also think you could add more details in the introduction part, like the proliferation of breast cancer cells (why you chose to get the data after 24 hours), the relation of ROS and ERK (is there any previous studies about this pathway? are there other possible pathways?), and the mitochondrial network in macrophages (has anyone also found M2 with smaller mitochondrial fragments before? and does the mitochondrial phenotype influence the cell functions?). Anyway I think this is a really good preprint, thank you for presenting your work here.

On 2021-10-20 06:09:07, user Seb Wang wrote:

Hi, I have a doubt about using photobleaching to generate ROS to test ERK activation. I feel like it is hard to draw comprehensive conclusions based on this experimental design. Even if under this condition, cell proliferation decreased, you can only conclude that ROS leads to increased cell proliferation, so transferred mitochondria can lead to cell proliferation through ROS accumulation, but you cannot eliminate the possibility that transferred mitochondria can induce cell proliferation through other mechanisms. I think another way to do it would be in cancer cells transferred or not transferred with mitochondria, either somehow knocking out all ROS or not knocking out all ROS, and comparing four experimental group’s proliferation state with each other. Then, we can know if transferred mitochondria induced proliferation is completely ROS dependent or not. I might be wrong, but I hope my comment makes sense!

On 2021-10-26 02:41:51, user CDSL JHSPH wrote:

Thank you for sharing your research. I found the paper to be very well-structured and I think the organized layout of the paper helps build a narrative that can be read by a wider, non-scientific audience as well. <br /> It was interesting to find that neutrophils in older adults had increased uptake and oxidative capacity, compared to younger adults (Fig 2B). I liked that you referenced research articles with findings that were in contrast to your own finding regarding neutrophil oxidative capacity in older adults. I think including such references is a great way to avoid confirmation bias in research.<br /> I hope to see future studies that build upon this research theme by using cohorts that span the entire age range of 1-80 years. It would also be interesting to see the variation in dendritic cells and B cells at human nasal mucosa with age. In addition to the challenge of obtaining human tissue samples, did you face any other challenges while investigating the immune cell composition at nasal mucosa? Also, were there any factors that led you to focus mainly on T cells and neutrophils for this study?

On 2021-10-26 02:22:47, user Lisa Pieterse wrote:

Very impressive and comprehensive work published on the changing immune landscape as a factor of ageing. Inclusion of immune cell composition and activation states provided an interesting gauge into the diminished nasal mucosal T cell reservoir, and perhaps overdependence on granulocyte function, in older adults. Has your group considered looking into Th17 subset composition amongst colonized and uncolonized children, young adults, and older adults? It would be immensely interesting to see if there would be a difference in Th17 populations within the nasal mucosa of older adults with or without S. pneumoniae colonization especially. As you point out in your paper, your group observed diminished T cell density and activation within older adults (Figures 1C, 1E, 1I), but did not include data on Th17 composition or Th17-producing cytokines such as IL-17, IL-21, or IL-22, although I may have missed this. Thanks in advance!

On 2021-10-26 01:54:16, user CDSL JHSPH wrote:

This was an interesting study to read, and a great analysis of how the immune system reacts as people age in their mucosal surfaces. After reading this study, I have a couple of questions regarding your experiment. In Figure.3 I understand that the data focuses on Spn colonization causes of inflammation in children. This test included young adults, but I noticed older adults were not included. Is there a specific reason older adults were not tested for this part of the experiment? Figure. 4G also omitted the older adults from this study. Lastly, the discussion noted that children ages 6 to 17 years old were not included in this study. Was there a particular reason for the age gap in subjects? Thank you for your responses in advance!

On 2021-10-25 21:17:54, user CDSL JHSPH wrote:

A novel and interesting read characterizing effect of age on the composition of immune cells in nasal mucosa. The paper certainly supports the authors’ argument of needing further mucosal studies to understand the changes in immune responses and risk of individuals with age. The key findings of reduced granulocytes and T cells in children and older adults respectively suggest new directions to explore the increased susceptibility to infections in these individuals. The figures were beautiful representations of the results and great analysis.<br /> The only source of confusion I encountered are the monocyte and neutrophil functional blood assays. Since it is presented in the abstract and experimentally shown in the latter figures that the immune functions in the blood are not reflected in the mucosal tissues, how is the importance of these functional blood assays to the paper justified? A few minor suggestions I would propose are to include a few lines explaining the intention behind an experiment and possible future studies in advancing the field.

On 2021-10-22 14:29:18, user Trupti Tripathi wrote:

I really enjoyed reading the paper. The main objective to investigate the composition, activation, and functionality of immune cells in the nasal mucosa was accomplished with sufficient supporting data. However, a lot of statistical data was represented in the figures with little or no information about how it was calculated, such as the Kruskal stress values and the multi-dimensional plots with Euclidian distances. The violin plots with box plots were the best way to represent the data in the paper and were easy to interpret. The correlation between the reduced number of nasal granulocytes and increased expression levels of CD66b, and increased MPO in children is profound. I wanted to know if it was challenging to collect nasal biopsies in so many individuals to measure monocyte functionality that you had to switch to blood. Could this comparison of immune cell functionality in mucosa samples vs. blood be made using mouse models? Do you think this study can be used for making personalized medicines against respiratory tract infections in children or older adults with underlying conditions? What would be the usefulness of this study during the current COVID-19 pandemic, where newer strains of viruses are growing rapidly? Thank you so much for this great read.

On 2021-10-21 01:28:21, user jt4444 wrote:

Great analysis of granulocyte and T-cell distribution with respect to ages. This paper also had a great introduction relating previous studies of pneumonia cases in children and adults. I do have some confusion, most of the data analyzing granulocyte are mainly subjected to neutrophils in this paper; however, there are many other immune cells with this that contains granulation activities and the same markers? <br /> Also, I understand that Spn are able to colonize the nasal cavity, but does this necessarily count for the colonization in lower respiratory tract (where most pneumonic activities happen). Also, it would have been great to analyze and differentiate adaptive activities such as immunoglobulin levels at this site since they do seem to occupy these linings. Lastly, could this also be tested in mice (i.e. acquiring respiratory tract samples even in the lower region with respect to lifetime and maintaining the same number of people for each group)? Overall, a good paper focusing on a subset of immune population and characteristics within the nasal lining cavity!

On 2021-10-25 19:58:44, user Joseph Binder wrote:

I think you had a great paper and it was quite insightful. These findings will help with conservation efforts in these regions. I know the focus of this study was in deserts. The extreme temperatures drive mammals to wetlands. Do you think that by going into a different environment such as a rocky mountainous region would provide similar results? Could this be a potential follow up study? For figure 1 I know that the x-axis was shared among each sub-figure. I think to make the figure more clear to the reader having the x-axis under each figure would have helped.

On 2021-10-25 18:36:26, user Michael Matthew wrote:

This was a great examination of the factors affecting ecosystem food webs. I have one question about predator-prey balance. While a major concern is the removal of feral donkeys and similar invasive megafauna, you also mentioned the importance of maintaining predator populations. Regarding optimal food ecosystem and web structure, what are the most effective methods of maintaining predator populations and introducing supplementary predators if needed? Does this depend on predator-prey relationship, time of year, or biome?

On 2021-10-25 19:52:39, user Mike Anderson wrote:

We are happy to say that we have just been notified that a peer-reviewed version of this manuscript has been considered acceptable for publication in TVST. We will be working with their Editorial Office to submit the final files for publication. We welcome any comments the community may have as this line of work and field continue to evolve!

On 2021-10-25 18:16:45, user Peter Rogan wrote:

This article is in press in the International Journal of Radiation Biology. DOI: 10.1080/09553002.2021.1998709

On 2021-10-25 12:35:33, user Maiwenn KERSAUDY-KERHOAS wrote:

Video of assembly and demo available here on Youtube https://www.youtube.com/wat...

On 2021-10-25 04:06:16, user Critical Dissections wrote:

Great article, I particularly appreciate that the research wishes to address SARS-CoV-2 viral transmission post-vaccination by stimulating mucosal immunity in addition to systemic immunity. Not to mention, the production of a shelf-stable, oral vaccine would be a great contribution to current global vaccination efforts.

The experiment design was planned and executed well to imitate and evaluate post-vaccination transmission of SARS-CoV-2. However, after thoroughly reading the paper, I was left searching for answers that would provide clarity to readers such as myself should they be addressed.

The first is regarding the use of an adenovirus type 5 vector when a large proportion of the population already has natural immunity to this serotype. Is there a benefit to using this vector as opposed to another subtype with less seroprevalence in the population?

The second unanswered question is regarding the IM vaccine used. Why was a SARS-CoV-1 spike protein vaccine used instead of a vaccine that expresses the SARS-CoV-2 spike protein, such as one of the current mRNAs vaccines currently in use?

My third unanswered question is regarding the high virus titer used to inoculate index hamsters. What was the rationale behind inoculating hamsters with a high physiological dose as opposed to a moderate dose that may better represent the amount that humans are actually exposed to?

A thought, not a question, but many of the experiments performed on samples were collected at several timepoints, yet the sVNT was performed at only 1 timepoint following full vaccination. It would have been informative had the sVNT been performed on samples from week 4 to determine the neutralizing ability of antibodies in between vaccinations and to compare to IgA and IgG levels which were determined at that time point.

As a final lingering question, IN- and oral- vaccinated index hamsters inoculated with the high dose viral titer showed more effective viral clearance than naïve hamsters that were exposed to these same index hamsters in the aerosol chambers. And as seen in figures 4b-g, though high viral and infectious viral loads were detected in the naïve hamsters they were protected from severe clinical disease outcomes at day 5 post-challenge. However, the same data from figure 4 b-d also seem to show that the unvaccinated, naïve hamsters were not protected from viral infection transmitted via aerosols from the infected, IN- and oral- vaccinated index hamsters. Yet, one of the conclusions from the paper is that IN- and oral- vaccinated index hamsters transmitted less aerosolized infectious virus to unvaccinated naïve hamsters. Have I misinterpreted the data from figures 4b-d?

The results of the study make a strong case for adenoviral-vectored SARS-CoV-2 vaccine candidate that induces mucosal immunity and I look forward to reading more studies from this team.

On 2021-10-24 17:46:08, user banksinoma spinifera wrote:

This is an interesting study about the effect of propionate on diabetes-induced neurological dysfunction.

Two small suggestions for the authors:<br /> - The title is a bit misleading. The role of PI3K-AKT-eNOS is not demonstrated. The changes are associated with the improvement but there is no loss-of-function or gain-of-function study to demonstrate it.<br /> - In Fig. 13a, the blots for p-PI3K and p-AKT are identical. Authors should check that out.

Best regards!

On 2021-10-24 04:52:42, user Kumiko Hayashi wrote:

Now published in Biophysical Journal https://doi.org/10.1016/j.b...

On 2021-10-23 10:13:43, user Michael Ailion wrote:

This manuscript examines the role of steroid hormones in<br /> regulating exit from the dauer diapause state in C. elegans. The manuscript presents a careful examination of partial dauers made by daf-9 mutants that are deficient in dafachronic acid (DA) steroid hormone production, concluding that these partial dauers are dauers that have initiated but failed to complete dauer exit, and that thus, complete exit from dauer requires DA steroid hormones. These major conclusions of the paper are well-supported by the<br /> experimental evidence. Convincing experiments demonstrate that daf-9 mutants make full dauers in unfavorable conditions, that these animals then become partial dauers when shifted to favorable conditions, and that many daf-9 partial dauers formed in favorable conditions transit through a transient full dauer state on their way to becoming partial dauers. Finally, it is shown that DA steroid hormones promote exit of arrested daf-9 partial dauers. This is a solid study and my critiques are all relatively minor.

Minor Points

1. The paper could be clearer on precisely how daf-9 and DA steroid hormones are required for dauer exit. Though it is true that daf-9 and DA are required for “complete” dauer exit as accurately stated in the Abstract, the take-home message of the paper often seems to be that daf-9 and DAs are required for “dauer exit” without getting into the nuance of which aspect of exit (e.g. lines 143-144 and elsewhere). The idea that DA hormones are required<br /> for exit is unintuitive and paradoxical given that daf-9 mutants actually initiate dauer exit rapidly under favorable conditions. Though the paper is generally careful in the precise way it talks about exit, I think it would be clearer to explicitly state the precise role of daf-9 and DA in dauer exit as<br /> follows. Based on the data in this paper, I would argue that there are two steps to dauer exit, only one of which depends on daf-9 and steroid hormones. First, an animal needs to initiate dauer exit, which leads it from the full dauer state to the partial dauer state. This step does not seem to require DA steroid hormones as daf-9 mutants initiate dauer exit from the full dauer state to the partial dauer state rapidly. Instead, this initial exit is likely regulated in a similar way as dauer entry – it is blocked by high levels of dauer<br /> pheromones, or in Daf-c mutants like daf-2 and daf-7. When pheromone is low (“favorable conditions”) initial exit occurs to the partial dauer state in a DA-independent fashion, and then there is a second stage of regulation going from partial dauers to non-dauers. daf-9 and DA steroid hormones are both necessary and sufficient for this second step, exiting from the partial dauer stage and<br /> resuming reproductive development. In one sense, it could be argued that daf-9 and DAs are required for full “execution” of dauer exit, not the “decision” to exit (just as daf-16 partial dauers are sometimes suggested to be defective in the execution of dauer formation rather than the decision). So ultimately, the action of DA hormones is required for reproductive development, whether it is<br /> to bypass dauer entirely at the L2 stage or whether it is to fully recover after forming dauers.

2. Some figures would benefit from non-dauer controls. For example, in Fig 2, it would be nice to see the pharynx width, speed, and pumping frequency of non-dauer L3s as a comparison. Without these controls, it isn’t clear which aspects of the daf-9 phenotype are dauer-like vs. non-dauer. Similarly, in Fig 3E, the daf-9 partial dauers have a speed similar to the recovered WT dauers. After 24 hours in favorable conditions, I would think the WT would be at L4 stage (i.e. fully non-dauer), yet this is one piece of evidence that<br /> daf-9 full dauers can become partial dauers upon a shift to favorable conditions. So does a partial dauer move at the same speed as an L4? If so, speed would not be a very useful measure in defining an animal as a partial dauer because it wouldn’t distinguish partial dauers and non-dauers. Including a WT L4 control would be useful. Pumping frequency in 3F is clearly less for daf-9 partial dauers than recovered WT dauers (presumed L4), but what about pharynx width shown in 3G? Is that more like a dauer or a non-dauer? Again, a WT L4 control would be useful.

3. Related to point 2 above, it would be nice to show that WT dauers at early stages of recovery (much less than 24 hours) resemble daf-9 partial dauers. That would demonstrate there is likely nothing abnormal or incomplete about daf-9 partial dauers and really nail the idea that the daf-9 partial dauer phenotype is due to defects in dauer exit rather than dauer formation.

4. Though strong evidence is presented that some daf-9 partial dauers transit through a transient dauer state (Fig. 4), it also seems<br /> likely based on the data in Figs 4 and S2 that daf-9 can also go from an L2d to a partial dauer without becoming a full dauer, though this possibility is not discussed. This raises several questions. Are the daf-9 partial dauers formed at 20° without going through a full dauer state (Fig S2) different than those formed at 25.5° (Fig 4)? If WT L2d are shifted to favorable growth conditions before becoming dauer, do they go through a partial dauer state that resembles daf-9<br /> mutants? Do daf-9 L2d look the same as WT L2d?

5. Related to point 4 above, the presumed daf-9 mutant L2d<br /> pharynx shown in Fig 4B looks somewhat slim – a WT L2d pharynx photo would be useful as comparison.

6. It is stated twice (lines 130 and 157) that there is only “anecdotal evidence” that daf-9(dh6) and daf-12(rh273) mutants form full dauers under unfavorable conditions. However, the two papers cited (Antebi et al.1998; Gerisch et al. 2001) both have tables with quantitative data showing full dauers formed in these mutants under starvation conditions. I would suggest to not use the phrase “anecdotal evidence.”

7. Lines 348-350: it is suggested that DA does not promote exit of full daf-9 dauers in the presence of pheromone because of<br /> inaccessibility of DA. Though this is possible, another possibility is that DA only promotes dauer exit after an initial decision to exit that does not occur in the presence of pheromone (see point 1 above).

8. It appears that the WT and daf-9 full dauer speed data shown in Fig 2C may be the same data shown in Fig 3E. If so, this should be<br /> stated explicitly, and any other data reused between different figure panels should be stated. Were all the data in Fig 3E from experiments performed in parallel? If not, this should be stated.

9. It would be helpful to show DIC micrographs of the pharynx without the yellow outline drawn on top since it is very hard to see<br /> the pharynx boundaries with the drawn outline (Figs 2A, 3G, 4B, S3). I would recommend a supplementary figure showing the same images without the outline.

10. It is unclear which daf-9 mutant is shown in Fig 3G<br /> (presumably dh6 but not stated). Please state this in the legend.

11. It would be helpful to state in the Fig 6C legend that these are WT dauers, not daf-9 dauers.

12. I was confused by Fig 7D. It seems that the presence of the UAS::ICE transgene promotes dauer exit, regardless of the presence of the XXX cGAL driver, which seems odd. Then I read in the Methods that dauers with the UAS::ICE transgene are SDS-sensitive. Are these real full dauers or partial dauers? If they aren’t normal full dauers, it seems questionable using them in an assay on dauer exit. Perhaps the UAS::ICE integration site or a mutation in<br /> that background caused by integration affect the dauer state? Given these concerns, I would recommend cutting this experiment from the paper. The laser ablation experiment in Fig 7C is much cleaner.

13. In Fig S3, it is stated that the labeled neuron is actually due to bleedthrough of RFP in a coinjection marker. This coinjection<br /> marker should be listed in the genotype of this strain in Table S1. Also, it appears that there might also be some GFP in the posterior pharynx in this image. Any explanation?

Reviewed (and signed) by Michael Ailion

On 2021-10-22 17:33:40, user Gregory Way wrote:

Thanks for posting this paper - there is a lot here to digest! I have one comment about, perhaps, a critical assumption that may not be true.

You cite reference 23, which is Cutiongco et al. 2019 (https://doi.org/10.1038/s41..., and state: "The assumption is that cells with more similar morphological features (e.g. cell size, nuclei size, granularity, or distribution) also have more similar transcriptional profiles23"

In Cutiongco et al., they tested 14 genes, and observed that they were associated with concomitant morphological changes as a result of different nanotopographies. I am not sure if this relationship will hold true for other genes and especially, outside of using nanotopography as an environmental stimulus. Also, evidence from other domains indicates that morphological differences may not necessarily indicate large gene expression differences (see Que et al. 2020: https://doi.org/10.1038/s41... where they found only 9 genes differentially expressed between neurons with different morphologies).

I suppose my point is that the field needs more evidence of the linkage expectation between expression and morphology, and, to the extent that this assumption impacts normalization, I would be a bit weary of over-interpretation.

On 2021-10-22 13:42:27, user disqus_ZTJxGjMaUL wrote:

what was the origin sample of the RNA fragments this test is tuned to?

On 2021-10-22 09:59:50, user Marc Gielen wrote:

Interesting read, thanks!<br /> Just two quick comments on figure 6 suppl 4 :<br /> 1) there is a microscopic reversibility issue in the kinetic scheme, which would be solved by increasing the AD --> ADI by 10^3 fold (i.e. k'on.10^3 rather than k'on)<br /> 2) by increasing 1000 fold the desensitization on-rate, it seems you are decreasing the lifetime of the open state rather than stabilizing the desensitized state (of course, there is a thermodynamic shift in favour of the desensitized state, but it doesn't sound to me like a stabilization in a kinetic sense). My bet is that if you rather decrease the desensitization recovery rate (d-), akin what you did for the flipped state in fig 6 suppl 3, you will end up with a similar observation to what we had for our 2018 review (GLIC & DHA): pretty much no effect on the peak current following preapplication of the inhibitor.<br /> Best,<br /> Marc

On 2021-10-22 06:28:34, user CC C wrote:

This is the first AI-designed PETase and it works so great with Microbes

On 2021-10-21 17:34:05, user Iklwa wrote:

Chimera formation during PCR is common.

On 2021-10-20 20:47:30, user Simón Villanueva Corrales wrote:

Hi, exciting development! As I was reading the paper and the main point of Polypolish is to reduce errors in repetitive regions, I was almost sure that Merqury was going to be used to assess the error reduction. Have you heard about it before?

I have not heard of ALE before, but by your description, it seems a mapping-based score, which would skew to non-repeats. I think Merqury might be a better fit because it is kmer-based, and thus it does not suffer from the mapping bias. Among other things, Merqury can provide a global QV score for your assembly, comparing the assembly kmer spectra to the short-reads kmer spectra. I suspect Merqury's QV scores can better reflect the reduction in repeat regions.

Furthermore, I wonder what one can achieve if you use both metrics. Being ALE a good representative of non-repetitive error reduction and Merqury's QVs of the global error reduction, one might think that something like the ratio or the difference of both could directly correlate to error reduction in repetitive regions. Although one would have to think about how to normalize the metrics to make them comparable, I think it is still an idea worth exploring.

I hope this comment helps and I wish you the best.

Regards,

Sivico

On 2021-10-20 20:03:08, user luis carlos LOPEZ GARCIA wrote:

This preprint was published on October 13th in Biomedicines (https://www.mdpi.com/2227-9....

On 2021-10-20 12:09:55, user Marcelo Briones wrote:

This work has been published in the journal "Viruses" (MDPI): https://www.mdpi.com/1999-4...

On 2021-10-20 01:58:44, user Joshua Mylne wrote:

I commend the authors for putting this openly online and I have some suggestions:

With the data shown, pqt11 does not to me seem hypersensitive to PQ. The error bars overlap and the images not compelling. Could I suggest you try more than one concentration of PQ and several sub-lethal doses? Also, only one pqt11 SALK allele is used. Others are available e.g. SALK_109667, SALK_148492 (confirmed homozygous) and more. Although it makes sense loss-of-function is sensitive if OX is resistant, it does not alway happen like that and needs better proof.

It would be nice to see the proteins are actually made in E. coli. There are no protein gels. Also, were you surprised that without adding IPTG you got life-saving expression of PQT11 from pET28a in E. coli on plates? I didn't think it was a very leaky plasmid?

Also, the in vitro assays you did not include a NADPH-cytochrome P450 reductase (e.g. ATR1 or ATR2)? I thought that was required for in vitro assays with P450s?

Finally. Did you consider trying to detect the N-demethyl paraquat in planta for WT vs pqt11 vs OX PQT11?

Happy to discuss offline if you prefer. Thanks for sharing your work openly online. I thought it very exciting.

On 2021-10-19 18:37:14, user Jin Yu wrote:

Accepted version:<br /> https://escholarship.org/uc...

On 2021-09-30 17:05:14, user Jin Yu wrote:

Published with MSDE from Royal Society of Chemistry:<br /> https://doi.org/10.1039/D1M...

On 2021-10-19 15:41:45, user Maria Paula Volpi wrote:

This preprint has just been accepted for publication in the journal Applied Microbiology and Biotechnology. Once the DOI is out, it will be posted here

On 2021-10-19 14:28:40, user Kimmo Palin wrote:

As pointed out in Twitter, competing interest statement needs updating

https://twitter.com/passDan...

On 2021-10-19 09:48:52, user Paul Macklin wrote:

Paper is now published as:

https://doi.org/10.1016/j.i...

On 2021-10-19 07:11:57, user Stefan Barakat wrote:

the peer reviewed version of this paper appeared now in Genome Medicine: <br /> Genome Med 13, 162 (2021). https://doi.org/10.1186/s13...

On 2021-10-18 07:49:31, user Ken Shirato wrote:

This preprint has been peer-reviewed and published as an article in Molecules. The information of the paper is shown below:

Title: "Standardized Extract of Asparagus officinalis Stem Attenuates SARS-CoV-2 Spike Protein-Induced IL-6 and IL-1β Production by Suppressing p44/42 MAPK and Akt Phosphorylation in Murine Primary Macrophages" by Shirato K, Takanari J, and Kizaki T<br /> Journal: Molecules 26(20), 6189, 15 pages, 2021.<br /> Doi: 10.3390/molecules26206189<br /> URL: https://www.mdpi.com/1420-3...

Ken Shirato, Ph.D.<br /> Kyorin University School of Medicine

On 2021-10-18 00:11:14, user Ellis Patrick wrote:

We also found that the choice of signature database made a big difference https://journals.plos.org/p...

I'm not sure if our dataset would help your validation?

On 2021-10-16 22:53:39, user Kaotic wrote:

Any update on when this may be peer reviewed?

On 2021-10-16 12:38:42, user Amartya Sanyal wrote:

This preprint is now published in Cancer Informatics (2021) and a link to the publication will be forthcoming.

Analysis of Aneuploidy Spectrum From Whole-Genome Sequencing Provides Rapid Assessment of Clonal Variation Within Established Cancer Cell Lines.

Ahmed Ibrahim Samir Khalil, Anupam Chattopadhyay, Amartya Sanyal

Cancer Inform. 2021; 20: 11769351211049236. Published online 2021 Oct 16.<br /> https://doi.org/10.1177/117...

PMCID: PMC8521761<br /> https://www.ncbi.nlm.nih.go...

On 2021-10-15 20:43:00, user Swineheart wrote:

I am wondering why this paper is stuck in preprint status. I can review if needed, having already rerun all the analysis independently.

On 2021-10-15 14:52:22, user Laurent Thomas wrote:

Hi, interesting work, I was just wondering about the choice of the U-Net architecture, ie a segmentation network, while the final readout is actually classification (event in the image yes/no). Is there a reason why you chose a segmentation architecture over an image-classification architecture, was the latter not efficient ?

On 2021-10-14 23:54:29, user Heinz V Bergen wrote:

You need to fix both abstract and PDF: You state W. somnifera (ATRI-CoV-E2) and further down we find W. somnifera (ATRI-COV-E5), which is it?

On 2021-10-14 15:16:12, user Claire Dessalles wrote:

This article has been accepted at the journal Biofabrication, and can be<br /> cited with its new DOI. <br /> https://iopscience.iop.org/...

On 2021-10-13 19:32:14, user Anon Anon wrote:

There are several things about the data presented here, along with the author’s interpretations, that are concerning and warrant close scrutiny.

1. From the Results section, it is unclear how many loci were retained for analysis. There was, evidently, more than one dataset that was generated because the authors refer to a “main dataset” in the Methods and to more than one dataset (with differing numbers of loci) in the Results. Also, at line 215, they state “3.935 samples remained from 55 individuals”. How can a fraction of a sample remain? Is a sample an individual animal? At line 219, they state that there is a dataset with as few as 7 SNPs. This is very concerning as any analysis on a RAD dataset consisting of 7 SNPs will likely not have any power to address their research questions. It is not stated which analyses are applied to which datasets; therefore, it is not possible for the reader to decide for themselves whether their results can be trusted.

2. There are two oddities in the PCA that need to be explained. First, the PCA is not centered on zero. Additionally, it shows that 98% of the variation in the data set is explained by the first PC; however, there is no separation of individuals into clusters along the first axis. The authors ignore both of these incredibly odd patterns and focus on the 2nd axis to explain that some population structure is revealed by the second PC. I’ve never seen a PCA with these two odd patterns in the published literature, and I don’t know how to explain it (neither do the authors, evidently, since they made no attempt to interpret what is responsible for explaining 98% of the variation in their dataset), but it reveals a serious flaw in either the dataset or methods of analysis and should have been a warning flag to the authors. No other results from this study can be trusted until the oddities of the PCA can be explained in such a way that lends confidence to the quality of the dataset itself.

3. Much of their conclusions are based on the results of STRUCTURE, but the STRUCTURE graph is not provided within the paper (nor is the supplemental documentation provided). The authors state that the best value for K is 3. If the pattern in the STRUCTURE graph does not show three clusters that correspond to geographically sampled locations, then this would need to be explained. It would be interesting to see what the STRUCTURE plot actually looks like given that the PCA is uninterpretable (the PCA and STRUCTURE plot should show comparable patterns). In the Discussion, the authors say that the inferred best value of K doesn’t match the STRUCTURE plot, stating at line 303 “However, examining the plots suggests that samples represent a single interbreeding population”. This discrepancy would need to be explained, and their explanation of more clusters, sampling scheme, and newly colonized subpopulations (lines 304-308) are very difficult to understand.

4. The authors cite a study by Pfau et al. (which appears to be the most relevant study of this species) and compare their findings in the following passage. However, they misinterpret the findings of Pfau et al. and apparently do not understand the data upon which Pfau et al. drew their conclusions.

The authors state at line 272 “More recently, [Pfau et al.] observed low mitochondrial DNA variation but high microsatellite diversity within the species. They concluded that genetic drift and not gene flow has had a greater impact on configuring D. elator genetic diversity. This result is possible because mitochondrial DNA has a lower effective population size than neutral nuclear markers such as RAD loci. Genetic drift could play a role in structuring mitochondrial DNA diversity, but more time would be needed to detect reduction of diversity in the nuclear genome using older markers such as microsatellites. An insufficient number of polymorphic microsatellite loci limits genetic resolution between individuals with supposed low population-level diversity. Our results suggest that RAD loci, that have a slower rate of mutation than microsatellites, are superior when investigating populations with weak population structure.”

The first sentence is correct. However, the second sentence was only a portion of the conclusions of Pfau et al. and unrelated to mtDNA variation because that particular conclusion was based only on the microsatellite dataset. Pfau et al. actually concluded-- using microsatellite markers--that "All methods addressing population structure indicated that the Iowa Park population was divergent from the others, with Vernon and Harrold showing a somewhat intermediate relationship but with a closer affiliation with Quanah than Iowa Park, despite their closer proximity to Iowa Park. This pattern did not conform to isolation by distance, thus genetic drift appears to have played a greater role than gene flow in establishing genetic structure."

Given the wording in the paragraph quoted above, the authors appear to be conflating the results from the mtDNA and microsatellite markers in Pfau et al. While it is true that mtDNA has a lower effective population size than nuclear marks such as RAD loci, Pfau et. al did not use RAD loci—they used microsatellites. It is also true that “Genetic drift could play a role in structuring mitochondrial DNA diversity, but more time would be needed to detect reduction of diversity in the nuclear genome”, but Pfau et al. provided explanations for why the mtDNA diversity was so low despite relatively high microsatellite diversity—going beyond just differences in effective population size.<br /> Furthermore, Pfau et al. found that nuclear microsatellites DID reveal population structure but that mtDNA did NOT reveal population structure (because there was essentially no mtDNA diversity which could be partitioned). The authors go on to say that “insufficient number of polymorphic microsatellite loci limits genetic resolution between individuals with supposed low population-level diversity”; however, microsatellite diversity was actually relatively high. The authors go on to state that RAD loci are superior to microsatellites. This is correct, but only because RAD datasets typically contain many more loci than microsatellite datasets. Surprisingly, the authors evidently didn’t notice that the microsatellite dataset of Pfau et al. actually revealed MORE genetic structure than their own RAD dataset (the opposite of what they predicted when comparing the two markers).

1. The authors state “Our samples were collected on opposite sides of a cline, separated by a region of inaccessible private land, so it was difficult to determine if the slight differentiation is due to that distance or if there is true population substructure and isolation from other habitat patches” (line 315). A cline is found when there is a continuous distribution, but since they also state that it is unknown if populations exist between their sampling locations, how do they know that they’ve sampled opposite ends of cline, two genetically isolated populations, or two sides of a panmictic population?

They attempted to use historical samples to fill in this distributional gap, but doing so ignores the likelihood that allele frequencies have changed within these populations over the sampled time frame. In addition, their predictions themselves are incorrect and difficult to interpret. They say “if the contemporary east and west subpopulations were indeed distinct” (line 320), which I interpret from the context to mean that the populations are geographically isolated rather than continuous. Given this hypothesis, they predict that the geographically intermediate population “the sampling hole” would be genetically intermediate between the two ends. This is actually the opposite of what is expected from population genetic theory which predicts that isolated populations diverge randomly with respect to one another. They predict the alternative hypothesis--that the species is one population--to reveal “greater differentiation between them and our sampling hole samples”. I do not understand what this statement is attempting to describe, but this hypothesis is the one that would predict the geographically intermediate population to be genetically intermediate—but only if gene flow is restricted sufficiently to produce isolation by distance. A panmictic population would show all three locations to be genetically homogenous.

1. The authors appear to make an attempt to discuss their findings in the context of metapopulation theory (lines 349-358), but it’s confusing because of their attempted elaboration on what constitutes metapopulation criteria. They never actually make an connection between their genetic findings and metapopulation theory, but using other sources of information, state “this connection to metapopulation theory is still tenuous”. Yet they go on to state “Should managers elect for extreme measures to manage D. elator populations, such as translocations or reintroductions, knowledge that the population is a metapopulation is critical”. It seems especially careless to state that it is critical that managers recognize the species as a metapopulation when the connection is still tenuous.

In conclusion, the striking oddities in the PCA demand a reanalysis on the quality of the dataset itself. If the dataset is in question, the results of all other analyses cannot be trusted. Furthermore, the many instances in which the authors misinterpreted their own results and made incorrect predictions from their hypotheses, are further indications that this study needs much attention before it can be used to understand the genetic patterns of Dipodomys elator for purposes of conservation and management.

On 2021-10-13 11:33:20, user Martin Humphries wrote:

An interesting paper. G3BP1, G3BP2, DDX3X, and RBM3 are all found in the meta adhesome defined in "Definition of a consensus integrin adhesome and its dynamics during adhesion complex assembly and disassembly" (PMID: 26479319).

On 2021-10-11 14:54:04, user A Scientist wrote:

I would like to raise some concerns about the methods used that I think put into question the conclusions regarding non-coding RNA functions:

1) Puro-PLA cannot be used to accurately localize translation, since contrary to what has been believed, even in the presence of elongation inhibitors the nascent polypeptides are released and can quickly diffuse in the cytosol. This has been quite convincingly demonstrated in the two reports listed below. Thus, the data presented cannot support a non-coding RNA function. They are more consistent with the idea that the mRNAs categorized in this work as ‘translated’ encode proteins that are FA constituents, whereas the mRNAs categorized as ‘untranslated’ simply encode proteins that do not associate with FAs and likely diffuse to other locations after translation.

Enam SU et al. ‘Puromycin reactivity does not accurately localize translation at the subcellular level’. (2020) eLife doi: 10.7554/eLife.60303

Hobson BD et al. ‘Elongation inhibitors do not prevent the release of puromycylated nascent polypeptide chains from ribosomes’. (2020) eLife doi: 10.7554/eLife.60048

2) The experiments using RNAse A-treated cells are perplexing. The authors are apparently assuming that RNAse A is cell-permeable, but that should not be the case. Are they certain that they are not looking at an effect caused by degradation of some extracellular RNAs? Different controls would be needed than what is provided in Figure S4A (RNase A will obviously degrade RNAs after the cells are broken up, but that cannot be taken to mean that RNAse A which was just added in the media was taken up by living cells and degraded intracellular RNA).

3) Ribosome profiling is performed from FA preparations that are isolated in buffer of low ionic strength and without Magnesium. However, under these conditions, ribosomes are expected to dissociate. Therefore, it is not surprising that no translation is detected for most mRNAs.

4) The fact that proteins encoded by FA-associated mRNAs are not detected as proteins in FAs cannot be taken as evidence for lack of translation. It is equally likely that the translated proteins diffuse away and have roles other than being FA constituents.

On 2021-10-12 22:38:55, user Larry Weisenthal wrote:

Here is what is of additional importance: At the time I submitted the preprint to biorxiv.org, the policy was not to accept actual clinical medicine data. So the editors obliged me to omit one of the most important points of all: to wit, human sera from different donors vary widely in ability to inhibit MCED.

REFERENCE: ascopubs.org/doi/abs/10.120...

I believe that differences in MCED inhibitor factor (in serum) explain differences in both response to bevacizumab in cancer therapy and susceptibility to atherosclerosis induced by pathogenic lipids. A potential treatment for atherosclerosis lies in identifying the serum factor (which is present in some people and absent in others) which inhibits MCED. Study differences in sera from high MCED inhibitor humans and low MCED inhibitor humans and isolate the MCED inhibitor. Turn this into a drug, and there could be a novel treatment/preventative for atherosclerosis.

On 2021-09-17 23:05:52, user Larry Weisenthal wrote:

See also: https://ascopubs.org/doi/10... AND https://vimeo.com/manage/vi...

On 2021-10-12 13:47:56, user Luisa Hugerth wrote:

The github link in the abstract leads to a 404 Page Not Found error. Maybe the repository is mistakenly marked as private?

On 2021-10-12 05:12:08, user Huigang Shi wrote:

Peer reviewed version is available! Low-cooling-rate freezing in biomolecular cryo-electron microscopy for recovery of initial frames https://cup.org/3v4Tah1

On 2021-10-11 22:19:36, user Jeremy Leipzig wrote:

We host the OQFE pipeline at Truwl https://truwl.com/search?q=...

On 2021-10-11 17:10:21, user Jessica wrote:

This is a really cool concept, thank you for sharing this method!

On 2021-10-11 16:32:32, user sanxiong liu wrote:

This preprint has been published on Molecular Cell and a link will be forthcoming. You can read the published version at https://www.cell.com/molecu....

On 2021-10-11 07:53:57, user S. Vianello wrote:

Is the 3D dataset only accessible by having to create an Elsevier account? I am unable to access it (even when clicking the "continue anonimously" option).

On 2021-10-10 19:23:38, user Equi Epi Vet wrote:

https://minister.awe.gov.au...

On 2021-10-10 11:22:11, user Zarul Hanifah wrote:

In the abstract, it says "77% of the SAR11 community was compromised of a small number of ASVs (7 of 106 in total). In this phrase, are you saying 77% of the SAR11 community by relative abundance? Which means the remaining 99 SAR11 ASVs made up the remaining 23% of SAR11 community relative abundance?

Also, should it be comprised or compromised?

On 2021-10-09 12:55:35, user Omar Vesga wrote:

Now published in PLOS ONE doi: 10.1371/journal.pone.0257474

On 2021-10-07 17:08:05, user Omar Vesga wrote:

This preprint is already published with all data in a peer-reviewed journal:

https://journals.plos.org/p...

https://doi.org/10.1371/jou...

On 2021-10-08 19:43:51, user Natascia Marino wrote:

This is the first molecular landscape of the NORMAL breast

On 2021-10-08 13:00:30, user Boštjan Kokot wrote:

The paper has been published online in Nanotoxicology journal on 6th of October 2021:

https://www.tandfonline.com...

On 2021-10-08 12:04:39, user Karel Muller wrote:

Very nice work, indeed. Although, the title sounds to "weak". :) I really appreciated the focus on role of individual GH3 genes in plant development. And I believe that there is still much to understand between inactivation of GH3 activity of the overall phenotype. I would like to know whether authors tested levels of other auxin metabolites in their material as well as for example transriptomic profiles of those plants. Thank you.

On 2021-10-08 11:58:06, user Eric Fauman wrote:

There are also lots of typos in the supplementary tables: e.g.: "signnifcant"

On 2021-10-08 11:57:25, user Eric Fauman wrote:

There are many p-values listed as 0 in the supplementary tables. You need to either report the -log10(p), or include the standard errors and subject counts for each variant so researchers can calculate the p-values for themselves.

On 2021-10-08 07:15:47, user ME wrote:

Note: this article is now published in Ecography :

https://doi.org/10.1111/eco...

On 2021-10-08 04:14:10, user Jeff Brender wrote:

Title is probably missing a few words:<br /> "The importance of residue-level filtering, and the Top2018 best-parts dataset of high-quality residues coordinates in protein structures"

On 2021-10-08 01:28:43, user Adrian Flierl ???????? wrote:

There is no question that ANTs are essential in mitigating environmental and cellular stress.<br /> Regarding the hypothesis of functional ANT at the epithelium cell membranes, these extraordinary claims require extraordinary evidence. <br /> In general, there are several technical and methodological points of concerns:<br /> Overexpression of ANT (mitochondrial ANT content is tightly regulated) can lead to miss-targeting, sorting or even excretion from cells, especially in cells with a high secretory capacity. It also has been shown that a significant portion of mitochondrial proteins are excreted through vesicles and Exosomes. <br /> Technically, there is the possibility of imaging immuno-histo/cyto-chemistry artifacts due to unspecific binding of primary and secondary antibodies (entrapment), when employing fluorescence-immunocytochemistry in this notoriously difficult cell type.<br /> It would have been nice to have additional evidence for ANT localization to the cell membranes, either by immunohisto/cyto staining controls, a secondary detection method (higher mag or EM) or simple biophysical cellular fractionation (lipid fraction) and protein detection (western).

As much as I'd like to see ANT to also fulfill a role of regulating airway epithelial cell membrane function, I would have to see additional evidence that would support the significant functional presence of ANTs at cytoplasmic membranes.

On 2021-10-07 19:41:10, user aquape wrote:

Congratulations with this paper that beautifully explains *how* we lost our tail. The *why* is perhaps less difficult: Miocene Hominoidea were "aquarboreal" (aqua=water, arbor=tree) in swamp forests: they frequently waded bipedally with stretched legs, and climbed arms overhead in the branches above their head. Nasalis larvatus (proboscis monkeys) often wade upright in mangrove forests, and already evolved shortened tails. Aquarborealism also helps explain why we became much larger than monkeys, why hominoids are also called Latisternalia ("broad-breastboned ones"), why humans & apes have broad thorax & pelvis, with dorsal scapulas, lateral movements of arms & legs, and more centrally-placed vertebral spines (monkeys have narrow bodies, laterally-places scapulas, dorsally-placed spines etc.). See e.g. our Trends paper (TREE 17:212-217), google "Aquarboreal Ancestors".

On 2021-09-20 14:18:42, user Bo Xia wrote:

Please note that the email address of Bo Xia should be Bo.Xia@nyulangone.org

On 2021-10-07 05:29:08, user Titus Ponrathnam wrote:

An additional question. Have you used this system with G-trace and LacZ? The 2017 paper which uses tetOff had used lacZ especially in their age related analysis, so it would be nice to attempt a few laZ stainings as a way to answer Laurent's question in older adults.

On 2021-10-06 20:34:51, user Titus Ponrathnam wrote:

This is very cool. On what chromosomes do you have the inserts, and when would you be comfortable to share the flies?

On 2021-10-06 17:15:23, user A.J. wrote:

Just a heads up VP1 stands for vacuolar-H+-pyrophosphatase not vacuole protein 1.

On 2021-10-05 18:24:10, user Marc RobinsonRechavi wrote:

Just after the discussion, the authors write:

The source code and weights for the trained models will be made available shortly.

This is a publication, i.e. it is made public as part of the scientific record and is citable, thus I strongly invite the authors to make the corresponding code available without delay.

On 2021-10-05 16:41:18, user Dr Anne Bishop wrote:

I found the paper interesting, but the current version of the manuscript reads as if there is no biological replication, just one culture of each of the two strains that was sampled for both RNA and metabolite profiling at three time-points. I thought I may have missed it when I read it, but searching the document for terms like "replicate" also yields nothing. In Figure 3 the PCA implies there are samples 3, 4 and 5 within each strain, but there is no explanation of this in the methods, legend or results text. I can't work out how t-tests for significance for 3h vs 4h or 3h vs 5h in Table 1 could be calculated if there is no replication to provide a measure of biological variation within the experiment. Please could the authors clarify what biological replication was used, if any?

On 2021-10-05 07:11:42, user Pradeep Gautam|प्रदीप गौतम wrote:

The current version of the manuscript is online at Supplementary figure 5. https://www.nature.com/arti...

On 2021-10-04 23:54:47, user Esperanza Bas, PhD wrote:

Knowing the titers of antibodies is a good start. I would encourage the authors to look into activation and response of memory T and B cells to APCs primed with antigens against different variants. Memory cells will remain long after the antibodies are cleared out of our bodies.

On 2021-10-02 20:13:44, user Whimsy wrote:

This is an important analysis. I suggest the authors provide the demographic information in this manuscript (age, sex, education and ethnicity) from whom the samples collected for these analyses was performed instead of linking to another paper. Some discussion on how this durability of traditional vaccines (such as the flu) compares to this finding would also help put things into context in respect to immunological memory incurred from mRNA based vaccines that have no endogenous adjuvant to engage immune co-activation typically required for immunological memory and training. - Shirin Kalyan, PhD

On 2021-10-04 12:07:39, user yadij wrote:

"Surprisingly, the response to glucocorticoids has not been previously examined at single cell resolution."

scRNAseq has been used to examine the response to glucocorticoids in several contexts, including:

https://doi.org/10.1126/sci...<br /> https://www.nature.com/arti...<br /> https://www.pnas.org/conten...

On 2021-10-04 11:06:26, user Shishir Gupta wrote:

The article is now published in 'Microorganisms'<br /> https://www.mdpi.com/2076-2...

On 2021-10-04 09:26:16, user Fernando Racimo wrote:

In "Ancestral contributions to contemporary European complex traits", Marnetto et al. look at enrichments of ancestral contributions to the genetic component of particular traits to modern individuals. They apply a newly-developed method to do so to a large cohort of individuals from the Estonian Biobank and find different contribution from ancestral populations on variants associated with pigmentation, anthropometric traits and blood cholesterol levels, among others. They also look at patterns of positive selection on a subset of these variants. The manuscript presents an interesting and extensive set of analyses informing on a fascinating question about the evolution of traits over time and adaptation, as well as the genetic make-up of trait variation in Estonia. The authors also introduce a useful statistic to measure these contributions. I detail some comments below that I think might help improve the text and analyses:

• While covA is extensively explained in the Methods, it has too brief of a description in the beginning of the Results section, given its importance to the associated conclusions. I would dedicate a few more sentences to explaining the reasoning behind covA, before assuming that the reader will automatically know what one means as they continue reading the results. I was a bit confused at first as to what covA was exactly measuring, and I think it's a smart way of measuring ancestral contributions to traits, so a bit more motivation for its use would be useful.

• I think it might be good to move the text describing the connection between covA and f-statistics to the Methods section, for people familiar with f-statistics to motivate its use this way. I also wouldn't say that "covA(i,j) has no interpretation in terms of branch length because of the fictitious nature of pA, an allele frequency which only serves as balanced comparison for the ancestries under analysis". After all, the "populations" used in f-statistics analysis are also artificial groupings of individuals that share more or less history, depending on the analysis at hand.

• It might be useful to try to separate the Eastern Hunter-Gatherer and Caucasus Hunter-Gatherer into 2 ancestral candidate groups, rather than treating Yamnaya as an ancestral source. EHG, CHG and WHG have quite differentiated component ancestries, so this might help solve some of the correlation issues mentioned at the bottom of page 3? See, e.g. Lazaridis et al. 2017.

• The study would greatly benefit from examining the behavior of the covA statistic using simulations of a phenotype (say, in SLiM, see, for example, the recent preprints by Yair and Coop (2021) and Carlson et al. (2021)). The authors mention the interdependence inherent to the fact that the same reference populations are used, and I am concerned about what other factors might contribute to the behavior of the statistic, including, for example: 1) varying levels of negative selection operating on a trait, 2) the bias inherent to the fact that some ancestral populations might be closer to the present-day population on which the trait was measured, 3) effective size of the ancestral contributing population under study, 4) heterogeneity of the average "reference" populations used as baselines for comparison and 5) sample sizes of the ancestral populations (e.g. low sample size for the Siberians).

• The authors say that "those counties for which the covA distribution is significantly different than the rest of Estonia (two-tailed Wilcoxon-Mann-Whitney test, p ≤0.001)." It seems like a large number of counties have an asterik. I wonder if *all* counties would have an asterisk if one had chosen a more standard p-value cutoff? Why was this cutoff chosen? What would be the null sampling distribution here for what one would call significant? That each county would have the same distribution of this statistic as the whole country?

• Following up on this, it seems a bit strange to say that one intends to control the pleiotropies that exist between traits by avoiding overly stringent multiple testing corrections as Bonferroni. Without an alternative way to control for multiple testing (e.g. FDR, or looking at genetic correlations between traits), I don't think the best course of action is to err on the side of anti-caution and go with a vanilla P-value cutoff designed for single tests. What does "significance" in this context tell one about the importance of different ancestral contributions?

• I commend the authors for adding a city/countryside residency covariate in their model, to control for socio-economic effects. I wouldn't suggest, however, that this covariate allows one to entirely control for other socio-economic effects (for example, socioeconomic or cultural effects that might not be well explained by place of residence). This is a very hard problem and the subject of extensive research (see for example Mostafavi et al. 2020). Maybe add some cautionary statements along these lines?

• I'd be happy to also share this review on other preprint peer review venues like Review Commons, PCI or Peerage of Science if the authors so desire. Thank you for posting this very interesting manuscript on bioRxiv and giving me the chance to review it!

On 2021-10-04 08:50:43, user Ramon Crehuet wrote:

I find it intriguing that AF2 outperforms experimental structures (Fig. 3A) when used to predict variant effects. That comes from the regions where AF2 has high confidence but where it could not use a template (Fig 2B)! Is this reporting on the quality of experimental structures? And does it depend on the experimental method? I guess this could be obtained from data in Fig. 6, but it is not straightforward.

The fact that the template reduces the quality when pLDDT also suggests that AF2 could be improved by allowing it to discard a bit experimental templates, right?

On 2021-10-03 05:35:59, user Ingo Bading wrote:

May be you should look more for plants like Amaranth than for Millet as the C4-plants in question? The mentioning of millet seems to be outdated (1). David Anthony writes 2007 about the Late Bronze Age Samara Region (2):

"The earliest permanent year-round settlements in the LBA contained no evidence of agriculture but abundant evidence for the gathering of wild plants - the nutritious seeds of Chenopodium and Amaranthus, which can grow in dense<br /> stands as productive in seed yield per hectar as einkorn wheat. (...) Wild plant resources have been largely ignored in arguments about the productive capacity and potential autonomy of steppe subsistence economies."<br /> _____________

1. Dragana Filipović, John Meadows, […] Tanja Zerl: New AMS 14C dates trackthe arrival and spread of broomcorn millet cultivation and agricultural change in prehistoric Europe. Published: 13 August 2020, Scientific Reports volume 10, Article number: 13698 (2020), <br /> https://www.nature.com/arti...
2. Anthony, David; Brown, Dorcas: The Herding-and-Gathering Economy at Krasnosamarskoe, Russia, and the end of the dependency model of steppe pastoralism. In: Social Orders and Social Landscapes, 2007, hrsg. von <br /> Charles W. Hartley, Laura M. Popova, Adam T. Smith, S. 393ff (see Google Books)

On 2021-10-02 21:20:51, user Darryl B McConnell wrote:

Great apo-SFX structure. This is the first apo-streptavidin structure with (for Chain A) a (I assume) fully solvated pocket with 7 waters. In the previous room temp structure 1SWB there were only three and for cryo structures you get mostly the cryo solvent. Well done. I wonder if the SFX approach is generally better for observing water molecules. If so this could be quite powerful for drug design.

On 2021-10-02 21:14:31, user Travis Wheeler wrote:

The results SEEM relevant and important ... but without code to test/review, there's really not much to say about the paper. The preprint has been posted for 2 months, but no RGN2 code is available; please share it.

On 2021-10-02 20:55:40, user Thomas Hanigan wrote:

Dear authors,

Thank you for sharing your work. Given the increasing importance of disseminating science through preprints, I want to help support research in drug development and metabolism by providing feedback and thoughts on preprints in these fields. Your preprint caught my eye, and I hope you will find the comments below useful.

This is an interesting paper proposing a method to target ARID1A mutant cells through PLK1 inhibition. The authors also show evidence that this mechanism is independent of PLK1's canonical role as a mitotic kinase and identify/investigate a novel function of PLK1 in mitochondria using an unbiased approach. Overall, the authors’ claims are well supported by initial evidence and communicated effectively but could be strengthened with additional experiments. Although, who hasn’t heard this before…<br /> Although there appears to be a difference in sensitivity to volasertib in ARID1A knockout cells in two non-transformed and one transformed cell line via a single timepoint celltiterglo and colony formation assay, the fold change in inhibitor potency is modest (Figure 1C/E), and this evidence connecting PLK1 and ARID1A could be strengthened by 1) comparing the growth rate of ARID1A knockout and control cells. By considering differences in growth rate, the authors might find an overall greater difference in volasertib sensitivity between ARID1A KO versus control cells. 2) adding an additional 12-point dose-response curve to Figures 1C and E using a non-selective cytotoxin (i.e. paclitaxel) in ARID1A KO/control cells, 3) adding a dose-response curve using the known SWI/SNF mutant selective OxPhos complex I inhibitor IACS-010759 (https://www.ncbi.nlm.nih.go.... As the authors find the underlying ARID1A vulnerability is mitochondrial in origin, IACS-010759 would be a useful control in other experiments throughout the paper as well.

The second main conclusion, which leads the authors to conduct a CRISPR screen, is that PLK1 inhibition does not differentially alter cell cycle in ARID1A KO versus control cells. Although this is clearly novel, it is a hard pill to swallow given the literature on SWI/SNF and PLK1 regulating cell cycle, and the connection between mitochondria fusion and cell cycle progression. I urge the authors to conduct additional experiments here. For example, make sure that the pHH3 staining in ARID1A KO versus control cells treated with 10nM Volasertib (Figure 2C) is not significant, use additional concentrations of volasertib (dose response for functional assays) and further investigate cell cycle checkpoints/signaling. The authors could also show the results from Gene Set Enrichment Analysis (GSEA) for mitosis/cell cycle related genes in addition to Ox Phos/Reactive oxygen species in Figure 4H/I. <br /> Beyond Figure 2, it would be beneficial to verify the basal ECAR rates in addition to oxygen consumption in Figure 4J/6B, as the difference in oxygen consumption between ARID1A KO/Control cells is minimal in Figure 4J, and as the authors found that ATP production between these cells was not significantly altered. I would assume that ARID1A KO cells would have diminished glycolysis and the difference in ratio of OCR/ECAR may help distinguish the overall significance of the difference in mitochondrial metabolism between ARID1A KO and control cells. Also, there is a discrepancy between Figure 4J and Figure 6B. Seemingly both are showing oxygen consumption via seahorse assay in GES1-ARID1A KO versus control cells, but the difference is substantially greater in Figure 6B? Additional controls for measurement of membrane potential and ROS, for example various OxPhos inhibitors, would also help pinpoint how an increase in oxygen consumption (Figure 4J/6B) could be accompanied by a loss in membrane potential (Figure 3a-f). Another useful control throughout the paper would be to reconstitute the ARID1A KO cells with a functional ARID1A construct and verify the functional changes in mitochondria are rescued.

The overall novelty of this paper is the role of PLK1 function and volasertib activity on mitochondria independent of cell cycle, and publication of these results could help shape the use of volasertib clinically, and our understanding of signaling between mitochondria and the nucleus. Although additional experiments would help substantiate the authors claims, the work is clearly important to communicate with basic researchers and clinicians looking at both mitochondria and cell cycle in relation to cancer, and this work helps solidify the functional relationship between these two fields.

Best,<br /> Thomas Hanigan

On 2021-10-02 19:15:17, user L. T. Fang wrote:

Here is a free SharedIt link for Nat Biotechnol: https://rdcu.be/cyJZk

On 2021-09-27 18:24:26, user L. T. Fang wrote:

A 2021 MAQC/SEQC2 presentation describing the method presented in this preprint:<br /> https://youtu.be/nn0BOAONRe8

On 2021-10-02 13:52:53, user Filippo Ferrario wrote:

The peer reviewed paper is now published. https://online.ucpress.edu/...

On 2021-10-01 16:39:48, user Robyn Wright wrote:

We discussed this preprint in a recent journal club meeting. Thanks for making it available as a preprint! You can read our thoughts on it here: http://dalmug.org/CAMI2_pre...

Best wishes,<br /> Robyn

On 2021-10-01 10:06:40, user Albert Cardona wrote:

Dear authors,

Thank you for putting together this very nice manuscript. I am writing to comment on the methods for tracking and analyzing Drosophila larvae. There are two most used methods so far. First, the multi-worm tracker (MWT) from Rex Kerr's lab initially developed for C. elegans (Swierczek et al. 2011) but then applied extensively to track populations of Drosophila larvae (e.g., Ohyama et al. 2013, Vogelstein et al. 2014, Ohyama et al. 2015, Jovanic et al. 2016, and more). Second, there's the system from Matthieu Louis' lab, Sensory Orientation Software (SOS; Gomez-Marin et al. 2012), which was used in subsequent projects (see citations to that paper. There is also JAABA from Kristin Branson's lab, which you already cite. Would be appropriate to compare the pros and cons of these softwares, if space allows it, in the discussion.

On 2021-10-01 02:23:20, user John McBride wrote:

Thanks for this work, it's good to see someone evaluating AlphaFold in this way.

Maybe I missed something in the article, but I fail to see why one should expect differences in pLDDT to correlate with either changes in stability, or fluorescence. To my understanding, changes in pLDDT can be either that the algorithm performs worse, or that the protein is predicted to be more disordered / flexible. My personal impression is that there is no robust rationale for expecting a priori that pLDDT should strongly correlate with either stability, or fluorescence. This point is pivotal to the entire paper, so I'd be more convinced if it were addressed.

On 2021-09-20 16:07:37, user Rath R. Weird wrote:

Not sure that entries like that would be sufficient to offset the advertisement machine unleashed in support of the AlphaFold, but I do welcome a systematic attempt to evaluate AlphaFold performance in realistic applications. There is lots of anecdotes out there, when people turned to AlphaFold for structural info, and got gobs of disordered strings in return. Personally I compared some of the 2-3 year old models built by I-TASSER to AlphaFold output and found no discernible difference - typical homology modeling performance, none have much in looking-ahead capabilities, but at least the guys behind I-TASSER don't claim to have it. Here we have a more deliberate evaluation of the heuristic AlphaFold imbalance between physical realism and template alignment. Template alignment wins, to the detriment of physical realism.

On 2021-09-30 16:59:12, user Sofia Duarte wrote:

Dear colegues,

Please note that in our article "Vimentin filaments interact with the actin cortex in mitosis allowing normal cell division" (2019) Nat Commun 10, 4200 (https://www.nature.com/arti..., we reported that "vimentin filaments redistribute to the cell cortex during mitosis, forming a robust framework interwoven with cortical actin and affecting its organization", and describe the "Intimate cortical vimentin–actin intertwining in mitosis", which particularly affects "the abundance and distribution of f-actin" (https://www.nature.com/arti....

However, although we report that the C-terminal of vimentin is necessary for its interaction with the cell cortex in mitosis, we do not demonstrate an interaction of this domain with actin or a direct binding of the two cystoskeletal polymers. Moreover, in our work, we do not address potential interactions of actin with keratin precursors, as stated in the introduction of the article by Wu et al. (https://doi.org/10.1101/202....

On 2021-09-16 22:05:04, user Dyche Mullins wrote:

Vimentin Intermediate Filaments and Filamentous Actin Form Unexpected Interpenetrating Networks That Redefine the Cell Cortex

https://doi.org/10.1101/202...

Members of the Mullins laboratory read this paper together and discussed it during a recent journal club meeting. Following the meeting, we drafted the following set of comment with the aim of helping the authors revise the work for final publication..

Briefly, this manuscript describes in vitro and live-cell experiments aimed at understanding the connection between intermediate filaments and the actin cytoskeleton. As the authors note, this is an understudied topic, with many fundamental unanswered questions. The experiments described in the paper employ a variety of approaches, including rheometry, fluorescence microscopy, and cryo-electron microscopy. The major results claimed by the authors include evidence for extensive interactions between vimentin IF and actin cytoskeletal systems in the cell periphery and effects of IF networks on actin monomer diffusion.

Specific comments/concerns:

1. In the Introduction the authors state that it is “...generally thought that F-actin and VIFs form two co-existing but separate networks.” This is not an entirely fair description of the field. There is a significant body of work on interactions between intermediate filaments and elements of the actin cytoskeleton [e.g. myosin II/IF interactions described in Svitkina et al. (1996). J Cell Biol 135(4):991-1007], and how the actin cytoskeleton can regulate IF network architecture [e.g. this review by Chang and Goldman (2004). Nat Rev Mol Cell Biol. 5(8):601-13; and numerous research papers, including Schoumacher et al. (2010) J Cell Biol. 189(3):541-56 and Serres et al. (2020) Dev Cell. 52(2):210-222]. A more inclusive discussion of previous work on the interaction of the two cytoskeletal components would improve the presentation.

2. The title mentions the cell cortex, but there is almost no mention of the cortex per se in the Results or Discussion.

3. The figure captions should contain more experimental details. For example, captions should contain information on cell types, microscopy methods, and molecular probes (e.g. is the actin in Figure 1 phalloidin stained?). They should also contain other relevant details, such as buffer conditions, temperatures, and how many times an experiment was repeated.

4. The authors should consider combining Figure 1 and 2. It is a little unclear what is being highlighted by the arrows in Figure 1, as the two networks appear to be quite separate, even in this region of the cell. Figure 2 makes a better case for overlapping network elements.

5. In Figure 2, it is unclear whether the two networks are actually interpenetrating or simply present in different focal planes. Because the raw data includes z-stacks, it would be useful to show an XZ or YZ or 3d projection of the data.

6. Figure 2 should contain some (even rough) quantification of the percentage of observed cells that displayed such unambiguously interpenetrating networks.

7. The electron microscopy in Figure 3 is beautiful and quite convincing. The minimum distance argument, however, does not rule out the possibility that the actin and intermediate filaments are linked by plectins. The crosslinks could be present in areas where the filaments are further apart. Also, to convince the reader that they can distinguish between filaments based on width, the authors should provide a histogram of all the observed filament widths. Does the histogram show a clear bimodal distribution?

8. For Figure 4, please cite some previous studies using this indenter method and/or provide further details on the method: does it apply uniaxial or isometric stretch? Some raw images before, during, and after stretching would be more useful than the recovery plot shown in B.

9. As noted above the caption (or description) of Figures 4&5 should contain the number of biological replicates. P values calculated with n=number of cells (rather than the number of replicates) are not appropriate. We recommend simply removing the P value calculations.

10. Unfortunately, the data presented in Figure 5 are not convincing:<br /> a. Firstly, P values calculated on the number of cells instead of the number of biological replicates are artificially small and should not be reported. The trend is intriguing, but does it hold up over multiple rounds of experiment?<br /> b. Secondly, the FRAP bleaches both monomers and filaments. To look at only monomers, a method like FCS would be better.<br /> c. Instead of the curves in B, raw images would be more informative.<br /> d. When vimentin is knocked out, does the rest of the cytoskeleton change to compensate? Currently the explanation for these results is that there could be direct protein-protein interactions between Vimentin and G-actin which affect G-actin diffusion. The author favored this over the model where G-actin diffusion is limited by physical/steric obstruction of Vimentin networks. However the current data is unable to fully delineate these two models as the Y117L mutant forms a sort of intermediate filament (“intermediate” in length relative to the long filament), which intuitively could physically obstruct G-actin diffusion in an intermediate fashion (which is observed by this experiment). Perhaps using another mutant form of Vimentin that is simply unable to form any filaments but is present in cellular protrusions could help clarify the mechanism of G-actin diffusion limiting.

11. It is unclear what precisely the reader is to take away from Figure 6. The concentration of actin filaments is so high that one cannot make out individual filaments or their interactions with vimentin. It does not seem very surprising that two proteins can polymerize in the presence of one another. Is the point mainly that these two proteins can mix? Are there examples of the literature of other filament-forming protein phase separating from actin, or other conditions where this mixing doesn’t occur? A negative control would make this data more convincing.

On 2021-09-30 13:37:19, user Juliano Bordignon wrote:

Really nice data! Congrats.

Human neutrophils are not activated by Zika virus but reduce the infection of susceptible cells

On 2021-09-30 12:28:37, user Dr.D wrote:

I would recommend the authors include information on IBC approval for this work in the manuscript.