Comprehensive Genomic Characterization of Staphylococcus aureus Isolated from Atopic Dermatitis Patients in Japan: Correlations with Disease Severity, Eruption Type, and Anatomical Site

ABSTRACT Atopic dermatitis (AD) shows frequent recurrence. Staphylococcus aureus is the primary microbial component in AD and is associated with disease activity. However, traditional typing methods have failed to characterize virulent AD isolates at the clone level. We conducted a comprehensive genomic characterization of S. aureus strains isolated from the skin of AD patients and healthy donors, comparing the whole-genome sequences of the 261 isolates with anatomical and lesional (AD-A)/nonlesional (AD-NL)/healthy sites, eruption types, clinical scores, virulence, and antimicrobial resistance gene repertoires in Japan. Sequence type (ST) diversity was lost with worsening disease activity; ST188 was the most frequently detected ST in AD-A and had the strongest correlation with AD according to the culture rate and proportion with worsening disease activity. ST188 and ST20 isolates inhabited all skin conditions, with significantly higher proportions in AD skin than in healthy skin. ST8, ST15, and ST5 proportions were equivalent for all skin conditions; ST30 was detected only in healthy skin; and ST12 was detected only in AD skin. ST97 detected in AD-A and healthy skin was clearly branched into two subclades, designated ST97A and ST97H. A comparison of two genomes led to the discovery that only ST97A possessed the complete trp operon, enabling bacterial survival without exogenous tryptophan (Trp) on AD skin, where the Trp level was significantly reduced. Primary STs showing an AD skin inhabitation trend (ST188, ST97A, ST20, and ST12) were all trp operon positive. The predominant clones (ST188 and ST97) possessed almost no enterotoxin genes, no mecA gene, and few other antimicrobial resistance genes, different from the trend observed in Europe/North America. IMPORTANCE While Staphylococcus aureus is a member of the normal human skin flora, its strong association with the onset of atopic dermatitis (AD) has been suggested. However, previous studies failed to assign specific clones relevant to disease activities. Enterotoxins produced by S. aureus have been suggested to aggravate and exacerbate the inflammation of AD skin, but their role remains ambiguous. We conducted a nuanced comprehensive characterization of isolates from AD patients and healthy donors, comparing the whole-genome sequences of the isolates with anatomical and lesional/nonlesional/healthy sites, eruption types, clinical scores, virulence, and antimicrobial resistance gene repertoires in Japan. We demonstrate that specific clones are associated with disease severity and clinical manifestations, and the dominant clones are devoid of enterotoxin genes and antimicrobial resistance genes. These findings undermine the established notion of the pathophysiological function of S. aureus associated with AD and introduce a new concept of S. aureus colonization in AD.

infection are frequent in AD patients. The strength of the study is the study design, incorporating multi-site sampling and large surface area of sampling and a significant number of AD patients. The methods are described in detail, and sequenced were uploaded to a public database. The quality of written English is acceptable. The weaknesses of the study/manuscript are the data presentation, e.g. redundancy in figures and text. One of the main issues is the conclusion that AD skin is more likely to be inhabited by S. aureus. This notion, however, is not fully clear since only S. aureus positivity over all swabs taken was presented. For this claim, it would be necessary to present the positivity rate in AD patients (x/135) and not (x/579).
Moreover, this study could explore the aspect of clonality in a single patient (did all isolates in different localizations from patients inhabited by S. aureus belong to the same clone?). This analysis would add valuable information to this manuscript/study. The results presented can be more concise to highlight the main points.
Major comments • There are too many figures/tables for the amount and depth of data presented. Some figures may not be necessary, Fig 1  (redundant to text), Fig 3-8 could be reduced. Many of these figures can be moved into supplementary data. The manuscript should be more focused on highlighting the main findings.
• Results lines 110-111: the denominator for percent positivity (% positivity) should be the number of patients in the first part of the results. This would give some impression on how many AD patients were colonized/infected and allow the readers to see how the S. aureus positivity in AD patients in Japan compared to other studies. Using the number of swabs as a denominator can distort the statistics. (Maybe I have missed this in the result since it was not mentioned in the beginning). • Relating to the previous comment, did all patients have active lesions or just a proportion of all AD patients? This was unclear since the number of swabs was used as a denominator for the calculations. In my opinion, the culture rates are secondary to the %patients positive with S. aureus inhabitation and %patients with active lesions. Similarly, it was mentioned in lines 278-279 that the rate of S. aureus in AD patients is high. Please provide the % of AD patients with S. aureus colonization/infection (x/135). • It was not mentioned whether the S. aureus isolates from a single patient are clonal (or not). This would be an interesting result for this specific study design. • Discussion lines 274-276: AD is classically linked to a stronger Th2 response and more S. aureus colonization, whereas Psoriasis is linked with a stronger Th1 response and less S. aureus colonization. You may want to mention that the reason might be immunological/priming and not primarily strain variations/virulence profiles. There are numerous studies on this. Th1/Th17 response is important for S. aureus elimination in the skin.
• The text of the results is redundant and can be shortened (in combination with figures). • Lines 309-310: strain abundance may be linked to molecular epidemiology of circulating strains in the region. Therefore, this argument is not convincing and is speculative. The concept of strain specificity for AD patients is not yet proven and established. The published data do not support this hypothesis since strain distribution is often linked to local/regional epidemiology. The study prepared by Obata et al., described the characterization of S. aureus isolates collected from atopic dermatitis patients in Japan. This is a very interesting study and likely of interest for the readership of the journal. The study was well performed and I would have just a few minor comments described below.
1. On line 141, it's not clear what are the ST belonging to CC188. Also, if an ST is not typeable, how it can be associated with an CC. Perhaps, 2. This reviewer is not really sure the phylogenetic analysis, as it seems that the analysis ended up clustering the isolates into CC anyway. So, the analysis seems unnecessary. Please, clarify. 3. On line 152, can you define "superantigen". I am not sure this term is necessary, and authors can simply say "antigen genes". 4. On line 186, you can probably describe a little more detail about what "truncated" means. Would that be a premature stop codon? 5. The sentence in lines 202 -204 reads as conclusion and it should be used in such section. The same for the sentence in lines 205 -206. 6. I am not why the authors prepared an specific analysis/section for the comparison between isolates from Japan and Denmark, and in fact, characteristics of isolates from Europe and NA are also described. This seems a little off, and perhaps fits best in the discussion section as part of a comparison analysis with previously published data in the literature. 7. The initial paragraphs of the conclusion section repeat to much all results generated and I would suggest authors to minimize that and go straight to the discussion, as possible.
Staff Comments:

Preparing Revision Guidelines
To 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. 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 of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. " 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. We thank you for your assistance in the submission process and look forward to hearing from you. Response: In accordance with this suggestion, we have dropped Fig. 1A, and Fig. 1B-1D have all been moved to the supplemental figures (Fig. S2A-C).     7 Lines 309-310: strain abundance may be linked to molecular epidemiology of circulating strains in the region. Therefore, this argument is not convincing and is speculative. The concept of strain specificity for AD patients is not yet proven and established. The published data do not support this hypothesis since strain distribution is often linked to local/regional epidemiology.
Response: The sentence in lines 309-310 of the original version has been deleted, indicated via the strike-through feature.

Minor comments:
8 Line 141, why is the ST untypable?
Response: For the untypable ST, the seven allele numbers were arcC_3, aroE_1* (novel allele), glpF_1, gmk_8, pta_1, tpi_1, and yqiL_1. Based on this allele pattern and the eBURST analysis, it was assumed to belong to CC188. For brevity in the text, "ST untypable" has been deleted (revised version line 148, indicated via the strike-through feature) and replaced by "the untyped STs". 10 Line 258: ethnic differences? Maybe "regional differences" is better?
Response: Line 351 (revised manuscript): We have revised this phrase to "regional differences," as suggested.
11 Lines 301-303: the meaning of the sentence is unclear Response: We have deleted this section since it is unclear.

Reviewer #2 (Comments for the Author):
The study prepared by Obata et al., described the characterization of S. aureus isolates collected from atopic dermatitis patients in Japan. This is a very interesting study and likely of interest for the readership of the journal. The study was well performed and I would have just a few minor comments described below.