The effect of site-specific recombinases XerCD on the removal of over-replicated chromosomal DNA through outer membrane vesicles in bacteria

ABSTRACT Outer membrane vesicles (OMVs) are universally produced by Gram-negative bacteria and play important roles in symbiotic and pathogenic interactions. The DNA from the lumen of OMVs from the Alphaproteobacterium Dinoroseobacter shibae was previously shown to be enriched for the region around the terminus of replication ter and specifically for the recognition sequence dif of the two site-specific recombinases XerCD. These enzymes are highly conserved in bacteria and play an important role in the last phase of cell division. Here, we show that a similar enrichment of ter and dif is found in the DNA inside OMVs from Prochlorococcus marinus, Pseudomonas aeruginosa, Vibrio cholerae, and Escherichia coli. The deletion of xerC or xerD in E. coli reduced the enrichment peak directly at the dif sequence, while the enriched DNA region around ter became broader, demonstrating that either enzyme influences the DNA content inside the lumen of OMVs. We propose that the intra-vesicle DNA originated from over-replication repair and the XerCD enzymes might play a role in this process, providing them with a new function in addition to resolving chromosome dimers. IMPORTANCE Imprecise termination of replication can lead to over-replicated parts of bacterial chromosomes that have to be excised and removed from the dividing cell. The underlying mechanism is poorly understood. Our data show that outer membrane vesicles (OMVs) from diverse Gram-negative bacteria are enriched for DNA around the terminus of replication ter and the site-specific XerCD recombinases influence this enrichment. Clearing the divisome from over-replicated parts of the bacterial chromosome might be a so far unrecognized and conserved function of OMVs.

Response to the raised concerns: 1) The article proposes a conserved mechanism based on similar results from a previously analyzed bacterial species to the one shown in this article (E. coli).this suggestion can be further supported, given that the methodology is in place.I suggest performing a similar analysis in at least two more bacteria of different lineages.As it stands the work is too preliminary for sending out to review.
We agree that for the proposal of a general mechanism prove from more bacteria is needed.As we cannot perform the experiments by ourselves anymore we searched the literature for other sequencing studies of DNA from OMVs.We identified three studies.For Cyanobacterium Prochlorococcus marinus and both chromosomes of Gammaproteobacterium Vibrio cholerae, we could clearly identify peaks at the dif sites as we found for D. shibae and E. coli.The OMVs isolated from Gammaproteobacterium Pseudomonas aeruginosa were predominantly formed during explosive cell lysis.However, we also could identify a peak at dif site that might have originated from overreplication repair.Thus, studies from other organisms support our hypothesis.
2) A phenotypic characterization of a mutant is not enough evidence to support a link between a gene and a phenotype.Complementation analysis is required to infer any gene-to-function link.
We fully agree that complemented strains would be needed for a full support of the proposed link.We hadn't finished the strains yet when the lab closed.However, both independent xerC and xerD mutants showing the same phenotype, provides already limited evidence for the proposed hypothesis.We made it now clear that these data are preliminary and have to be treated with caution.
3) Please ensure that the methodology is described in the manuscript.The statement "Details on all methods, including DNA isolation, can be found in [4] is not acceptable.
We submitted the study as an observation with very limited space.We felt that the required shortening of the methods section would provide no benefit to the reader over the methods section of our first OMV paper that we cite here and that is very detailed.We are convinced It will be a reliable guide to other researchers.
3) The coverage analysis shown in Fig. 2 is different from the one referenced in (4) and is not described in the manuscript.Log2 coverage, fold change etc. Please ensure that the methodology for data analysis and the statistical significance of the fold changes are reported.
We are sorry for the missing information that we added now to the methods section.
The title does not represent the data covered in the research.The release of DNA through OMVs is not the topic of the experiments.The effect of XerCD on the removal of over-replicated chromosomal DNA is the main focus of the experiments presented.
Lines 24-26, "Our data suggest a conserved mechanism for repair and removal of over-replicated DNA through outer membrane vesicles and an active role of the site-specific XerCD recombinase complex therein."and Lines 134-135: "The site-specific recombinases XerC and XerD might play an active role for enrichment of their recognition sequence dif in the lumen of OMVs."My interpretation of these sentences is that XerCD is actively involved in the translocation of DNA to the lumen of OMVs.The data presented here do not cover the removal of over-replicated DNA from the cell through OMVs.The issue of how the excised DNA ends up in the OMVs is not studied here.The OMVs are used as a source of the DNA and assumed to contain a representative sample of the excised DNA.Maybe the authors can rephrase these sentences and be more specific on the role of XerCD.
Lines 57-59: "Are the XerCD enzymes directly involved in the excision of over-replicated DNA fragments around ter, implicating that they might have a second function in addition to resolving chromosome dimers?" The experiments presented in this manuscript do not address whether the XerCD enzymes are directly or indirectly involved in this process.Indirect effects can not be excluded and I would therefore suggest to remove the word "directly" and rephrase the second half of the sentence to reflect the fact that the observed effects might be an indirect result of the absence of one of the enzymes.
Lines 59-60: "Therefore, we investigated the DNA composition in the lumen of OMVs produced by deletion mutants of XerC and XerD in E. coli."Deletion mutants will not answer this question, they will just answer if the enzymes (or actually the genes) are essential for this process.
Since it is the genes that are deleted I would refer to xerC and xerD rather than XerC and XerD.
Lines 103 -104: "... then the mutants apparently had more waste to get rid of."It is intriguing that at time 0 the number of vesicles per cell are the same (~10 vesicles per cell) for the three strains, indicating that during preculturing conditions the vesicle formation was likely similar.But during the 12 h cultivation the wild-type cells hardly produce vesicles while the xerC and xerD mutant strains continue vesicle production.This means that the vesicles after 12 h cultivation were produced during a different stage of culturing.For the wild-type a major part of the vesicles were produced before the 12 h cultivation, while >95% (rough calculation, going from 2x10^8 to 5x10^9 vesicles/mL) of the vesicles obtained from the mutant strains were produced during the 12 h cultivation.
To be able to claim that the mutants had more waste to get rid of you would need to show that the newly produced vesicles contain DNA.Another question is whether the vesicles are produced as a means to get rid of the excess DNA (active process, as is suggested by the phrase "had more waste to get rid of") or if the DNA just happens to be incorporated in the vesicles (passive process).
Maybe the authors can rephrase this sentence.
Lines 116-117: "These data suggest impaired binding or functioning of the XerCD recombinase complex at the dif site when one of the homologs is knocked out." In the deletion mutants there is no XerCD recombinase complex.
Please rephrase this sentence.
Suggestions regarding potential typos: Line 80: Supplementary Line 91: remainder Line 134: is Reviewer #2 (Comments for the Author): Overall comments: The authors' condensed resubmitted article describes reanalyzed and newly collected DNA sequencing data from bacterial OMV isolations.Some proteobacterial species they have explored in this and prior studies was examined and appears to show that some OMVs are enriched with chromosomal DNA regions (dif and ter) specifically over-generated during cell division.The study is provocative and likely impactful, but the article's current very condensed/ shortened format makes the findings and methods briefly described a difficult and at times confusing to follow in places.Many questions related to how methods were compared (eg.were vesicle particles/ml standardized for DNA extraction) and other questions about the aim of the study are posed in the comments below to help the authors enhance a reader's understanding of the content (as they didn't have the benefit of the longer marked up manuscript version).The article would also benefit from more citations to prove arguments/ statements made, as well as additional proofreading and grammar editing; many grammar editing programs will catch many of these sentence structure/grammar issues that weren't noted in comments but present in the manuscript.

Specific comments:
ABSTRACT: Line 12.This statement about OMVs being produced by all prokaryotes is incorrect as written; not all prokaryotes have outer membranes.Please correct this statement (as on lines 28-29) or omit it.A similar oversimplification of prokaryotes occurs in the author's conclusions statement made on lines 135-136 as well.
Line 18-20.This statement regarding E.coli deletion mutants lacks a clear interpretation or explanation of outcomes.Please consider rephrasing findings here to interpret the outcomes of results for the reader as few read beyond the abstract.

IMPORTANCE:
Lines 22-26.The authors do not prove or show evidence of how ter is enriched in the species OMVs they DNA sequenced.This section should be carefully revised to highlight why dif and ter DNA enrichment in OMVs is important in Gram-negative species OMV formation.Explain to a reader why this is potentially important to study/ understand.

INTRODUCTION:
Line 30 "...vesicles of Gram negative bacteria (OMV) have often been found to contain DNA,..." Can the authors clarify the following in this statement: i) What articles are being referred to here, because reference 3 does not sufficiently address this content.ii) Is the DNA referred to here actually confirmed to be located within the OMV or was it just carried over with the isolated vesicles (artifactual)?How common is DNA containment in vesicles?As this is a major rationale for the study these could be more clearly justified/ explained in this paragraph.
Line 34: "Based on multiple evidence..." Please cite this evidence here (either reference papers and summarize main arguments) as it was not clear from the first statement about prior alphaproteobacterial work.
Line 35 "...vesicles, which were ejected from the dividing cell´s division plane,.." Please discuss this evidence in more detail as it is again a major argument poorly described in the hypothesis but not well argued in the introduction.
Line 45: "The DNA enriched in OMV might therefore originate..." Grammar issues related to "therefore usage and placement.Also the authors should briefly describe what is known about the DNA enriched in OMVs?Are DNA verified to be inside the vesicles (see carryover question above)?DNA may be attracted to the surface of the vesicles drawn by Mg2+ and Ca2+ that associate with outer layer OMV lipopolysaccharides.
Lines 46-50.The aim and arguments posed to rationalize the study, were a bit confusing.Do the authors seek to clarify why over-replication by XerCD occurs?It was not clear how monitoring OMV DNA content is relates and was a logic gap based on the introductory content provided.What are the authors trying to prove, unequal XerCD replication or the fact that these overreplicated DNA products are dealt with by cells through their removal in OMVs during cell division?The study does not really explain how that happens as alluded to in the "Importance section".Were mid-log cells selected for this study versus the more common stationary phase cells to enrich cell replication-derived OMVs versus stationary; this was not clear in the methods?
Line 49-50.More effort needs to be made before this statement and in the methods to explain/ prove how DNA was extracted from within OMVs for sequencing (and not also contaminating carryover with vesicles) or from other descriptions of explosive cell lysis in lines 81-93.Were all OMVs DNA sequenced in the study collected at the same stages of culturing (mid or late log, stationary phase?).Did growth have an impact?Additionally, tangential flow filtration (tff) used by the authors to isolate OMVs could just as easily increase and concentrate released DNA along with accumulated OMVs.The authors need clarify and prove this DNA is inside or attached to the vesicle based on these methods, as this is major part of their hypothesis and main argument at many points in the paper.The hypothesis may still hold even if the DNA is on the surface of vesicles and not inside, so some effort should be made on this topic.
Line 59-60.Please explain how luminal DNA only was isolated from OMVs here.After reviewing the fluorescent staining images from cited study, the vesicles shown with FM1-43 lipid stained and DAPI stained OMVs did not always show that DAPI-stained DNA was solely located within OMV lumen (Fig 1G -H) of the cited reference #4.How much DAPI signal was in the background of the image and was the image background corrected?Was this cited study (reference 4) the only evidence of luminal DNA in OMVs to prove the author's arguments?
Line 85. "It is therefore expected that it might contain also DNA other than dif."Hard to follow statement; many grammar issues hindering meaning.This type of situation occurs often in this short article.Please carefully edit the article for grammar.
Line 81 (Paragraph).This whole section is missing DNA GenBank accession numbers and reference citations for the DNA described and used in the study.The authors should make more effort to properly cite prior collected data used in the study.Figure 1 and 2 of the author's study should provide an added chart or quantification of the total sequenced DNA reads, where the % of ter and dif identified in coverage is summarized with respect to other possible identified or unknown DNA sequences recovered from the OMVs.How much of the OMV DNA recovered (% of total) is actually ter or dif DNA as opposed to other random DNA fragments?
Line 94-104.Given that XerCD mutants produced very different amounts of vesicles per ml amounts, did DNA extractions from all OMV preps account for the differences in vesicles per ml?Please return the manuscript within 60 days; if you cannot complete the modification within this time period, please contact me.If you do not wish to modify the manuscript and prefer to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by Microbiology Spectrum.
If your manuscript is accepted for publication, you will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail.Arrangements for payment must be made before your article is published.For a complete list of Publication Fees, including supplemental material costs, please visit our website.
Corresponding authors may join or renew ASM membership to obtain discounts on publication fees.Need to upgrade your membership level?Please contact Customer Service at Service@asmusa.org.
Thank you for submitting your paper to Microbiology Spectrum.
While we are willing to consider a revised version of this paper at Spectrum, it would be in your best interest to improve the writing.I recommend that you ask a colleague of yours who is a native English speaker to read and provide you some feedback on the writing.You are also welcome to use one of the services here: https://journals.asm.org/content/language-editing-services Line 1: The title does not represent the data covered in the research.The release of DNA through OMVs is not the topic of the experiments.The effect of XerCD on the removal of over-replicated chromosomal DNA is the main focus of the experiments presented.
Lines 24-26: "Our data suggest a conserved mechanism for repair and removal of over-replicated DNA through outer membrane vesicles and an active role of the site-specific XerCD recombinase complex therein."and Lines 134-135: "The site-specific recombinases XerC and XerD might play an active role for enrichment of their recognition sequence dif in the lumen of OMVs."My interpretation of these sentences is that XerCD is actively involved in the translocation of DNA to the lumen of OMVs.The data presented here do not cover the removal of over-replicated DNA from the cell through OMVs.The issue of how the excised DNA ends up in the OMVs is not studied here.The OMVs are used as a source of the DNA and assumed to contain a representative sample of the excised DNA.Maybe the authors can rephrase these sentences and be more specific on the role of XerCD.
Lines 57-59: "Are the XerCD enzymes directly involved in the excision of over-replicated DNA fragments around ter, implicating that they might have a second function in addition to resolving chromosome dimers?" The experiments presented in this manuscript do not address whether the XerCD enzymes are directly or indirectly involved in this process.Indirect effects can not be excluded and I would therefore suggest to remove the word "directly" and rephrase the second half of the sentence to reflect the fact that the observed effects might be an indirect result of the absence of one of the enzymes.
Lines 59-60: "Therefore, we investigated the DNA composition in the lumen of OMVs produced by deletion mutants of XerC and XerD in E. coli."Deletion mutants will not answer this question, they will just answer if the enzymes (or actually the genes) are essential for this process.
Since it is the genes that are deleted I would refer to xerC and xerD rather than XerC and XerD.
Lines 103 -104: "… then the mutants apparently had more waste to get rid of."It is intriguing that at time 0 the number of vesicles per cell are the same (~10 vesicles per cell) for the three strains, indicating that during preculturing conditions the vesicle formation was likely similar.But during the 12 h cultivation the wild-type cells hardly produce vesicles while the xerC and xerD mutant strains continue vesicle production.This means that the vesicles after 12 h cultivation were produced during a different stage of culturing.For the wild-type a major part of the vesicles were produced before the 12 h cultivation, while >95% (rough calculation, going from 2x10^8 to 5x10^9 vesicles/mL) of the vesicles obtained from the mutant strains were produced during the 12 h cultivation.
To be able to claim that the mutants had more waste to get rid of you would need to show that the newly produced vesicles contain DNA.
Another question is whether the vesicles are produced as a means to get rid of the excess DNA (active process, as is suggested by the phrase "had more waste to get rid of") or if the DNA just happens to be incorporated in the vesicles (passive process).Maybe the authors can rephrase this sentence.
Lines 116-117: "These data suggest impaired binding or functioning of the XerCD recombinase complex at the dif site when one of the homologs is knocked out." In the deletion mutants there is no XerCD recombinase complex.
Please rephrase this sentence.

Reviewer comments:
Reviewer #1 (Comments for the Author): Line 1: The title does not represent the data covered in the research.The release of DNA through OMVs is not the topic of the experiments.The effect of XerCD on the removal of over-replicated chromosomal DNA is the main focus of the experiments presented.
Answer: Title was changed as suggested.
Lines 24-26, "Our data suggest a conserved mechanism for repair and removal of over-replicated DNA through outer membrane vesicles and an active role of the site-specific XerCD recombinase complex therein."and Lines 134-135: "The site-specific recombinases XerC and XerD might play an active role for enrichment of their recognition sequence dif in the lumen of OMVs."My interpretation of these sentences is that XerCD is actively involved in the translocation of DNA to the lumen of OMVs.The data presented here do not cover the removal of over-replicated DNA from the cell through OMVs.The issue of how the excised DNA ends up in the OMVs is not studied here.The OMVs are used as a source of the DNA and assumed to contain a representative sample of the excised DNA.Maybe the authors can rephrase these sentences and be more specific on the role of XerCD.
Answer: Thank you for your comments.We have rephrased the "Importance" paragraph (formerly lines 24-26) in the following way: "Our data show that OMVs from diverse Proteobacteria are enriched for over-replicated sequences around the terminus of replication ter and the site-specific XerCD recombinases influence this enrichment.Clearing the divisome from over-replicated parts of the bacterial chromosome might be a novel function of OMVs." Regarding lines 134-135: Of course we did not want to imply that the XerCD recombinases are actively involved in the translocation of DNA into the lumen of OMVs.We showed, however, that the enrichment of ter and dif changes when either xerC or xerD is deleted, meaning the activity of these enzymes is essential for the enrichment.We have rephrased the statement in the following way: "To conclude, we show that the enrichment of the ter region of the bacterial chromosome in OMVs is not restricted to D. shibae.The site-specific recombinases XerC and XerD are essential for enrichment of their recognition sequence dif in the lumen of OMVs.Given their almost universal presence in Bacteria [36] and the strong conservation of the cell division molecular machinery it would be interesting to unravel the underlying mechanisms in more detail." Lines 57-59: "Are the XerCD enzymes directly involved in the excision of over-replicated DNA fragments around ter, implicating that they might have a second function in addition to resolving chromosome dimers?" The experiments presented in this manuscript do not address whether the XerCD enzymes are directly or indirectly involved in this process.Indirect effects can not be excluded and I would therefore suggest to remove the word "directly" and rephrase the second half of the sentence to reflect the fact that the observed effects might be an indirect result of the absence of one of the enzymes.

Answer:
We have carefully considered your comments above and elsewhere and rephrased the sentence in the following way: "Deletion of xerC or xerD in E. coli reduces the enrichment peak directly at the dif sequence, while the enriched DNA region around ter becomes broader, demonstrating that either enzyme influences the contents of DNA inside the lumen of OMVs.We hypothesize that the intra-vesicle DNA is the result of over-replication repair and the XerCD enzymes contribute to this, providing them with a new function in addition to resolving chromosome dimers." Lines 59-60: "Therefore, we investigated the DNA composition in the lumen of OMVs produced by deletion mutants of XerC and XerD in E. coli."Deletion mutants will not answer this question, they will just answer if the enzymes (or actually the genes) are essential for this process.
Since it is the genes that are deleted I would refer to xerC and xerD rather than XerC and XerD.
Answer: Was rephrased as suggested.Sentence now reads: "(2) Are the XerCD enzymes essential for the enrichment of ter and dif in the DNA inside OMVs?When these enzymes are resolving chromosome dimers, no fragments containing dif are produced.Therefore, we investigated the DNA composition in the lumen of OMVs produced by deletion mutants of xerC and xerD in E. coli." Line 74: This is a reference to a reference.Please refer to Genome Biol Evol 10:359-369.https://doi.org/10.1093/gbe/evy005directly.
Answer: Thank you, we added the correct reference.In addition, we also uploaded the used scripts to github.The link is provided in the material and methods section.
Lines 103 -104: "... then the mutants apparently had more waste to get rid of."It is intriguing that at time 0 the number of vesicles per cell are the same (~10 vesicles per cell) for the three strains, indicating that during preculturing conditions the vesicle formation was likely similar.But during the 12 h cultivation the wild-type cells hardly produce vesicles while the xerC and xerD mutant strains continue vesicle production.This means that the vesicles after 12 h cultivation were produced during a different stage of culturing.For the wild-type a major part of the vesicles were produced before the 12 h cultivation, while >95% (rough calculation, going from 2x10^8 to 5x10^9 vesicles/mL) of the vesicles obtained from the mutant strains were produced during the 12 h cultivation.
To be able to claim that the mutants had more waste to get rid of you would need to show that the newly produced vesicles contain DNA.Another question is whether the vesicles are produced as a means to get rid of the excess DNA (active process, as is suggested by the phrase "had more waste to get rid of") or if the DNA just happens to be incorporated in the vesicles (passive process).Maybe the authors can rephrase this sentence.
Answer: Your careful observations are really interesting.However, we do not know the half-life of the produced vesicles.The samples from 12 h might contain none or all or a certain fraction of the vesicles produced at 1 h.What we do see, though, is that the number of vesicles per cell is similar between wt and mutants during the first 2 h of cultivation.Then, at 4 h, it declines in the wt and remains high in the mutants.We think it is possible to conclude that the mutant cultures produced more vesicles per cell, particularly during stationary phase.
Regarding the passive or active incorporation of the DNA into the vesicles: Since we do not know the mechanism, we rephrased the sentence accordingly.Paragraph now reads: "The ratio of vesicles per cell was similar for wild-type and mutants during the first 2 h of growth.Then, at 4 h, it dropped to 0.6 -0.2 for the wild-type while it remained between 3 and 10 for both mutants (Supplementary Figure ).If our hypothesis is true and the DNA in OMVs represents overreplicated fragments, then more such waste was produced in the mutant OMVs." Lines 116-117: "These data suggest impaired binding or functioning of the XerCD recombinase complex at the dif site when one of the homologs is knocked out." In the deletion mutants there is no XerCD recombinase complex.Please rephrase this sentence.
Answer: Right.Thank you.Sentence now reads: "Since the XerCD-FtsK-complex cannot be formed when either xerC or xerD are knocked out, these data reflect the activity of the remaining recombinase homolog." Suggestions regarding potential typos: Line 80: Supplementary Line 91: remainder Line 134: is Answer: Thank you, was corrected or reformulated.
Reviewer #2 (Comments for the Author): Overall comments: The authors' condensed resubmitted article describes reanalyzed and newly collected DNA sequencing data from bacterial OMV isolations.Some proteobacterial species they have explored in this and prior studies was examined and appears to show that some OMVs are enriched with chromosomal DNA regions (dif and ter) specifically over-generated during cell division.The study is provocative and likely impactful, but the article's current very condensed/ shortened format makes the findings and methods briefly described a difficult and at times confusing to follow in places.Many questions related to how methods were compared (eg.were vesicle particles/ml standardized for DNA extraction) and other questions about the aim of the study are posed in the comments below to help the authors enhance a reader's understanding of the content (as they didn't have the benefit of the longer marked up manuscript version).The article would also benefit from more citations to prove arguments/ statements made, as well as additional proofreading and grammar editing; many grammar editing programs will catch many of these sentence structure/grammar issues that weren't noted in comments but present in the manuscript.
Answer: Thank you!We appreciate the valuable criticism.

Specific comments:
ABSTRACT: Line 12.This statement about OMVs being produced by all prokaryotes is incorrect as written; not all prokaryotes have outer membranes.Please correct this statement (as on lines 28-29) or omit it.A similar oversimplification of prokaryotes occurs in the author's conclusions statement made on lines 135-136 as well.
Answer: Of course, sorry for the mistake.We replaced "prokaryotes" by "Gram negative Bacteria" as only those have outer membranes.In the Introduction we generalized that membrane vesicles are produced by cells from all domains of life.
Line 18-20.This statement regarding E.coli deletion mutants lacks a clear interpretation or explanation of outcomes.Please consider rephrasing findings here to interpret the outcomes of results for the reader as few read beyond the abstract.
this study versus the more common stationary phase cells to enrich cell replication-derived OMVs versus stationary; this was not clear in the methods?
Answer: Those are a lot of questions!The rationale of the introduction is the following: (1) DNA has routinely been found inside OMVs.Many references are now provided.(Line 35 -54).
(2) Given the mechanism of OMV biogenesis by blebbing from the outer membrane, it is unclear how DNA can be incorporated into OMVs.(Line 55 -64).
Then, we continue to explain which part of this hypothesis is explored in the current paper.
Regarding your suggestion to compare the DNA content of exponential to stationary phase OMVs: It is certainly a good idea.Here (and in our previous work) we harvested OMVs at the end of the exponential growth phase.
Line 49-50.More effort needs to be made before this statement and in the methods to explain/ prove how DNA was extracted from within OMVs for sequencing (and not also contaminating carryover with vesicles) or from other descriptions of explosive cell lysis in lines 81-93.Were all OMVs DNA sequenced in the study collected at the same stages of culturing (mid or late log, stationary phase?).Did growth have an impact?Additionally, tangential flow filtration (tff) used by the authors to isolate OMVs could just as easily increase and concentrate released DNA along with accumulated OMVs.The authors need clarify and prove this DNA is inside or attached to the vesicle based on these methods, as this is major part of their hypothesis and main argument at many points in the paper.The hypothesis may still hold even if the DNA is on the surface of vesicles and not inside, so some effort should be made on this topic.
Answer: Thank you for this comment.We dedicate now one section to the discussion of the vesicle DNA content: "For P. marinus, D. shibae, V. cholerae and the newly analyzed E. coli strains, the vesicles were treated by DNase prior to analyzing the DNA inside the vesicle lumen.However, the effectiveness of DNAse treatment plays a large role in the enrichment of protected DNA and a complete removal of extra-vesicular DNA cannot be guaranteed [39].For D. shibae we previously sequenced DNA from both, DNase treated and untreated vesicle enrichments and could show that digestion of unprotected DNA results in a reduction of reads mapping outside the ter region.In case of V. cholerae, the sampling time point was chosen to minimize DNA originating from lysed cells and two consecutive digestion steps were performed.Besides enrichment of the ter region, some other short specific regions and in particular phage DNA was found to be overrepresented.DNA within a phage is also shielded from DNase activity.The Membrane vesicles from P. aeruginosa biofilms were not treated by DNase prior to isolating DNA [31].In those vesicles, also mRNA was found and sequenced.Transcripts of the SOS response were over-expressed relative to stationary culture cells, while in the DNA we found a coverage gradient along the ori-ter axis, indicating that the DNA originates from cells lysed while actively replicating.In summary, while remnants of DNA originating from outside the vesicles cannot be completely excluded, there is strong indication that it is really the DNA inside the vesicles that is enriched for the dif site." Line 59-60.Please explain how luminal DNA only was isolated from OMVs here.After reviewing the fluorescent staining images from cited study, the vesicles shown with FM1-43 lipid stained and DAPI stained OMVs did not always show that DAPI-stained DNA was solely located within OMV lumen (Fig 1G -H) of the cited reference #4.How much DAPI signal was in the background of the image and was the image background corrected?Was this cited study (reference 4) the only evidence of luminal DNA in OMVs to prove the author's arguments?
Answer: We expanded the methods section to provide a detailed description of the vesicle isolation and removal of external DNA.Indeed, the fluorescence micrographs of D. shibae OMVs showed rare cases where the DAPI signal might originate from the presence of DNA attached to the vesicles (there was almost no background-signal and no subtraction had been performed).However, when we developed the method further, we sequenced DNA from both, DNase treated and untreated vesicle enrichments and could show that digestion of unprotected DNA results in a reduction of reads mapping outside the ter region.Although we cannot exclude remaining residual external DNA, the observed enrichment of mapped reads clearly shows that DNase treatment was successful.There are numerous studies that showed DNA within vesicles, although for most of them no sequencing data exist.We expanded the introduction accordingly.
Line 85. "It is therefore expected that it might contain also DNA other than dif."Hard to follow statement; many grammar issues hindering meaning.This type of situation occurs often in this short article.Please carefully edit the article for grammar.
Answer: We modified this part in order to make it clearer: "For P. aeruginosa the sequenced DNA reportedly originated mainly from OMVs formed in biofilms during explosive cell lysis.In this process the whole cellular DNA content is released and can be attached to the surface or included in the lumen of newly formed vesicles.It is therefore expected that over-replicated DNA from the last stage of cell division might not be particularly enriched." Line 81 (Paragraph).This whole section is missing DNA GenBank accession numbers and reference citations for the DNA described and used in the study.The authors should make more effort to properly cite prior collected data used in the study.Figure 1 and 2 of the author's study should provide an added chart or quantification of the total sequenced DNA reads, where the % of ter and dif identified in coverage is summarized with respect to other possible identified or unknown DNA sequences recovered from the OMVs.How much of the OMV DNA recovered (% of total) is actually ter or dif DNA as opposed to other random DNA fragments?
Answer: Accession numbers and references had been provided in Supplementary Table S1 before.
We provide this information now in Table 1 together with read mapping statistics showing an enrichment of the ter region in the analysed strains.As the dif site itself is much smaller than the read size, it is in our opinion not of much use to calculate mapping statistics based on read counts.We compared the counts mapped to the ter region to counts from random regions outside ter.As the random number generator had the tendency to produce smaller numbers within the required range, samples near the start of the DNA sequence were overrepresented.Therefore, we divided the chromosome into 10 equal parts excluding ter and sampled 20 times within each segment.As we have only one ter site (sample size = 1) compared to many randomly sampled sites, no statistical test could be applied in a meaningful way.However, we think that the enrichment of ter becomes clear from the test as well as graphical inspection shown in the figure.
Line 94-104.Given that XerCD mutants produced very different amounts of vesicles per ml amounts, did DNA extractions from all OMV preps account for the differences in vesicles per ml?This could significantly skew the arguments made in the next paragraph which state fold change increases and decreases in ter and dif DNA content.But, this may actually be due to reductions in the isolated particles/ml of OMVs recovered by these mutant OMVs.Can the authors comment on this?
Answer: The same amount of input DNA was used for all samples and correction for sequencing depth was performed prior to analysis.The different amount of vesicles in the enriched fraction from the mutants should therefore not influence the read composition of the sequenced samples.Furthermore, the very specific changes in both directions directly at the dif site are a strong indication that the observed patterns are not due to such a general effect.
Line 105: is this paragraph continuing arguments from the paragraph above?It is not clear what results are being discussed?Assume OMV DNA from XerC/D mutants?Unclear and hard to follow.

of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. "
This could significantly skew the arguments made in the next paragraph which state fold change increases and decreases in ter and dif DNA content.But, this may actually be due to reductions in the isolated particles/ml of OMVs recovered by these mutant OMVs.Can the authors comment on this?Line 105: is this paragraph continuing arguments from the paragraph above?It is not clear what results are being discussed?Assume OMV DNA from XerC/D mutants?Unclear and hard to follow.Lines 119-122: Unclear wording; what does "The peak of DNA in OMVs" really refer to here?What peak?The statement in lines 120-122 is very hard to follow as written.What is meant or actually referred to by "independent mutants" here?
Line 123: Sentence fragment."TheXerCDrecombinationcomplex is highly adaptable."Adaptabletowithrespecttowhat?Line 128: Assume E. coli gene deletion mutants are meant here in the statement "...functionally impaired ΔxerC and ΔxerD mutants..." ?Line 134: Please clearly state other species in the study ter region enrichment also occurs in in this statement.Staff Comments:Preparing Revision GuidelinesTo submit your modified manuscript, log onto the eJP submission site at https://spectrum.msubmit.net/cgi-bin/main.plex.Go to Author Tasks and click the appropriate manuscript title to begin the revision process.The information that you entered when you first submitted the paper will be displayed.Please update the information as necessary.Here are a few examples of required updates that authors must address:• Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER.•Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file.•Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file.• Manuscript: A .DOC version of the revised manuscript • Figures: Editable, high-resolution, individual figure files are required at revision, TIFF or EPS files are preferred For complete guidelines on revision requirements, please see the journal Submission and Review Process requirements at https://journals.asm.org/journal/Spectrum/submission-review-process.Submissions