New insights into the role of Cutibacterium acnes -derived extracellular vesicles in inflammatory skin disorders

  1. Maria Pol Cros
  2. Júlia Mir-Pedrol
  3. Lorena Toloza
  4. Nastassia Knödlseder
  5. Marc Güell
  6. Julien Maruotti
  7. Christos C. Zouboulis
  8. Maria-José Fábrega Fernández

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Background

Cutibacterium acnes ( C. acnes ) is one of the most prevalent bacteria that form the human skin microbiota and, depending on multifactorial conditions it can help to maintain the skin homeostasis. Actually, different phylotypes of C. acnes have been associated with different degrees of acne vulgaris development, while others, such as the H1 subtype, have been detected in patients with non-acneic skin. However, due to the physiology of the skin, the skin microbiota neither has direct access to the skin’s sebaceous glands nor to the main immune cells, as they are protected by a sebum layer. Therefore, the inter-kingdom communication relies on secreted factors and bacterial extracellular vesicles (EVs). In this context, the purpose of this project was to study the role of EVs secreted by three different phylotypes of C. acnes (A1 as pathogenic, H1 as beneficial and H2 as commensal).

Results

Main findings showed that the proteomic profile of the cargo embodied in the EVs reflects unique characteristics of the different C. acnes phylotypes in terms of lifestyle, survival and virulence. Moreover, in vitro skin models showed an extended pro-inflammatory modulation of A1 EVs, while H1 EVs displayed a high sebum-reducing potential.

Conclusions

This study has highlighted the role of C. acnes EVs as key modulators during skin alterations, specially H1 EVs as an alternative based-natural treatment to fight acne vulgaris symptomatology.

Article activity feed

  1. Review Commons

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

    Learn more at Review Commons

    Reply to the reviewers

    'The authors do not wish to provide a response at this time.'

  2. Review Commons

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

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    Summary:

    Extracellular vesicles (EV) from three different phylotypes (A1, H1, H2) of Cutibacterium acnes were analyzed for size distrubution, protein content and inflammatory effects on cellular systems in vitro. Main findings were that EV composition differs between the the phylotypes that by other have been suggested to have pathogenic or more beneficial properties. Furthermore, A1 EVs induce more proinflammatory signals than H1 EVs. The conclusion are that EVs are key modulators of during skin alterations, and H1 EVs are suggested as a treatment of acne vulgaris symptomatology.

    Major comments:

    1. The purpose is stated as "to study the role of EVs secreted by three different phylotypes of C. acnes (A1 as 25 pathogenic, H1 as beneficial and H2 as commensal)". This is in my mind a quite vague purpose that needs to be sharpened. What role are you talking about and where is it relevant? During infection, in experimental systems or as potential treatment?
    2. The result section in the abstract is very short. At present it is not possible to understand what you have done by reading this. I suggest reducing the introduction part of the abstract and focus on your detailed results instead.
    3. The conclusions in the abstract is a bit difficult to understand a potentially very far reaching. You have not in my mind shown that EVs are "key modulators during skin alterations". You have shown that EVs can modulate cellular responses in vitro which is far from skin alterations, but important in its own right.

    Furthermore, "EVs as an alternative based-natural treatment to fight acne vulgaris symptomatology" is not really a supported conclusion but rather a potential future implication. The wording is also very odd; what does "alternative based-natural treatment" mean? You also come back to this in the main conclusion in the end. This needs to be clarified!

    1. Line 465. You assume that your strains have lost virulence factors during evolution. This is just a wild guess without genetic analysis. It could be anything from real gene loss, via gene regulation, to post-translational regulation.
    2. Line 510. You should tone done your claims about a closer picture of real skin in acne vulgaris. You are using models!
    3. Since you are claiming an essential role for EVs during skin alterations, you are missing essential controls in your system. What happens if you add just washed bacteria to your systems? If you get the same signals, EVs are not essential but can have the same effects as the parent bacteria.

    Minor comments:

    Reference list All species should be in italics, No Upper Case Within Article Titles, and journals should be abbreviated consistently.

    Significance

    General assessment:

    The most important aspect of this study is that EVs from different phylotypes of C. acnes have cellular effects that correlates with the pathogenic or beneficial profile of the parent bacteria. The major limitations are that the EVs are not compared with the parental bacteria in the systems and unsupported far reaching conclusions about using EVs as treament.

    Advance:

    The study is the first one looking at detailed protein patterns in different phylotypes of C. acnes and trying to link this to biological activity.

    Audience:

    Basic researchers in microbiology will read this with interest. At the moment, the translational/aspects are just suggested and not tested.

    Expertise:

    Infection medicine, experimental treatment, development of biological drugs, inflammatory diseases, anaerobic bacteria, antimicrobials, bacteriophages, commensals

  3. Review Commons

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

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    The work by Pol et al. describes the proteome and the in vitro effect, in four different cell culture models, of extracellular vesicles (EVs) isolated from Cutibacterium acnes of three different phylotypes found in acneic (A1) and normal skin (H1 and H2). They found that the EVs from H1 and H2 strains seem to evoke less pro-inflammatory effector mechanisms and even anti-inflammatory cytokines. A1 EVs apparently carry a higher number of proteins that include virulence factors and possible pro-inflammatory molecules that have not been specified. Although the number of proteins tends to be quite high in microorganism EVs, the presence of other groups of molecules can not be ignored when drawing conclusions, however. Their results seem to support the conclusions that should anyway be softened, however some points related to the EV concentrations used in the analysis should be cleared, besides other details, in order to make the results stronger to the eyes of expert readers. 1. The abstract needs improvement. You can decrease the background information to a minimum necessary and give more result information.2. In introduction and throughout the text be careful to cite highly updated works when starting with "recent data..." - for e.g., in lines 39-43 the reference cited dates to 2017.3. The Brucella culture medium was chosen to grow C. acnes for isolation of EVs. This is a complex rich medium whose components could eventually associate to the bacteria and EV surfaces, resulting in proteome artifacts. How was that controlled? On the other hand, culture time was long (7 days) for that species. By 7 days in that medium, in which phase of the growth curve are the bacteria, considering that at stationary phase membranes from dead cells could co-precipitate with EVs? 4. EV preps were kept at -20oC. For how long? Are the properties of these EVs maintained fairly intact at these conditions?5. l. 317: I suppose NTA is a more quantitative imaging technique, but not more precise.6. l. 321: Can you present the EV yield/bacteria for each sample? Do the phylotypes grow at similar speed rates? 7. l. 328-331: There seems to be some artifactual effect in the Qubit protein estimation because it's clear from the SDS-PAGE gel that the 3 samples do not have the same 7 micrograms that they should according to the dosage methodology. Sample A1 possibly does considering that many of the protein bands are quite fat. Therefore, you can not compare the samples in terms of diversity or amount if there is not an internal quantity control in the proteome analysis. All the experiments that compared different concentrations of EVs among phylotypes could be compromised by an artifactual protein estimation. Please comment and justify.8. Gene ontology analysis refers to total EV proteins analyzed in each phylotype, is that correct? Supporting information Tables 1, 2, 3: The table headlines should specify that those identified proteins were found EXCLUSIVELY in each of the haplotypes. 9. l. 356: The extension of red staining within the cells seems to reflect that many EVs were internalized. Could you estimate the number of internalized EVs/cell? Please change "positive control" in the figure for EV or anything you choose, considering that this is your experimental result and not a positive control.10. l. 406 and others: the EVs induced or stimulated (not displayed) secretion of mediators.11. In Discussion, please exploit even further the individual proteins found in the proteome to suggest their presumed function in the  various effects observed in the work. In l. 555-556, please rephrase the sentence to make it clear, also softening the conclusion by using "suggested" instead of "proved". Although English is readable throughout the text, edition is needed to improve grammar specially in Discussion where there are numerous inadequacies.

    Significance

    The characterization of the microorganism EVs and their role in pathogenesis and eventual protection of the diseases they cause has incresed significantly and any new investigation brings new light to the subject. With the present work, the authors claim that EVs isolated H1 strains from normal skin could be useful in the treatment of acne based on their set of results.

  4. Review Commons

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

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary:

    This manuscript describes differences between three different C. acnes strains in their proteomics, internalization, and induction of host cell gene expression. These differences are assessed through what the manuscript reports to be extracellular vesicles, but the relevance of these particles is not clear. The strong differences in immunomodulatory phenotypes across strains is of potential interest, but more work needs to be done to demonstrate rigor, relevance, and clear interpretation.

    Major concerns:

    My primary concern is the relevance of these purported extracellular vesicles, which could have been produced by the filtering and centrifugation process. Can these particles be observed without this processing? Alternatively, can the authors demonstrate that these particles have distinct proteomic properties from the cells from which they were isolated?

    A second major concern is the framing of these 3 strains. The evidence of association of type I strains and acne is weak. Most papers making this claim have not compared the same sampling sites on subjects of similar ages, or at least have not reported subject information well enough for this evidence to be clear. Moreover, it is not clear if the difference across 3 strains here are generalizable to these large clades or specific to these chosen strains. Notably, H1 strains are much more closely related to H2 strains than either is to A1 strains, challenging the meaning of the distinction between H1 and H2 strains and more generally the classification as probiotic, commensal, and pathogen. Type IB strains (which includes H1 and H2) have even been suggested to be more harmful in prosthetic infections [PMID: 34361935]. Regardless of the naming, a comparison between A strains and H strains is of interest - but a second A strain is required to make this evolutionary comparison.

    • Line 321: A claim is made about EV production rate across strains, but no statistics are performed and a number is only provided for one of the three strains.
    • Line 332: H1 and H2 are much more similar in their genomic content. Do the authors have a proposed reason for A1 and H1 sharing more proteomic similarity than H1 and H2?
    • Which genome was chosen for analysis of the proteomic data? Might classification be biased towards strains for which the reference genome more closely matches the amino acid content of the analyzed strains?
    • Line 341: For the biological function analysis, it is not clear if enrichment presented is relative to other strains or to the reference genome.
    • Figure 6 needs to show individual dots of experimental replicates to enable assessment of variation.
    • How were the probes for qPCR chosen? This is important for understanding multiple hypothesis correction. Were any others tested?

    Minor concerns:

    • Line 39: What about the vaginal microbiome?
    • Line 51: Context for this depth is needed. How deep is a pilosebaceous unit?
    • Line 56: It would be more correct to say that microbiome alterations are associated with the development of acne.
    • Line 62: C acnes is known to break down sebum, so why is it assumed that sebum is a barrier for C acnes contact with host cells? What about secreted products that aren't in EVs?
    • Line 64: 500 nm is getting close to the size of a bacterial cell
    • Line 66: Citation needed.
    • Line 92: Is this also done inside of a bag system as in the above section or in an anaerobic chamber?
    • Line 96: The point of a 75 mm filter is not understood by this reviewer.
    • Line 371: I could not locate a p-value in this entire section around gene expression induction in host cells. I see that there are statistics in the figures, but this needs to be indicated in the text as well.

    Significance

    This manuscript describes differences between three different C. acnes strains in their proteomics, internalization, and induction of host cell gene expression. These differences are assessed through what the manuscript reports to be extracellular vesicles, but the relevance of these particles is not clear. The strong differences in immunomodulatory phenotypes across strains is of potential interest to those studying C acnes biology, but more strains would need be tested to understand the relevance of these differences evolutionarily and more work would have to be done to understand the relevance of these EV particles. Claims about relevance to acne overstate the knowledge and consensus in the filed.

  5. Version published to 10.1101/2022.12.15.520547v2 on bioRxiv
  6. Version published to 10.1101/2022.12.15.520547v1 on bioRxiv