Multi-modal cryo-EM reveals trimers of protein A10 to form the palisade layer in poxvirus cores

Poxviruses are among the largest double-stranded DNA viruses, with members such as variola virus, monkeypox virus and the vaccination strain vaccinia virus (VACV). Knowledge about the structural proteins that form the viral core has remained sparse. While major core proteins have been annotated via indirect experimental evidence, their structures have remained elusive and they could not be assigned to individual core features. Hence, which proteins constitute which layers of the core, such as the palisade layer and the inner core wall, has remained enigmatic. Here we show, using a multi-modal cryo-electron microscopy (cryo-EM) approach in combination with AlphaFold molecular modeling, that trimers formed by the cleavage product of VACV protein A10 are the key component of the palisade layer. This allows us to place previously obtained descriptions of protein interactions within the core wall into perspective and to provide a detailed model of poxvirus core architecture. Importantly, we show that interactions within A10 trimers are likely generalizable over members of orthopox- and parapoxviruses.


Referee expertise:
Referee #1: virology Referee #2: virology, cryo-ET Reviewers' Comments: Reviewer #1: Remarks to the Author: A. Summary Poxviruses have a special place in virology owing to their profound impact on human health as both the cause and potential solution to some of the most dreadful diseases.The study of these viruses led to major discoveries in cell biology.Yet, significant blindspots remain in our understanding of their biology.This manuscript presents major advances in our understanding of the poxvirus architecture and composition.It extends the pioneering cryo-ET study by Cyrklaff et al (2005) using modern approaches that prove to be particularly powerful.
I enjoyed the way the paper is written describing general features in MVs first and elucidating its components at progressively higher-resolution in subsequent sections.Key findings are that trimers of A10 form the palissade contrary to previous suggestions and do not have strong lateral interactions.This is unlike classical capsids or nucleocapsids observed in other viruses which typically feature a continuous, interconnected lattice.It suggests an unusual mechanism to achieve the brick shape typical of poxvirus cores.
The other components are identified with less confidence but the location of A3 is supported by converging evidence (published mass spectrometry and immunolabelling, and structural features in this study).
The overall model integrating these findings is novel and is likely to represent the working model for poxvirus core until higher resolution is (painstackingly) achieved.

B. Originality and significance (cf. above too)
The imaging presented in this study is a major advance compared to available data (Cyrklaff, 2005) and ingeniously combined with SPA.It changes our view of poxvirus core architecture and suggests mechanisms important for assembly.C-D.Data, methodologoy and stats State of the art approaches are used.Technical comments are added below.No major issues were identified in the validation of the SPA maps and structure.
Maps: provide local resolution maps for SPA.Movie: add the electron density when showing A10 and the side chains involved in domain swapping.
Model building: was one chain refined and then copied to the other positions (l.537)?It would make more sense to apply NCS restraints if that's the case.Six Ramachandran outliers remain.They should be eliminated if possible or commented on otherwise.Fig. S4 should be bigger given there's no space restriction in supplementary material E. Conclusions Overall, despite a limited resolution for most components the conclusions are supported by appropriate experimental evidence.
F. Suggested improvements 1.The fold of A10 should be described in more detail.Does it have any similarities with other viral or cellular proteins (Foldseek, Dali)?Even if they are models, it would also be interesting to discuss these points for A3 (and to a lesser extent L4). 2. The authors are cautious in analysisng molecular details given the limited resolution of the maps.At 3.8Å resolution, the A10 trimer structure is quite reliable.An analysis of residue conservation (as implemented in Consurf for example) would greatly support the interactions described as important for the stability of the trimer (e.g.domain swap and disulfide bond, l. 190-196) or its interaction with A3 (both interacting surfaces are expected to be conserved in addition to their complementary charges; l. 220-24 -Fig.S6).
3. A4 is discussed as a potential matrix protein.Its location, interaction on A10 trimer sides and predicted extended conformation seem more reminiscent of tape-measure or cementing proteins seen in phages (e.g.P30 of PRD1) or adenovirus (rope-like minor capsid proteins).If described as a matrix protein, hydrophobicity should be discussed.
4. Findings are extrapolated to the whole family.It is remarkable that entomopoxviruses should be overlooked here.They are the most diverse members of the family in terms of the ultrastructure and sequence, and would be the strongest support for a broad conservation of the proposed findings.If this paragraph is to stay, the findings should be at the very least extended to parapoxviruses and more divergent members of the family.Given how AF2 works, it is not surprising to see high similarities between proteins that have 95%+ sequence identity (i.e.Fig S10 is not informative).G. References: appropriate.

H. Clarity
Overall, the manuscript is pleasant to read and the figures of high quality.
Introduction: mention EVs (line 43) L.120: The lattice is also reminiscent of the D13 pseudo-hexagonal matrix reconstituted in vitro (Hyun, 2011 and2022) and visualised in vivo (Heuser, 2005) L.127-129: It would be useful to mention here that cores were treated with detergent and DTT so that the reader doesn't have to check the method section Sucrose concentrations: w/v or v/v? 0.25% trypsin: what is this in terms of mg/ml or activity?
Reviewer #2: Remarks to the Author: In this manuscript, Datler et al study the structure of several components of mature vaccinia virus particles, primarily the so called palisade layer.Vaccinia virus is a poxvirus, and this family of viruses have been hard to study structurally due to their massive size (at the limit of what cryo-EM can handle) and the number of proteins that potentially make up the particle structure.The recent outbreaks of Mpox is one reason, amongst many, why it is important to better understand poxvirus architecture.
Datler et al use state-of-the-art methodology to provide important and substantial new insights into poxvirus architecture.The text is overall very well written and the figures are clearly presented and beautiful.In my opinion it would fit excellently in NSMB.I only have relatively minor comments aimed at improving the presentation of the article.
1. Since the inner core wall volume is shown in a figure, it should be deposited at EMDB.Prior to final acceptance the authors should submit a validation report.2. With the preface that I am a structural virologist but not a poxvirologist, I wonder if the authors should also describe their particles in terms of the slightly different nomenclature that exists.In that nomenclature, which I find less ambiguous, the particles that the authors study are referred to as intracellular mature virus (IMV), distinguishing it from e.g. the extracellular enveloped virus (EEV).Perhaps it is at least worth mentioning that the particles are also referred to as intracellular mature virus (IMV)? 3.There should be a space between number and unit, for instance as "3.8 Å". 4. As a general comment, some of the supplementary material is of very high quality and could warrant being moved to main figures, if there is space.5. line 106: The term "high-resolution cryo-ET" is ill defined.To most structural biologists, these tomograms don't count as "high-resolution", so I suggest removing that designator.6. line 124: "average number of ~2280 trimers" Average of how many virions?What is the standard deviation?7. line 140-142: Some statement about the inner core are made here without justifications.At least, a better visual presentation should be made, or a power spectrum shown.Alternatively, the discussion of the inner core architecture could be deferred to the place where the average is presented.8. line 144-145: The statement "Our cryo-ET data suggested the … trimer."should be justified if kept here.Or deferred to the discussion of the average.9. line 182: It could be worth explicitly mentioning what part of the 614 residue A10 is not built into the density.If I understand right, it is the C-terminus with low predicted pLDDT scores?10. line 194 and Fig. S4D: the statement about a possible disulphide should be backed up by showing the density.11. line 209: I find this phrase very unclear: "likely represents a finding with biological significance".The authors are probably trying to say that the findings are in line with each other.The "biological significance" should be removed.12. line 232: As far as I understand the entire central ring is unidentified.There might thus be little need to discuss the "unidentified donut-shaped density" separately (it is also very small and feature-less).13. line 233-235: This statement relies on two assumptions which are not explicitly (quantitatively) shown: (i) that A10 dimers are shed elsewhere (it does seem to be the case), and (ii) that the central ring density without attached A10 trimers would still be particle-picked and correctly averaged.Since the authors don't explicitly show that, they could perhaps make a weaker version of this statement, essentially saying that the trimers are still there but not suggesting that the interaction is "stronger".14.Line 237: Even though the structure predictions make this statement probably, I would still suggest a less definitive subheading since the authors show no experimental data.As structural biologists, it is in our interest to distinguish predictions from experimental measurements.The subheading could simply say something like "is likely conserved".15.The discussion is very long.I think it is ok that way but the readability of the article may be improved if some discussion sections were shortened a bit.Specifically, I am thinking of the subheadings "A potential role for A10…" and "Structural characterization of novel….".16. line 252: This is nit-picking of me, but if you say you are proposing a "revised" model, that would mean that a similarly detailed other model existed, and you are proposing changes to it.I think the authors may instead want to say they are proposing a "detailed" or "more detailed" or "more accurate" model.17. line 286: The authors say "pore hexagon" but have not provided sufficient proof that the central ring density is indeed a hexagon.Clearly, the A10 trimers around it are arranged in hexagonal fashion, but given the extent of symmetry mismatches in this virus and the limited resolution I would suggest removing the word "hexagon".18. line 319: If I understand right, the reference to Table S1 is supposed to support the discussion of the multiple other complexes/symmetries found in the SPA data.But they are not mentioned in Table S1, or? 19.line 407: Says "Aliquots were thawed" should say "Aliquots of X were thawed" (X probably being virus…) 20.line 471: "five rounds of alignment applying no symmetry".It should be made clear what the initial alignment reference was.21. line 480: The expression "were space cleaned (6 pixels)" is understandable only with some interpretation.I suggest to use some more standard phrase such as "overlapping particles were removed using a 6 pixel distance cutoff".22. Fig. S3B: Are the two volumes at the top the same thing from different orientations?If so, please mark it with a rotation symbol.23.Fig. S4: In the legend for panel C, might "inter-subunit" be a better word than "hybrid"?24.Fig. S7C: Is the rotation arrow not in the wrong direction in panel C? 25.Fig. S7D: I realise that this is a low-resolution average and the model fitting is thus very approximate.But the authors should still mention something more about it, for instance if the subunit models also fit in other directions (or not) and possibly show such alternative fits.26.Fig. S9A: The C6 symmetrisation of the central density appears quite speculative at this resolution.I suggest that the authors show C6 and C1 side by side already in panel A. 27.Table S1: Underfocus is typically denoted with negative values.Since the authors do that in the M&M text, I suggest doing it here too.28.Table S2: Perhaps it would be helpful to add a column saying if the protein is a host or virus protein?For some of the proteins, that is not immediately obvious to me.

Author Rebuttal to Initial comments
We thank the reviewers for their evalua2on of our manuscript and have responded to their ques2ons point-by-point below.Please note that during the course of revision Lukas W Bauer has been added to the author list, which has been approved by all other authors.We also now added one addi2onal supplementary figure (Figure S2), which shiGed the numbering of the other supplementary figures accordingly.

Reviewer #1:
Remarks to the Author: A. Summary Poxviruses have a special place in virology owing to their profound impact on human health as both the cause and poten2al solu2on to some of the most dreadful diseases.The study of these viruses led to major discoveries in cell biology.Yet, significant blindspots remain in our understanding of their biology.This manuscript presents major advances in our understanding of the poxvirus architecture and composi2on.It extends the pioneering cryo-ET study by Cyrklaff et al (2005) using modern approaches that prove to be par2cularly powerful.
I enjoyed the way the paper is wriUen describing general features in MVs first and elucida2ng its components at progressively higher-resolu2on in subsequent sec2ons.Key findings are that trimers of A10 form the palissade contrary to previous sugges2ons and do not have strong lateral interac2ons.This is unlike classical capsids or nucleocapsids observed in other viruses which typically feature a con2nuous, interconnected laZce.It suggests an unusual mechanism to achieve the brick shape typical of poxvirus cores.
The other components are iden2fied with less confidence but the loca2on of A3 is supported by converging evidence (published mass spectrometry and immunolabelling, and structural features in this study).
The overall model integra2ng these findings is novel and is likely to represent the working model for poxvirus core un2l higher resolu2on is (painstackingly) achieved.
We thank the reviewer for the posi2ve comments.

B. Originality and significance (cf. above too)
The imaging presented in this study is a major advance compared to available data (Cyrklaff, 2005) and ingeniously combined with SPA.It changes our view of poxvirus core architecture and suggests mechanisms important for assembly.C-D.Data, methodologoy and stats State of the art approaches are used.Technical comments are added below.No major issues were iden2fied in the valida2on of the SPA maps and structure.
Maps: provide local resolu2on maps for SPA.We now provide local resolu2on maps and show them as new panels in Figure S4 and S9.
Movie: add the electron density when showing A10 and the side chains involved in domain swapping.We acknowledge the reviewer's comment.However, we have refrained from adding the electron density to the movie as this made the informa2on difficult to interpret, given that for some of the smaller and nega2vely charged side chains no clear density is visible.This informa2on is already stated in the manuscript: "We note that at ~4 Å resolu0on structural features, such as small and/or nega0vely charged side chains are not clearly visible and hence impose a limit to interpretability." As the corresponding EM density map and model are deposited to the EMDB/PDB, interested readers can inspect the map and model in more detail than what we would be able to convey in the video Model building: was one chain refined and then copied to the other posi2ons (l.537)?It would make more sense to apply NCS restraints if that's the case.Six Ramachandran outliers remain.They should be eliminated if possible or commented on otherwise.
We agree with the reviewer and we have now performed a final itera2on of RoseUa refinement into the cryo-EM density with NCS applied to the model.The backbone RMSD between our new model and the previous one without NCS is 0.873Å for 563 pruned atom pairs, and 1.065Å across all 599 pairs.Molprobity sta2s2cs are improved compared to our previous model, and there are no longer Ramchandran outliers.We thank the reviewer for their sugges2on.We have updated the model sta2s2cs table (Table S3) and the methods sec2on to include the RoseUa refinement with NCS restraints.We have also updated the figures and movies with the new model where necessary.
Fig. S4 should be bigger given there's no space restric2on in supplementary material We have changed the layout of Figure S4 to increase its size.

E. Conclusions
Overall, despite a limited resolu2on for most components the conclusions are supported by appropriate experimental evidence.
F. Suggested improvements 1.The fold of A10 should be described in more detail.Does it have any similari2es with other viral or cellular proteins (Foldseek, Dali)?Even if they are models, it would also be interes2ng to discuss these points for A3 (and to a lesser extent L4).
We thank the reviewer for these sugges2ons.We have now performed a Dali analysis of A10, which revealed no convincing similari2es to any other proteins.We repeated this search using Foldseek against the PDB or AlphaFold-Database, again obtaining the same result.For example, for A10 Foldseek reports only hits with small protein stretches, all with high E-value (where lower is beUer) and low TMscore (where higher would be beUer).
In the case of A3, Foldseek and Dali reveal similari2es in its protein fold to deubiquina2ng enzymes, such as ubiqui2n carboxyl-terminal hydrolases.This is very interes2ng and poten2ally warrants further analysis, but currently goes beyond the scope of this manuscript.Similarly, for L4 and 23K the Dali and Foldseek hits were not providing convincing results with high scores.
We now briefly men2on these results in the main text (changes in bold)."No experimentally-derived structures of the major structural core protein candidates are available.Hence, we used AlphaFold (Jumper et al. 2021) to computa2onally predict models of the main core protein candidates, A10, 23K, A3, A4, and L4 (Figure S1C) to facilitate the interpreta2on of cryo-EM densi2es obtained in our downstream workflow.The Foldseek analysis of A4 is explained below, when answering comment #3 about the role of A4 in the viral core.
2. The authors are cau2ous in analysing molecular details given the limited resolu2on of the maps.At 3.8Å resolu2on, the A10 trimer structure is quite reliable.An analysis of residue conserva2on (as implemented in Consurf for example) would greatly support the interac2ons described as important for the stability of the trimer (e.g.domain swap and disulfide bond, l. 190-196) or its interac2on with A3 (both interac2ng surfaces are expected to be conserved in addi2on to their complementary charges; l. 220-24 -Fig.S6).
The reviewer is correct that given the high sequence conserva2on of the A10 protein within the orthopoxvirus genus (Figure S11A in the revised manuscript), the role of the discussed amino acids in forming interac2ons is somewhat obvious, and hence results in the same AF2 predic2ons.
As suggested by the reviewer, we now have performed ConSurf-analysis of A10, which looks at sequence conserva2on beyond the orthopoxvirus genus.Both cysteine residues (C31/C569) in A10 are highly conserved according to our Consurf analysis.Similarly, the salt bridges described to tether central helices together within the trimer are also showing high conserva2on.Interes2ngly, the residues stretching the region forming the three-stranded beta-sheet show an average to variable conserva2on.However, it is notable that these residues are conserved within orthopoxviruses, but less so among all other poxviruses included in ConSurf analysis (see also our comparison to parapoxviruses and entomopoxviruses below).We have also extended this ConSurf analysis to the other structural core proteins and provide these results in a new Figure S2 and also an updated Figure S5 (previously Figure S4), which displays the conserva2on of residues ploUed back onto the 3D model.
The base of the A10 trimer is more strongly conserved than its top.This includes several of the posi2vely charged residues which have an average to high conserva2on scale according to Consurf.The increased conserva2on of the base of A10 is interes2ng given the similarity of structure predic2ons for this region among orthopox, parapox and entomopoxviruses (see please our answer to comment #4 below).
The side of A3 we predict to be contac2ng the base of the A10 trimer also shows an average level of conserva2on, where several nega2vely charged residues are conserved.This supports our specula2on about how A10 and A3 could be in contact with each other.However, given that based on our structures we are not able to unambiguously define poten2al residue-residue contacts (or also exclude the presence of an addi2onal protein in between the trimer and inner core wall, as discussed in our manuscript could be 23K), we prefer not to speculate in too much detail about conserva2on of a specific interface.2).The residue labels are colored according to the color of the chain they are in (as in panel A).B) salt bridge contacts from panel A with residue numbers indicated.C) Top, a salt bridge and hydrogen bond network at the inter-subunit beta-sheet formed at the core of the trimer.Bottom, hydrophobic packing on the underside of the beta-sheet.D) intramolecular disulfide (cysteine residues indicated).The orientation of the monomer is identical to panel A.

New Figure S2 below:
4. Findings are extrapolated to the whole family.It is remarkable that entomopoxviruses should be overlooked here.They are the most diverse members of the family in terms of the ultrastructure and sequence, and would be the strongest support for a broad conserva2on of the proposed findings.If this paragraph is to stay, the findings should be at the very least extended to parapoxviruses and more divergent members of the family.Given how AF2 works, it is not surprising to see high similari2es between proteins that have 95%+ sequence iden2ty (i.e.Fig S10 is not informa2ve).
Given that comment #2 and #4 deal both with conserva2on of residue and structure, we decided to answer #4 before providing answers to comment #3.
The reviewer is correct that our extrapola2on of the role and structure of the A10 trimer to the en2re poxvirus family does not include poten2al differences in more diverse members.Hence, we have now extended our analysis of the A10 protein structure conserva2on sec2on to also include statements about Entomopoxviruses and Parapoxviruses and extended Figure S11 with this addi2onal informa2on.
The A10 homolog in Orf virus (a member of the Parapoxvirus genus), having only ~40% sequence iden2ty, resulted in an AF-predic2on almost iden2cal to what we observe in orthopoxviruses.However, the fold of the heterodimeric three-stranded beta-sheet is slightly different compared to as seen in orthopoxviruses.Despite this small difference, given the overall iden2cal trimer predic2on for Orf virus, it is s2ll fair to assume that the trimer structure is likely conserved among both ortho-and parapoxviruses.
We have also extended our comparison to Entomopoxviruses.Specifically, we performed a sequence alignment between protein A10 of VACV and the respec2ve homologs in Amsacta moorei entomopoxvirus (AmEPV) and Melanoplus sanguinipes entomopoxvirus (MSEPV).Here the sequence conserva2on was substan2ally lower with ~22%.For these two viruses, AlphaFold predicted an overall different protein structure for the protein A10 homologs.However, a region in these entomopoxvirus proteins adopted an almost iden2cal conforma2on compared to the base of the A10 trimer in orthopoxviruses, Specifically, AmEPV residues 618-850 and MSEPV residues 784-1014 adopt a fold that is structurally remarkably similar with VACV protein A10 residues 370-602.This is intriguing as it suggests that the region in A10 that is posi2oned towards the inner core wall might be a defining structural part among different more distantly related poxvirus species.
To illustrate all these comparisons, we have now extended the table shown in Figure S11A with the sequence conserva2on of Orf virus, AmEPV and MSEPV.We have also included the AF-predic2on of Orf virus into Figure S11 panel B and created a new Figure S11 panel C for the entomopoxvirus predic2ons.Further, we changed the corresponding sec2on 2tle to "The A10 trimer is likely conserved across ortho-and parapoxvirus species" and added there the results obtained for the other poxvirus members.
(changes in bold): "The protein sequence of A10 is highly conserved among the orthopoxvirus genus, including the key residues forming interac2ons in the trimer, with an average ~97% sequence iden2ty between VACV Western reserve (WR), Variola virus, Monkeypox virus, Rabbitpox virus, Cowpox virus and Ectromelia virus (Figure S11A).Correspondingly, AlphaFold predic2ons of these proteins as monomers or trimers are very similar (Figure S11B).More distantly related poxviruses show a substanAally lower sequence idenAty of their A10 protein homologs, with ~40% idenAty in Orf virus (a parapoxvirus) and ~22% in two members of the entomopoxvirus genus, Amsacta moorei entomopoxvirus (AmEPV) and Melanoplus sanguinipes entomopoxvirus (MSEPV) (Figure S11A).AlphaFold predicAons for Orf virus showed a similar A10 trimer structure (Figure S11B).AmEPV and MSEPV protein A10 homologs displayed a different overall fold.However, one region in AeMV and MSEPV A10 homologs adopted a similar conformaAon compared to the base of A10 trimer in orthopoxviruses (Figure S11C).This is intriguing as it suggests that the region in A10 that is posiAoned towards the inner core wall might be a defining structural part among different poxvirus species.Overall, these findings indicate that the interacAons formed within the trimers consAtuAng the palisade layer are similar among members of the orthopox and parapoxvirus genus, but potenAally display interesAng differences to entomopoxviruses." New Figure S11 below:  3) and AF2 predictions of Variola virus A10 and the parapoxvirus Orf virus P4a (A10) residues 1-599.This comparison shows the strong similarity between the protein folds between these virus species.It further shows that the biggest difference between the predicted and refined VACV WR A10 model is at the top of the trimer, facing the outside of the core.The color code displays RMSD (root-mean-square deviation) variations, with lower values equaling a more similar structure.C) Comparison of the refined VACV A10 model to the predicted putative core protein models of MSEPV and AmEPV.Analysis revealed that despite an overall more different fold, MSEPV residues 784-1014 and AmEPV residues 618-850 adopt a highly similar fold compared to VACV A10 residues 370-599.Please note that in the AmEPV AF2-prediction a slightly different angle in a connecting loop between two halves of the fold (annotated with an arrow) leads to a different orientation.
3. A4 is discussed as a poten2al matrix protein.Its loca2on, interac2on on A10 trimer sides and predicted extended conforma2on seem more reminiscent of tape-measure or cemen2ng proteins seen in phages (e.g.P30 of PRD1) or adenovirus (rope-like minor capsid proteins).If described as a matrix protein, hydrophobicity should be discussed.This is an excellent point and indeed the analogy of A4 to cement proteins in Adenoviruses is appealing.Interes2ngly, when using the FoldSeek server, the only hit for a protein matching A4 within the PDB is for Minor/Cement protein IX of human Adenovirus (hAdV) 5, as reported in pdb 6B1T.Specifically, the C-terminal helix of A4 (which is also more highly conserved than the rest of the protein, see new Figure S2) matches the C-terminal helix of protein IX, which is involved in a 4-helix bundle.However, this match has a high E-value (9.29e+0) and low TM-score (0.344), which is subop2mal.There were only two other matches in the AFDB50 for two uncharacterized proteins.One in the barn swallow (Hirundo rus/ca), and the other in the rainbow trout (Oncorhynchus mykiss), both with high E-values and low TM-scores.Hence, while the match to hADV5 protein IX is very interes2ng, it is also s2ll a preliminary analysis.Therefore, we refrain to discuss this in too much detail in our manuscript.Instead, we now briefly men2on in the discussion that the role of A4 as matrix protein would be reminiscent of cement proteins in Adenoviruses, also based on their extended conforma2on.
With our conclusion that A4 could act as a matrix-like protein, we have referred to the report by Cudmore et al. 1996, who characterized the membrane-binding features of A4 (or p39 called in their manuscript) in great detail.Specifically, they reported A4 to be of predominantly hydrophilic nature and not to bind directly to membrane.Instead, they concluded that A4 would be posi2oned to link the core with surrounding membranes by interac2ng with other viral membrane proteins.We acknowledge that our statement referencing this study was not concise enough and hence modified this sec2on to clarify this aspect.

"The extended conforma0on of A4 is reminiscent to minor coat proteins found in Adenoviruses, termed cement proteins due to their role in assembling and maintaining the virus shell (Dai et al. 2017, Gallardo et al. 2021). Hence, A4 could have a similar role compared to these proteins in providing addi0onal stabiliza0on to the palisade layer. A4 has previously been also described as matrix-like protein, which for example could establish a link between the core and surrounding membranes via binding other viral membrane proteins (Cudmore et al. 1996)."
G. References: appropriate.

H. Clarity
Overall, the manuscript is pleasant to read and the figures of high quality.Thank you!Introduc2on: men2on EVs (line 43) We have done this as suggested.
L.120: The laZce is also reminiscent of the D13 pseudo-hexagonal matrix recons2tuted in vitro (Hyun, 2011and 2022) and visualised in vivo (Heuser, 2005) This statement and the respec2ve references have now been added.L.127-129: It would be useful to men2on here that cores were treated with detergent and DTT so that the reader doesn't have to check the method sec2on We have added this informa2on."In order to improve the resolu/on of core structural features, we decided to reduce the complexity of our experimental system and therefore isolated VACV cores via op/mizing established protocols using the detergent NP40 and dithiothreitol (DTT) (Joklik 1962, Esteban 1984, Dubochet et al. 1994).
Sucrose concentra2ons: w/v or v/v? 0.25% trypsin: what is this in terms of mg/ml or ac2vity?We now specify in the methods that the sucrose concentra2ons are w/v.

Reviewer #2:
Remarks to the Author: In this manuscript, Datler et al study the structure of several components of mature vaccinia virus par2cles, primarily the so called palisade layer.Vaccinia virus is a poxvirus, and this family of viruses have been hard to study structurally due to their massive size (at the limit of what cryo-EM can handle) and the number of proteins that poten2ally make up the par2cle structure.The recent outbreaks of Mpox is one reason, amongst many, why it is important to beUer understand poxvirus architecture.
Datler et al use state-of-the-art methodology to provide important and substan2al new insights into poxvirus architecture.The text is overall very well wriUen and the figures are clearly presented and beau2ful.In my opinion it would fit excellently in NSMB.I only have rela2vely minor comments aimed at improving the presenta2on of the ar2cle.
We thank the reviewer for the encouraging comments.
1. Since the inner core wall volume is shown in a figure, it should be deposited at EMDB.Prior to final acceptance the authors should submit a valida2on report.
The volume for the inner core wall has been deposited to the EMDB under deposi2on code EMD-18452.A valida2on report is now provided with the revision.
2. With the preface that I am a structural virologist but not a poxvirologist, I wonder if the authors should also describe their par2cles in terms of the slightly different nomenclature that exists.In that nomenclature, which I find less ambiguous, the par2cles that the authors study are referred to as intracellular mature virus (IMV), dis2nguishing it from e.g. the extracellular enveloped virus (EEV).Perhaps it is at least worth men2oning that the par2cles are also referred to as intracellular mature virus (IMV)?
We have changed this sec2on, also following the recommenda2on of reviewer 1, and now dis2nguish between intracellular mature virus (MV) and extracellular enveloped virus (EV).We agree with reviewer 2 that the usage of the abbrevia2on IMV, instead of just using MV, is less ambiguous.However, we have aimed to be consistent also with the accompanying manuscript by Liu & Corroyer-Dulmot et al., and also with the current nomenclature in the poxvirus field.
3. There should be a space between number and unit, for instance as "3.8 Å".We have inserted a space between a number and the Å symbol, for each instance where it occurs.
4. As a general comment, some of the supplementary material is of very high quality and could warrant being moved to main figures, if there is space.We thank the reviewer for this nice comment.However, due to space reasons we have refrained from moving supplementary figures to main figures.

line 106:
The term "high-resolu2on cryo-ET" is ill defined.To most structural biologists, these tomograms don't count as "high-resolu2on", so I suggest removing that designator.We have removed the term "high-resolu/on" as suggested.
6. line 124: "average number of ~2280 trimers" Average of how many virions?What is the standard devia2on?
The average number was calculated from 15 virions.The standard devia2on is ±309.We now provide these numbers also in the revised manuscript (see below, changes in bold).
"Using the ini/al mesh defined on the surface of the core wall for extrac/ng subtomograms, and the measured size of a trimer within a hexamer of trimer unit, we calculated an average number of ~2,280 trimers (standard devia0on (SD) = ±309, n = 15 virions) to cons/tute the palisade layer, not considering the presence of gaps and cracks."7. line 140-142: Some statement about the inner core are made here without jus2fica2ons.At least, a beUer visual presenta2on should be made, or a power spectrum shown.Alterna2vely, the discussion of the inner core architecture could be deferred to the place where the average is presented.We realized that our statement about the inner core at this posi2on in the text was not clear.Hence, we have changed this sec2on to make it more explicit how we have obtained the measurements of the inner core wall units.Specifically, we believe that in Figure 1C the shape of the inner core wall units is already easily appreciated.Together with the Movie S2, where the organiza2on of the inner core wall is also visible, we are confident that our statements about the inner core wall unit at this posi2on of the manuscript are warranted.
"More importantly, we could for the first /me observe the structural arrangement of the inner core wall (Figure 1C, inner core wall panel, see also Movie S2).Each inner core wall unit adopted a squarelike shape of ~7.4nm x 7.4nm (as measured directly from tomogram slices containing the inner core wall units) and appeared to be following at least 2-fold symmetry, which is consistent with our interpreta/on from intact viruses that the inner core wall is not following the organiza/on of the palisade layer."8. line 144-145: The statement "Our cryo-ET data suggested the … trimer."should be jus2fied if kept here.Or deferred to the discussion of the average.We have removed this sentence.9. line 182: It could be worth explicitly men2oning what part of the 614 residue A10 is not built into the density.If I understand right, it is the C-terminus with low predicted pLDDT scores?
We now state in the methods and Table S3 which residues are built into the EM-density (residues 1-599).
10. line 194 and Fig. S4D: the statement about a possible disulphide should be backed up by showing the density.We acknowledge that we have been not precise in this statement in the ini2al version of the manuscript.We made the conclusion that a disulfide bridge could exist based on the proximity of the two cysteines in our model.However, we did not actually observe a clear density for it, which is also reasonable given that we purified virus cores under reducing condi2ons.We s2ll found it relevant to men2on these cysteines, considering a study that suggested that within assembled MVs, core proteins including A10 are able to form disulfide bonds that are required for maintaining stability of released virus par2cles (Locker andGriffiths 1999, DOI: 10.1083/jcb.144.2.267 ).Also, given that the two cysteines are close to each other in space, but not sequence, they might have a role defining protein conforma2on.We have now changed this part of the manuscript to make this clearer.

"A previous study suggested that VACV core proteins form disulfide bonds within MVs relevant for maintaining stability of released virus par0cles (Locker and Griffiths 1999). When analysing our A10 model, we found two highly conserved cysteine residues (C31 and C569) in close proximity to each other (Figure S5D) which conceivably could clamp the N-terminal and C-terminal parts of A10 monomer together, stabilizing its conforma0on."
We provide informa2on about the conserva2on of cysteines already in our response to reviewer 1. 11. line 209: I find this phrase very unclear: "likely represents a finding with biological significance".The authors are probably trying to say that the findings are in line with each other.The "biological significance" should be removed.We no2ced that the sentence containing this phrase was ambiguous in its meaning and we have therefore removed it.
12. line 232: As far as I understand the en2re central ring is uniden2fied.There might thus be liUle need to discuss the "uniden2fied donut-shaped density" separately (it is also very small and featureless).
We have removed the sentence men2oning the uniden2fied donut-shaped density in the main text.However, we have kept the reference men2oning it in the legend of Figure S10.
13. line 233-235: This statement relies on two assump2ons which are not explicitly (quan2ta2vely) shown: (i) that A10 dimers are shed elsewhere (it does seem to be the case), and (ii) that the central ring density without aUached A10 trimers would s2ll be par2cle-picked and correctly averaged.Since the authors don't explicitly show that, they could perhaps make a weaker version of this statement, essen2ally saying that the trimers are s2ll there but not sugges2ng that the interac2on is "stronger".
The reviewer is correct, we have not quan2ta2vely analysed shedding and how efficiently we can detect the central ring density when no trimers are aUached.Hence, we have now removed the en2re sentence.
14. Line 237: Even though the structure predic2ons make this statement probably, I would s2ll suggest a less defini2ve subheading since the authors show no experimental data.As structural biologists, it is in our interest to dis2nguish predic2ons from experimental measurements.The subheading could simply say something like "is likely conserved".We have now changed the subheading and also extended the rest of the sec2on based on comments of reviewer 1 (please see our answer above).
15.The discussion is very long.I think it is ok that way but the readability of the ar2cle may be improved if some discussion sec2ons were shortened a bit.Specifically, I am thinking of the subheadings "A poten2al role for A10…" and "Structural characteriza2on of novel….".
We have removed one sentence in each of the men2oned sec2ons, given that they were not absolutely necessary.We have also shortened the very first paragraph of the discussion.
16. line 252: This is nit-picking of me, but if you say you are proposing a "revised" model, that would mean that a similarly detailed other model existed, and you are proposing changes to it.I think the authors may instead want to say they are proposing a "detailed" or "more detailed" or "more accurate" model.As suggested by the reviewer, we have changed the phrase "revised model" to "more detailed model" in the discussion and also the abstract.17. line 286: The authors say "pore hexagon" but have not provided sufficient proof that the central ring density is indeed a hexagon.Clearly, the A10 trimers around it are arranged in hexagonal fashion, but given the extent of symmetry mismatches in this virus and the limited resolu2on I would suggest removing the word "hexagon".We have removed the word "hexagon".
18. line 319: If I understand right, the reference to Table S1 is supposed to support the discussion of the mul2ple other complexes/symmetries found in the SPA data.But they are not men2oned in Table S1, or?
We thank the reviewer for no2cing this error and have removed the reference to Table S1 and this  posi2on.19.line 407: Says "Aliquots were thawed" should say "Aliquots of X were thawed" (X probably being virus…) We have changed the statement to now read: "Aliquots of isolated cores were thawed……" 20.line 471: "five rounds of alignment applying no symmetry".It should be made clear what the ini2al alignment reference was.We now explain how the ini2al reference was obtained (see below, changes in bold).
"To generate a de novo reference, subtomograms (cubic size 464 Å³) were then extracted from IsoNetcorrected bin8 tomograms and subjected to five rounds of alignment applying no symmetry.The first ini0al alignment reference was generated by averaging all par0cles, using their non-refined star0ng posi0ons." 21. line 480: The expression "were space cleaned (6 pixels)" is understandable only with some interpreta2on.I suggest to use some more standard phrase such as "overlapping par2cles were removed using a 6 pixel distance cutoff".
We have changed the expression as suggested (see below, changes in bold).
"A^er the first two alignments in bin8, overlapping par0cles were removed using a 6 pixel distance cutoff and……" 22. Fig. S3B: Are the two volumes at the top the same thing from different orienta2ons?If so, please mark it with a rota2on symbol.Yes, the volumes on top are iden2cal, but rotated by 90 degrees.We have added rota2on symbol to illustrate this.23.Fig. S4: In the legend for panel C, might "inter-subunit" be a beUer word than "hybrid"?We have changed the word "hybrid" to "inter-subunit".24.Fig. S7C: Is the rota2on arrow not in the wrong direc2on in panel C?
We thank the reviewer for no2cing this error.We have corrected the orienta2on of the arrow.25.Fig. S7D: I realise that this is a low-resolu2on average and the model fiZng is thus very approximate.But the authors should s2ll men2on something more about it, for instance if the subunit models also fit in other direc2ons (or not) and possibly show such alterna2ve fits.
We acknowledge that the current map is limited in resolu2on and is anisotropic, preven2ng the possibility to unambiguously assign map handedness and orienta2on of an A3 dimer.We now more explicitly acknowledge these limita2ons within the text (see below).Nonetheless, we have aUempted fits of the A3 dimer model into our volume with and without flipping handedness (see figure R1, below), and found that the current fit has the highest correla2on value.
For fiZng the highest ranking Alphafold2 model of A3 dimer into the inner core density we rigid body fit using chimera's "fit in map" tool.Op2ons were set for real-2me correla2on using a simulated map at resolu2on at 20 Å (es2mated resolu2on from cryoSPARC) and op2miza2on by correla2on.We fit the dimer in all possible orienta2ons and, as expected with low resolu2on volumes, correla2on values were similar for all fits although our current fit resulted in the highest value.The current orienta2on also filled the innermost density most completely, while maintaining an apparent two-fold symmetry (as appears in 2D class averages).We provide a figure for the reviewer to allow a beUer apprecia2on of these fits.We have modified the methods to describe how we carried out fiZng, and modified the main text to acknowledge that alterna2ve configura2ons are possible: 27.Table S1: Underfocus is typically denoted with nega2ve values.Since the authors do that in the M&M text, I suggest doing it here too.We have changed Table S1 as suggested.

"Although the low resolu0on and anisotropy of this map prevented determina0on of map handedness and unambiguously assigning orienta0on of A3 within the volume, our model with A3's nega0vely charged patch facing towards A10 sa0sfied the inner density most completely."
28. Table S2: Perhaps it would be helpful to add a column saying if the protein is a host or virus protein?For some of the proteins, that is not immediately obvious to me.This informa2on was already provided in the legend for Table S2: "List of mass spectrometry results from soluble frac/on core sample, filtered for VACV proteins." We have now changed the wording to make this more explicit "List of mass spectrometry results from soluble frac/on core sample, filtered to exclusively show proteins encoded by VACV."

Decision Letter, first revision:
Message: 11th Oct 2023 Dear Dr. Schur, Thank you again for submitting your manuscript "Multi-modal cryo-EM reveals trimers of protein A10 to form the palisade layer in poxvirus cores".We now have comments (below) from the 2 reviewers who evaluated your paper.In light of those reports, we remain interested in your study and would like to see your response to the comments of the referees, in the form of a revised manuscript.
You will see that while reviewer #2 has no further comments, reviewer #1 has remaining concerns, which we would expect to be addressed in a last round of revision.Specifically, please make sure to represent the densities in the figures, where these are discussed in the text.Please also revise the reporting of the ConSurf analysis, as well as description of A10 and A3 folds.
Please be sure to address/respond to all concerns of the referees in full in a point-by-point response and highlight all changes in the revised manuscript text file.If you have comments that are intended for editors only, please include those in a separate cover letter.
We are committed to providing a fair and constructive peer-review process.Do not hesitate to contact us if there are specific requests from the reviewers that you believe are technically impossible or unlikely to yield a meaningful outcome.
We expect to see your revised manuscript within 6 weeks.If you cannot send it within this time, please contact us to discuss an extension; we would still consider your revision, provided that no similar work has been accepted for publication at NSMB or published elsewhere.
As you already know, we put great emphasis on ensuring that the methods and statistics reported in our papers are correct and accurate.As such, if there are any changes that should be reported, please submit an updated version of the Reporting Summary along with your revision.
Please follow the links below to download these files: Reporting Summary: https://www.nature.com/documents/nr-reporting-summary.pdfPlease note that the form is a dynamic 'smart pdf' and must therefore be downloaded and completed in Adobe Reader.
When submitting the revised version of your manuscript, please pay close attention to our href="https://www.nature.com/nature-portfolio/editorial-policies/image-integrity">DigitalImage Integrity Guidelines.</a> and to the following points below: --that unprocessed scans are clearly labelled and match the gels and western blots presented in figures.
--that control panels for gels and western blots are appropriately described as loading on sample processing controls --all images in the paper are checked for duplication of panels and for splicing of gel lanes.
Finally, please ensure that you retain unprocessed data and metadata files after publication, ideally archiving data in perpetuity, as these may be requested during the peer review and production process or after publication if any issues arise.
If there are additional or modified structures presented in the final revision, please submit the corresponding PDB validation reports.
Please note that all key data shown in the main figures as cropped gels or blots should be presented in uncropped form, with molecular weight markers.These data can be aggregated into a single supplementary figure item.While these data can be displayed in a relatively informal style, they must refer back to the relevant figures.These data should be submitted with the final revision, as source data, prior to acceptance, but you may want to start putting it together at this point.SOURCE DATA: we urge authors to provide, in tabular form, the data underlying the graphical representations used in figures.This is to further increase transparency in data reporting, as detailed in this editorial (http://www.nature.com/nsmb/journal/v22/n10/full/nsmb.3110.html).Spreadsheets can be submitted in excel format.Only one (1) file per figure is permitted; thus, for multipaneled figures, the source data for each panel should be clearly labeled in the Excel file; alternately the data can be provided as multiple, clearly labeled sheets in an Excel file.When submitting files, the title field should indicate which figure the source data pertains to.We encourage our authors to provide source data at the revision stage, so that they are part of the peer-review process.
Data availability: this journal strongly supports public availability of data.All data used in accepted papers should be available via a public data repository, or alternatively, as Supplementary Information.If data can only be shared on request, please explain why in your Data Availability Statement, and also in the correspondence with your editor.Please note that for some data types, deposition in a public repository is mandatory -more information on our data deposition policies and available repositories can be found below: https://www.nature.com/nature-research/editorial-policies/reportingstandards#availability-of-data We require deposition of coordinates (and, in the case of crystal structures, structure factors) into the Protein Data Bank with the designation of immediate release upon publication (HPUB).Electron microscopy-derived density maps and coordinate data must be deposited in EMDB and released upon publication.Deposition and immediate release of NMR chemical shift assignments are highly encouraged.Deposition of deep sequencing and microarray data is mandatory, and the datasets must be released prior to or upon publication.To avoid delays in publication, dataset accession numbers must be supplied with the final accepted manuscript and appropriate release dates must be indicated at the motifs if any…).Also, it is impossible to know where the N-and C-termini of A10 are from the main text figures.This is discussed in the text (e.g.line 209) and important given the maturation events preceding the assembly of the palissade.For A3, this could also be a point to mention when the fit is discussed (l.258-260).
2. The electron density map must be shown for each section where side chain residues are discussed and the domain swapping.A close-up of the electron density map is only shown for one region, which is not directly relevant to most of the discussion and insufficient.As it stands, the statements lines 201-206 are not supported by the presented data.I strongly disagree with the suggestion that readers should refer to the deposited map.There is no reason why the figures can't be presented in supplementary material and in the movie.The absence of some side chains is expected at this resolution and doesn't preclude their discussion.However, the reader (and reviewer) should be able to decide by themselves whether the provided information is supportive enough.

Minor points:
As anticipated, the ConSurf results are valuable and support the discussion.However, the main parameters for the analysis are missing.The most important one is which sequences were included (orthopox, parapox, whole family?).
Inclusion of the suggested comparison between VACV A10 to the parapoxvirus and entomopoxvirus A10s adds interesting results.The findings are intriguing for EPVs but, more importantly, the data now supports the hypothesis that the structure of A10 is conserved within the sub-family, which was one of the main conclusions of the manuscript.Please make sure to check the current nomenclature for entomopoxviruses (EV vs. EPV, capitalisation) and complete the statement 22% sequence identity (e.g. over XX residues out of YY).
The refinement of A10 is significantly improved and now satisfactory.
Reviewer #2: Remarks to the Author: The original submitted version of this manuscript was already very good, and in my opinion the authors have now completely addressed all minor questions I had.I would like to congratulate all the authors on this beautifully presented, landmark study.Also, it is impossible to know where the N-and C-termini of A10 are from the main text figures.This is discussed in the text (e.g.line 209) and important given the matura3on events preceding the assembly of the palissade.For A3, this could also be a point to men3on when the fit is discussed (l.258-260).
We have changed Figure 3 to now annotate the N-terminus and C-terminus of A10.We also hope that the new supplementary figure 2 showing the rainbow representa3on for A10 and A3 and their protein topology further helps in their interpreta3on.
For A3, we also now men3on in the main text that in our fit the N-terminus if A3 is facing the core interior.Changes in bold: "Although the low resolu?on and anisotropy of this map prevented determina?on of map handedness and unambiguously assigning orienta?on of A3 within the volume, our model with A3's nega?vely charged patch facing towards A10, and its N-terminus facing the core interior sa?sfied the inner density most completely." 2. The electron density map must be shown for each sec3on where side chain residues are discussed and the domain swapping.A close-up of the electron density map is only shown for one region, which is not directly relevant to most of the discussion and insufficient.As it stands, the statements lines 201-206 are not supported by the presented data.I strongly disagree with the sugges3on that readers should refer to the deposited map.There is no reason why the figures can't be presented in supplementary material and in the movie.The absence of some side chains is expected at this resolu3on and doesn't preclude their discussion.However, the reader (and reviewer) should be able to decide by themselves whether the provided informa3on is suppor3ve enough.
We now provide an updated Movie S4 that displays the densi3es.The densi3es are now also shown in Figure S6B and C (see below), when presen3ng the side chain residues and the domain swapping.We hope that these changes now allow the best possible interpreta3on of our results.Acceptance is conditional on the manuscript's not being published elsewhere and on there being no announcement of this work to the newspapers, magazines, radio or television until the publication date in Nature Structural & Molecular Biology.
Over the next few weeks, your paper will be copyedited to ensure that it conforms to Nature Structural & Molecular Biology style.Once your paper is typeset, you will receive an email with a link to choose the appropriate publishing options for your paper and our Author Services team will be in touch regarding any additional information that may be required.
After the grant of rights is completed, you will receive a link to your electronic proof via email with a request to make any corrections within 48 hours.If, when you receive your proof, you cannot meet this deadline, please inform us at rjsproduction@springernature.com immediately.
You will not receive your proofs until the publishing agreement has been received through our system.
Due to the importance of these deadlines, we ask that you please let us know now whether you will be difficult to contact over the next month.If this is the case, we ask you provide us with the contact information (email, phone and fax) of someone who will be able to check the proofs on your behalf, and who will be available to address any last-minute problems.
To assist our authors in disseminating their research to the broader community, our SharedIt initiative provides all co-authors with the ability to generate a unique shareable link that will allow anyone (with or without a subscription) to read the published article.
Recipients of the link with a subscription will also be able to download and print the PDF.
As soon as your article is published, you can generate your shareable link by entering the DOI of your article here: <a href="http://authors.springernature.com/share">http://authors.springernature.com/share<a>.Corresponding authors will also receive an automated email with the shareable link Note the policy of the journal on data deposition: http://www.nature.com/authors/policies/availability.html.
Your paper will be published online soon after we receive proof corrections and will appear in print in the next available issue.You can find out your date of online publication by contacting the production team shortly after sending your proof corrections.Content is published online weekly on Mondays and Thursdays, and the embargo is set at 16:00 policies">self-archiving policies</a>.Those licensing terms will supersede any other terms that the author or any third party may assert apply to any version of the manuscript.
In approximately 10 business days you will receive an email with a link to choose the appropriate publishing options for your paper and our Author Services team will be in touch regarding any additional information that may be required.
You will not receive your proofs until the publishing agreement has been received through our system. If Foldseek (van Kempen et al. 2023) and Dali (Holm 2022) analysis revealed no similariAes in protein fold of A10 to other cellular or viral proteins in the Protein Data Bank (PDB) or AlphaFold-databases. InteresAngly, A3 showed highest similarity to deubiquinaAng proteins.For the other proteins Foldseek and Dali hits had low probability or TM scores, further highlighAng the structural dissimilarity of these proteins to cellular proteins or other viral proteins outside of the poxvirus family."

Figure S2 :
Figure S2: Sequence conservation analysis of structural core proteinsAlphaFold-predicted models of A10, A3, 23K, A4 and L4, shown in the same orientation as in FigureS1Cand colored according to their ConSurf score 29 , highlighting the predicted conservation of individual residues.The modeled residues are indicated in brackets.

Figure S5 :
Figure S5: Inter-and intramolecular interactions of the A10 trimerA) Diagrammatic illustration of key inter-chain contacts at the oligomerization interface between two monomers (labeled I and II).The diagram shows monomer II pulling away from the trimer to reveal underlying contacts, shown in yellow (hydrophobic), orange (H-bond), and red (salt bridges).B-D) Details of the inter-and intramolecular interactions.Residues are colored according to their conservation as determined via ConSurf (see also Figure2).The residue labels are colored according to the color of the chain they are in (as in panel A).B) salt bridge contacts from panel A with residue numbers indicated.C) Top, a salt bridge and hydrogen bond network at the inter-subunit beta-sheet formed at the core of the trimer.Bottom, hydrophobic packing on the underside of the beta-sheet.D) intramolecular disulfide (cysteine residues indicated).The orientation of the monomer is identical to panel A.

Figure S11 :
Figure S11: Comparison of Vaccinia virus Western Reserve A10 trimers to other members of the poxvirus family A) Sequence identity of A10 protein of different viruses of the poxvirus family in comparison to VACV WR.B) Comparison of the initial VACV WR A10 AlphaFold (AF2) prediction to the refined VACV WR A10 structure (also shown in Figure 3) and AF2 predictions of Variola virus A10 and the parapoxvirus Orf virus P4a (A10) residues 1-599.This comparison shows the strong similarity between the protein folds between these virus species.It further shows that the biggest difference between the predicted and refined VACV WR A10 model is at the top of the trimer, facing the outside of the core.The color code displays RMSD (root-mean-square deviation) variations, with lower values equaling a more similar structure.C) Comparison of the refined VACV A10 model to the predicted putative core protein models

Figure R1 .
Figure R1.Possible fits of alphafold A3 dimer model into density for inner core wall volume with and without map handedness flipped (figure not included in the revised manuscript).

Figure S2 :
Figure S2: Protein topology of A10 and A3 Illustration of protein topology and arrangement for A10 (panel A-B) and A3 (panel C-D).The coloring of the models and topology diagrams is according to residue positioning from the N-terminus (blue) to the C-terminus (red).A) Cartoon ribbon representation of the Alphafold-predicted model of A10, colored from N-terminus to C-terminus.The two sub-domains (SD1 and SD2) of the A10 fold are annotated.B) Protein topology diagram of A10, showing the positioning of protein regions, such as specific secondary structures with respect to each other.Beta-strands and alpha-helices are shown as arrows and cylinders, respectively.Their length is proportional to their number of residues.Colored boxes represent motifs which are positioned nearby relative to one another and therefore are likely to form a structural group.C) Cartoon ribbon representation of the Alphafold-predicted A3 colored from N-terminus to C-terminus.D) Protein topology diagram of A3, in the same depiction style as in (B).The topology diagrams shown in (B) and (D) were generated using Pro-Origami 66 .

Figure S6 :Final
Figure S6: Inter-and intramolecular interactions of the A10 trimerA) Diagrammatic illustration of key inter-chain contacts at the oligomerization interface between two monomers (labeled I and II).The diagram shows monomer II pulling away from the trimer to reveal underlying contacts, shown in yellow (hydrophobic), orange (H-bond), and red (salt bridges).B-D) Details of the inter-and intramolecular interactions.Residues are colored according to their conservation as determined via ConSurf (see also FigureS2).The residue labels are colored according to the color of the chain they are in (as in panel A).B) Salt bridge contacts from panel A with residue numbers indicated.C) Top, a salt bridge and hydrogen bond network at the inter-subunit beta-sheet formed at the core of the trimer.Bottom, hydrophobic packing on the underside of the beta-sheet.Electron microscopy density is shown for all images in B-C.D) Potentially interacting cysteines (residues indicated).The orientation of the monomer is identical to panel A.
you have any questions about our publishing options, costs, Open Access requirements, or our legal forms, please contact ASJournals@springernature.com