Fwd: Major Revision Decision - Blue-201312-2129PP

[Katrine Whiteson]

Hi Guys,

Hope you have all had a chance to enjoy the holidays! I'm got home from Northern California and I'm diving back in... 

We got reviews back on the perspective! They request revisions within 90 days, and the first 2 reviewers especially make helpful comments. 

Some reviewers suggest broadening this to include more than just CF, it's a reasonable point. The last reviewer suggests having a broader model including many other ecological principles, and I'm not sure if that's feasible. They also assume that respiratory infections are characterized by less diversity/richness, so they find the island biogeography prediction to be opposite of reality. There are other polymicrobial infections where greater diversity/richness is associated with infection (i.e. gingivitis), this may be an important point to emphasize. In CF I think an important reason for decreasing diversity with age and severity of disease is antibiotics. We certainly state this but maybe it needs to be introduced differently.

I will try to tackle these and send a draft around, if you have any suggestions based on the reviewer comments I am all ears.  

:)

Katrine

Begin forwarded message:

From: g-m...@northwestern.edu

Subject: Major Revision Decision - Blue-201312-2129PP

Date: December 23, 2013 1:57:36 PM PST

To: katrine...@gmail.com

23-Dec-2013

American Journal of Respiratory and Critical Care Medicine

Dear Dr. Whiteson:

I have evaluated the critiques of your manuscript entitled "The upper respiratory tract as a microbial source for pulmonary infections: Parallels from Island Biogeography".  Our reviewers raised concerns which preclude acceptance of your paper in its current form. We would, however, like to give you an opportunity to respond to reviewer comments in a revised version of your paper within 90 days (15 days if this is an Editorial).

Please understand an invitation to submit a revision does not imply a commitment to publish, since a revision still may not achieve a sufficient priority to warrant acceptance.

In your revision, please ensure that your references include the most current articles and information. In compliance with the Journal's policy on Prior Publication, you need to cite any abstracts (related to the research contained in the manuscript) in the last sentence of the Introduction section and also include those citations in the list of References. The recommended format is: "Some of the results of these studies have been previously reported in the form of an abstract(s)(References)." In addition, the abstract(s) must be listed in the References section of the manuscript. If the abstract(s) were distributed to meeting attendees, please cite as an unpublished paper; please refer to the appropriate sections of the Chicago Manual of Style (17.215, 17.68-71) for citation guidelines.

In addition to addressing the comments of reviewers, please provide point-by-point responses to all of the issues raised in my letter. When responding to the points in my letter and to each comment of the reviewers, please reproduce the pertinent sentences or word fragments from the reviewers' critiques and my letter immediately above each of your responses; number each point as comment 1 (C1), response 1 (R1), and so on.

Your revised manuscript should include:

1. a "Marked up" version of your previous manuscript (Use red font to indicate the revised portions of the manuscript.)
2. a "Clean" version of your revised manuscript (use black fonts, no red marked-up fonts)
3. point-by-point responses to reviewer comments and to any issues raised in this letter

To upload the revision and enter the "responses to reviewer comments", please log on to Manuscript Central at http://mc.manuscriptcentral.com/ajrccm and follow the steps below:

1. Go to the "Author Center" and click on the "Manuscript with Decisions" to locate the manuscript.

2. Click the "Create a Revision" link. This will bring you to the screen where you can enter the revision.

3. You will first see the "Decision letter with the Reviewer Comments”, which was also sent to you by email.

4. Underneath the decision letter, you will see a box labeled as "Response to Decision Letter", this box is your allotted space for entering the "Author Responses to Reviewers and Editors".

5. When completed, click the "Save and Continue" button, this will bring you to the screen where you can enter the submission questions.

6. When you have completed the submission questions, upload the files and select the appropriate "File Designation". The system will combine the files for the manuscript (i.e. text, figures, online data supplement, if included) into one PDF file.

7. Upload your clean version files (all black fonts, no red mark-up fonts) and your marked-up files. (Use red font to indicate the revised portions of the manuscript.)

8. You only need to upload one set of high-resolution figures if there are no changes. However, if you have previously submitted low-resolution artwork files, please submit the original source files for all figures with your revision and follow the journal artwork policies and guidelines outlined in the Instructions for Authors.

9. View the converted PDF, checking for errors or anything out of order. Then click the 'submit' button to finalize your submission.

In submitting a revision of your paper, please be sure that each author has submitted a disclosure of financial interest form (http://www.icmje.org/coi_disclosure.pdf). The form is also available from Manuscript Central under Instructions and Forms (http://mc.manuscriptcentral.com/atsjournals). The submitting author must upload the co-authors’ disclosure forms, if these have not been provided with the initial submission. When uploading forms, choose File Designation: "Disclosure of Financial Interest Form(S)."

At the time of submitting your revision, please ask all authors to submit a signed ATS journals copyright form (electronic signatures are acceptable). Click here to obtain a form: http://mc.manuscriptcentral.com/atsjournals. When completed, email the copyright form to fo...@thoracic.org. Please be sure to include the manuscript ID number in the email subject line. Co-authors may submit their copyright forms individually.

Thank you for submitting your paper to the Journal. I look forward to your response. If, however, I have not heard from you about this manuscript in 90 days, I will presume you have withdrawn the paper from consideration.


Sincerely,


Gökhan M. Mutlu, M.D.
Associate Editor
American Journal of Respiratory and Critical Care Medicine


Reviewer: 1

Comments to Authors
Blue-210312-2129PP

The upper respiratory tract as a microbial source for pulmonary infections: Parallels from Island Biogeography.


This opinion piece is well written and nicely introduces the readership to central tenets of ecological theory that apply to the human ecosystem. It also addresses a major contentious issue in the field of respiratory microbial research, in a manner which is both objective and informative for the readership of AJRCCM.  

My main concern is that a large section of the manuscript seems to target a single publication to address the question whether sputum is relevant to lower airway disease. I know that this is primarily by necessity because of the dearth of studies that examine both upper and lower airway samples in parallel in CF patients, but it would be useful to point out the benefits of that study prior to indicating it’s deficiencies.

The issue of whether sputum is a useful proxy for lower airway samples or whether lower airway samples are highly contaminated with oral secretion is pervasive across the spectrum of respiratory disorders studied. Therefore broader inclusion of studies beyond that of CF would make this a more comprehensive perspective piece, that has implications for other respiratory diseases. There have been other studies (beyond the CF field) that have examined upper and lower airway colonization patterns e.g. one by Morris et al, ((Am J Respir Crit Care Med. 2013 May 15;187(10):1067-75) using the neutral model theory, as well as others e.g. Iwai et al, J Clin Microbiol. 2012 Sep;50(9):2995-3002 that focuses on oral and airway microbiota in HIV-infected patients. It would be useful to bring such studies into the discussion to illustrate the points made.

It is useful also to include simple studies that demonstrate source and sink – Neils Hoiby has published a study demonstrating that the Pseudomonas aeruginosa strains found in the newly transplanted lungs of CF patients are identical to those residing in the sinuses of those patients (Ciofu, J Cyst Fibros. 2013 Dec;12(6):729-36. doi: 10.1016/j.jcf.2013.02.004.). Again, these studies are used to support the points made in the manuscript.

Minor points.

It would be useful to initially provide the bigger picture by mentioning that beyond island biogeography, the human ecosystem more broadly exhibits central tenets of ecological theory  – dispersal, succession etc. maybe useful to direct readers to a recent review article by Gonzalez et al (Gonzalez, A. et al., EMBO reports (2011) 12, 775 – 784) that quite nicely illustrates this point. Following this, it makes sense to hone in on island biogeography as a framework for tackling the issues discussed in the perspective.

Page 3, line 119-120. Should also mention other lung-specific immune responses e.g. alveolar macrophages that are dysfunctional in CF

Page 4, line 128 – add chronic sinusitis to the list

Page 4 line 144, please replace “flora” with microbes

Page 4, line 146

Suggest restructuring this paragraph:
However, invasive lower airway sampling methods such as bronchoalveolar lavage cannot be performed regularly. Thus, while sputum samples cannot provide good insight into spatial distribution of microbes within the airways, given that they are non-invasive, standardized, and exhibit similar microbial composition between lower and upper airways (8, 15, 16, 34), it seems sensible to use the information they provide to increase our understanding of airway microbial communities.

Given the readership, it would be very useful to make the point that standardization is key, irrespective of the sample type used for a study, as is the resolution of the approach used

Also useful to reinforce the point that the biological question should dictate the most appropriate sample type and that if it is not available, the next best proxy for that sample should be used.



Reviewer: 2

Comments to Authors
General Comment:
This manuscript submitted for a Pulmonary Perspective piece is well-organized and written fairly clearly that even readers who are not intimately engaged in pulmonary microbiome investigation should be able to follow. The authors propose application of the Island Biogeography model as a framework for understanding recent insights (and accompanying controversy) regarding the continuum of microbiota from the oropharynx to lower airways/lung.  It introduces another refreshing ecological perspective to this area of study that should help inform understanding of current evidence.

There are a few areas, however, in which the manuscript could be improved. These include some discrepancies/inaccuracies that ought to be corrected, as well as suggestions for clarifying arguments made by the authors.

Major comments:
The series of critiques based on conclusions drawn in the paper by Goddard et al. are pointed. Although already fairly well written, addressing the below could make the authors' counterarguments come across even stronger.

1. 1st critique- could the authors expound a little further on why examining the community in explant lung parenchymal samples may not be most relevant towards understanding community evolution factors at play in lung disease progression (i.e. patients with less severe CF; this is likely important even in COPD).  This reviewer wholeheartedly agrees with the authors, however, the underlying premise could be more explicitly stated to drive the point home for less expert readers. The fact being it is important to identify and understand the role of other microbiota, such as those represented in sputum, for their contributions to community function and potential influence on behavior of traditional pathogenic species. It seems likely that these interactions, in addition to other selective pressures like antibiotics, are important factors in the evolution of the “Climax” community which is most represented in end-stage lung disease.

2. The 2nd critique reads a little weak, particularly in lines 187-188 regarding the throat being devoid of microbes. The potential reasons mentioned for low recovery/burden of microbial DNA in throat swabs are all true. But another factor is simply the smaller represented area sampled using throat swabs vs. sputum.  In addition to inter-individual variability in how throat swabs are collected, the tool also just does not afford sampling as broad an area of the respiratory tract compared to sputum (especially if the latter were induced, but even with spontaneous expectorated sputum).  

3. The 3rd critique- last sentence (lines 202-203) is I think a very important argument and inclusion of a reference would bolster it. A suggestion is the 2011 paper by Hajishengallis et al. (Low-Abundance Biofilm Species Orchestrates Inflammatory Periodontal Disease through the Commensal Microbiota and Complement, Cell Host & Microbe)

Minor comments:
1. Pg 3 lines 98-101.  What the authors state in this sentence is not entirely congruous with the arguments made by Sethi in reference #13, in particular that sputum "should be disregarded."  In the referenced article Sethi does describe that sputum is subject to contamination by oral secretions but the overarching message is that sputum is still a useful specimen and research tool. Suggest modifying the sentence or the reference.

2. Pg 4 lines 128-130. Could the authors include references for the disease examples mentioned, especially for those without references made earlier (e.g. periodontitis, otitis media, COPD). This would strengthen the points being made.

3. Pg 4, lines 154-147. It should be noted that the explant lung study being discussed is CF.  Also in the earlier sentence Line 153, regional heterogeneity in bacterial community composition has also been described in COPD explant lung by Erb-Downward et al.






Reviewer: 3

Comments to Authors
The conceptual treatise entitled "The upper respiratory tract as a microbial source for pulmonary infections: parallels from island biogeography" is an interesting concept.  Unfortunately, it is superficial and does not discuss multiple published studies in pulmonary microbiology and infection.  The paper also really lacks good discussion and integration with pulmonary physiology, cell biology and medicine.  There are numerous concerns about both the discussion and utilization of the ecological model and the cited (or lack-thereof) data from the current state of the field in respiratory infection studies.   Furthermore, this paper discusses individual subjects as islands, which is completely inaccurate and inconsistent with the island biogeography model.  A major concern is that the proposed model is intended to inform our understanding of respiratory infection, but it actually predicts only a limited aspect of microbial diversity (species richness), and it appears to predict the opposite of what is truly observed in respiratory infections.  

General Concerns:

1.  While this model may be useful for understanding respiratory infections, all the discussion in this paper is on cystic fibrosis, so the title needs to be changed to include this caveat.  "The upper respiratory tract as a microbial source for pulmonary infections in cystic fibrosis..."

2.  The paper is written with the biogeography model as the primary point and the respiratory microbiology as an example, rather than the other way around.

3.  At multiple points the authors interchangeably use "species richness" with "species diversity" (e.g. Figure 1 legend, line 32). The distinction between the two is critical and very relevant to their model. Species evenness is an important component of community diversity that is unaddressed by the model depicted in Figure 1.

4.  A comprehensive model of lung microbial ecology should include at the least 1) species richness, 2) species evenness, 3) alpha and 4) beta diversity and 5) total bacterial burden. I fear the authors' island model only attempts to address one of these (species richness). And this element of diversity is actually _low_ in the case of respiratory infection, not high as this model would predict:

5.  Despite the authors' observation that some infections are polymicrobial (line 126), it remains the case that the vast majority of lung infections (pneumonia, COPD and bronchiectasis exacerbations, etc.) are caused by one or a small number of pathogens (even when studied using culture-independent techniques). This by definition reflects a low-diversity, low-species richness state. Yet the authors' proposed model predicts that in conditions of high immigration and low extinction (such as with aspiration and impaired ciliary clearance) there should be high species richness. Given that this model aims to explain the pathogenesis of respiratory infections, this is a critical and unadressed limitation of their proposed model. Doesn't it predict the opposite of what actually occurs in respiratory infection?

6.  There have been dozens of studies to date using culture-independent techniques to study bacteria in the lung, and nearly all have reported measures of community diversity. The authors do not discuss whether these findings to date support or refute their proposed model. For instance, Blainey et al found that diversity was significantly decreased in CF airways when compared to controls (Sci Transl Med. 2012 Sep 26;4(153):153ra130.). Wouldn't this island model predict the opposite, given the decreased extinction rate in CF airways due to impaired ciliary clearance?

7.  Throughout the manuscript, the authors use the verb "to colonize" in two importantly distinct senses: in the microbiologic sense ("microbial communities colonize various points in the respiratory tract") and in the ecological sense ("distance from colonization sources, colonization rates...") (both examples from the first sentence of Abstract). This is a potential source of considerable confusion. Would consider alternate terms to keep distinct concepts distinct, e.g. "inhabit" and "immigrate."

This is more than a semantic point. The microbiological sense of "colonize" implies that resident microbes are reproducing and susceptible to selective pressures in the host environment. A very active debate is ongoing whether the bacteria detected via culture-independent techniques in the lungs are alive and reproducing or rather transiently present after transit from the upper airways (see Morris et al. "Comparison of the Respiratory Microbiome in Healthy Nonsmokers and Smokers." American journal of respiratory and critical care medicine 187.10 (2013): 1067-1075.). The culture-independent techniques used in the studies cited by the current authors are based upon identification of bacterial DNA, and thus cannot discriminate between living and dead bacterial cells (see Pezzulo et al. "Abundant DNase I-Sensitive Bacterial DNA in Healthy Porcine Lungs and Its Implications for the Lung Microbiome." Applied and environmental microbiology 79.19 (2013): 5936-5941.).

Thus it is far from settled within lung microbiology to what degree the lung harbors "colonizers" in the microbiological sense that the authors use.

8.  The MacArthur model illustrated in Figure 1 is most coherent when explaining variation in species richness among multiple points/specimens/islands that are of varying sizes and distances from a common source of immigrating species. But in the authors' application of this to respiratory microbiology, they alternately refer to the same sites as "islands" as well as "sources" of immigrating species, which makes the model conceptually difficult to follow. For instance: in the legend to Figure 1, the oral cavity (circle #2) is described as analogous to a "large proximal island with high species diversity." But elsewhere, the oral cavity is described as the source of immigrant species to the lung (e.g. the title, Figure 2, line 67, line 70, line 91). If the oral cavity is the dominant source of lung species shouldn't it be on the X-axis in figure 1? Or shouldn't the "colonization rate" line in this figure be vertical (if the distance from source is zero)? What is the meaning of a "colonization rate" when referring to a single site (the mouth) as both the source and destination of immigration?

9.  In Figure 2 and various places in the text (e.g. line 207), other people are referred to as other islands. But the MacArthur model is useful only in predicting species richness in various islands of varying distance from a single immigration source. It is not useful in the prediction of immigration from island to island. As such it is not clear to me how the "people as other islands" concept would be illustrated on Figure 1. I worry the authors' broad use of "island" analogy makes the model considerably harder to understand and apply.

10.  The MacArthur/Wilson model of island biogeography was of considerable historical importance and has been of great influence in the field of ecology, but it was as speculative as it was empirical at the time of its publication. The current authors at times present the model as if it were a validated empirical observation (e.g. line 32, Figure 1 legend). They would do well to acknowledge its numerous limitations given its centrality to their argument (e.g. Lomolino, Mark. "A call for a new paradigm of island biogeography." Global Ecology and Biogeography 9.1 (2000): 1-6.)

11.  For most of the manuscript the authors appear to be proposing a general model of lung microbiology and diversity, but at many times their discussion is rather narrowly limited to the conditions of cystic fibrosis (e.g. paragraph starting on line 70, line 92 ["...so that their connection with the CF lung can be understood"], lines 134-205, 228-9). It is unclear to me how generalizable the authors feel this discussion to be.

12.  In their discussion of the potential utility of the model (line 213), the authors mention potentially finding sources of opportunistic pathogens and quarantining infectious patients. These activities are already widely undertaken, and it's not clear to me how this model would change our practice or research strategies. The authors should make a stronger case for why this model is a conceptually or practically useful one.



Specific comments:

13.  Line 65: What is meant by "microbes immigrating to the lung come from various sources including air, water and food, but also from the oral cavity..."? Are the authors referring to aspiration in the cases of water and food, or via hematogeneous spread of microbes from swallowed food via the GI tract? It would seem they mean the latter (given the "but also from the oral cavity" clause), but this seems improbable and unusual to me given the infrequency of detectable bacteremia. Don't virtually all microbes in the lungs originate from above the vocal cords?

14.  Line 83: This paragraph is highly speculative (oral microbes transmitting metabolites and DNA-encoded traits to the lungs and impacting communities without physically migrating there) and feels out of place. It's not clear to me how it fits into their island model.

15.  Line 109: I believe the authors have misunderstood or miscommunicated the findings of the Morris 2013 study. That study's finding was that certain species appear to be enriched in the microbial communities detected in BAL, not in the upper airways (as summarized by the current authors).

16.  Line 111: The authors write: "Healthy lung communities often resemble those in CF lungs except they lack key pathogens, e.g. Neisseria." No source is provided for this, and I believe it is incorrect. The largest study to date comparing CF subjects to healthy controls was Blainey et al. "Quantitative analysis of the human airway microbial ecology reveals a pervasive signature for cystic fibrosis." Science Translational Medicine 4.153 (2012): 153ra130-153ra130. In that study, CF samples had consistently distinct microbiota from healthy subject samples, even when pathogenic species were removed from analysis.

17.  Line 137: "...most lung airways are sampled by inducing sputum..." I believe this is incorrect both clinically (in which spontaneously expectorated sputum is more common) and regarding lung microbiota studies (in which BAL is the most commonly used specimen). I suspect the authors are restricting their scope to CF. Similarly, the discussion between lines 146-149 appears limited to CF. Sputum is rarely used in lung microbiome studies of other lung diseases or of healthy control subjects.

18.  Lines 134-205: This whole section appears to be a defense of the continued use of sputum for study of CF, which seems like a technical question beyond the scope of the island model discussion. Much of it seems like a targeted editorial of the Goddard PNAS study, which is rather narrow for a Pulmonary Perspectives manuscript.

19.  Line 248: The authors write "When microbes that are typical of the oral cavity are found in sputum or bronchoalveolar lavage samples from CF or COPD patients, they should be considered colonists, rather than contaminants." Surely this is sometimes true, but I fear the authors aren't allowing for true contamination with oral microbiota, which is unquestionably a real phenomenon. They should either narrow this claim or propose means for discriminating between "colonists" and "contaminants."

[mfip mfip]

Dear Katrine,

Dear all,

I've just opened my gmail box and found your message.

Please find an attached doc that I've prepared to summarize diversity in chronic inflammatory pulmonary diseases using the different articles I'm currently reading for our Vertex Grant application.

Cheers

2013-12-30 Katrine Whiteson <katrine...@gmail.com>

--
You received this message because you are subscribed to the Google Groups "CFTelluride2013" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cftelluride20...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.