Integrated Phenotypic, Proteomic (MALDI-TOF MS), and Genomic (WGS) Investigation of a Prolonged Hospital Outbreak of Pseudomonas aeruginosa with High Biofilm-Forming Capacity

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents an integrated investigation of a prolonged ICU outbreak of Pseudomonas aeruginosa, combining antimicrobial susceptibility testing, biofilm phenotyping, MALDI-TOF MS clustering, and whole genome sequencing. The integrated design is a clear strength, and inclusion of an environmental sink isolate adds epidemiological relevance. The comparison between MALDI-TOF MS clustering and WGS phylogeny has practical translational potential. The manuscript is well organized, and the research question is clearly stated. However, there are several concerns that I believe need to be addressed. I have organized my comments into major and minor categories below.

Major Comments

1. Supplementary Table S2 shows variable BUSCO completeness among sequenced isolates, ranging from approximately 50% (PA_11) to 83.5% (PA_14), with several isolates below 70%. Because the manuscript interprets gene absence biologically in the resistome and virulome heatmaps (Figures 2 and 3)—for example, T3SS effectors or pyoverdine components—incomplete assemblies could produce false negatives. I would recommend adding a concise limitation statement acknowledging this, or a brief justification for why these assemblies are adequate for the specific analyses performed.
2. If I understand correctly, the abstract states that “the blaVIM gene was detected in 32 isolates,” but the results section (lines 134–135) reports 22 isolates (56.4%), consistent with Table 2. The number 32 matches the meropenem-resistant isolate count reported in line 131, which I suspect is the source of the confusion. I would recommend correcting the abstract to clearly distinguish meropenem-resistant isolates (n = 32) from VIM-positive isolates (n = 22).
3. Selection criteria for sequenced isolates. The authors state that 12 isolates were selected for WGS to capture outbreak diversity, but the specific selection logic is not described. Since the genomic conclusions depend entirely on this subset, I would suggest a brief, concrete explanation—for example, whether isolates were chosen to represent different MALDI-TOF clusters, time points, resistance profiles, or specimen types.
4. Clarification of “ND” in Table 2. Multiple meropenem-resistant isolates are labeled “ND” under the carbapenemase report column. It is unclear whether “ND” means “tested and not detected” (true negative) or “not tested.” This distinction is important for interpreting resistance mechanisms and carbapenemase prevalence. I would recommend clarifying this explicitly.
5. The manuscript interprets the sink isolate PA_35 as a potential reservoir and notes its clustering with clinical isolates (PA_27, PA_22), but pairwise SNP distances are not reported. Including key distances or a small supplementary matrix would make the environmental transmission inference more concrete and easier to evaluate.

Minor Comments

1. The results jump from section 2.4 to 2.6; section 2.5 appears to be missing.Please correct it.
2. VIM is incorrectly defined as “Verona interferon-encoded metallo-β-lactamase” in line 69 and in the abbreviations table. The correct term is “integron-encoded.”
3. ST11 vs. ST111. Line 348 refers to “ST11” which should be “ST111.”
4. Section 4.10 gives the approval date as 06 April 2022, while the IRB statement says 06/04/2023. I would suggest clarifying which is correct.
5. Section 4.8 lists accessions for 12 isolates, but the Data Availability Statement lists only 9 (PA_02, PA_11, and PA_14 appear to be missing from the latter).
6. The resistome analysis specifies AMRFinderPlus, but the virulome analysis is described only as “targeted gene searches” without naming the database or thresholds used. I would suggest specifying the database and the identity/coverage parameters for reproducibility.
7. Terms such as “macro-groups,” “clusters,” and “internal groupings” are used somewhat interchangeably. I would suggest defining these consistently upon first use.

Author Response

Based on the comments and suggestions made by Reviewer 1, we have made respective modifications to the original manuscript above.

Major Comments

Comment 1: Supplementary Table S2 shows variable BUSCO completeness among sequenced isolates, ranging from approximately 50% (PA_11) to 83.5% (PA_14), with several isolates below 70%. Because the manuscript interprets gene absence biologically in the resistome and virulome heatmaps (Figures 2 and 3)—for example, T3SS effectors or pyoverdine components—incomplete assemblies could produce false negatives. I would recommend adding a concise limitation statement acknowledging this, or a brief justification for why these assemblies are adequate for the specific analyses performed.

Author’s answer: We want to thank reviewer for his/her constructive comments and thoughtful suggestions that allowed us to improve the original manuscript. We are pleased to inform you that thanks to his/her recommendations and we have included a paragraph on limitations in the Discussion section (lines 374-381) acknowledging that absences in resistome and virulome reports should be interpreted with caution, especially in assemblies with lower completeness, since sequencing was performed only with long reads from Oxford Nanopore, which may underestimate the gene content in genomes with lower completeness.

Comment 2: If I understand correctly, the abstract states that “the bla_VIM gene was detected in 32 isolates,” but the results section (lines 134–135) reports 22 isolates (56.4%), consistent with Table 2. The number 32 matches the meropenem-resistant isolate count reported in line 131, which I suspect is the source of the confusion. I would recommend correcting the abstract to clearly distinguish meropenem-resistant isolates (n = 32) from VIM-positive isolates (n = 22).

Author’s answer: We appreciate reviewer comment. The number of VIM-positive isolates (n=22) was corrected in the abstract, and the number of CRPA isolates (n=32) was added. Thank you for your correction that it will avoid misunderstandings by the Readers.

Comment 3: Selection criteria for sequenced isolates. The authors state that 12 isolates were selected for WGS to capture outbreak diversity, but the specific selection logic is not described. Since the genomic conclusions depend entirely on this subset, I would suggest a brief, concrete explanation—for example, whether isolates were chosen to represent different MALDI-TOF clusters, time points, resistance profiles, or specimen types.

Author’s answer: Thank you very much for this observation. We have included an explanation in the Materials and Methods section (lines 428-432) indicating that the isolates were chosen to represent the diversity of the outbreak, considering the clusters obtained in the proteomic analysis, incorporating different moments of the outbreak, variation in resistance profiles, and different types of origin, including the environmental isolate.

Comment 4: Clarification of “ND” in Table 2. Multiple meropenem-resistant isolates are labeled “ND” under the carbapenemase report column. It is unclear whether “ND” means “tested and not detected” (true negative) or “not tested.” This distinction is important for interpreting resistance mechanisms and carbapenemase prevalence. I would recommend clarifying this explicitly.

Author’s answer: We thank the reviewer for this comment and clarify that in Table 2, the “MEM” column corresponds to the meropenem susceptibility profile (R, S, or I) that is determined for all isolates. On the other hand, the “Carbapenemase report” column does not represent the general screening for carbapenemases, but specifically the determination of VIM (either the blaVIM gene or the VIM protein). To avoid confusion, we have changed the name of the column to “VIM determination”. In addition, ND means “not detected”, the isolate was evaluated for VIM, but neither the blaVIM gene nor the protein was detected, so we have clarified in the footnote that ND corresponds to “evaluated and not detected”. Thank you for calling this lapse to our attention.

Comment 5: The manuscript interprets the sink isolate PA_35 as a potential reservoir and notes its clustering with clinical isolates (PA_27, PA_22), but pairwise SNP distances are not reported. Including key distances or a small supplementary matrix would make the environmental transmission inference more concrete and easier to evaluate.

Author’s answer: We thank the reviewer for the suggestion and have written a paragraph to report the pairwise SNP distances related to the possible reservoir (PA_35) (lines 246-251). In addition, we have compiled a complete matrix of pairwise SNP distances, which is attached as Supplementary Table S5. With this, the structure of the phylogenetic tree is supported by quantitative evidence.

Minor Comments

Comment 1: The results jump from section 2.4 to 2.6; section 2.5 appears to be missing. Please correct it.

Author’s answer: Thank you very much for your comment. Section 2.4 was duplicated, but we have corrected this error. We apologize for this lapse.

Comment 2: VIM is incorrectly defined as “Verona interferon-encoded metallo-β-lactamase” in line 69 and in the abbreviations table. The correct term is “integron-encoded.”

Author’s answer: We appreciate the reviewer's comment. The definition of VIM has been corrected throughout the manuscript, replacing “Verona interferon-encoded metallo-β-lactamase” with the correct term: “Verona integron-encoded metallo-β-lactamase”. We apologize for this gap.

Comment 3: ST11 vs. ST111. Line 348 refers to “ST11” which should be “ST111.”

Author’s answer: We appreciate the reviewer's comment. The typographical error in the indicated line has been corrected, replacing “ST11” with “ST111.”

Comment 4: Section 4.10 gives the approval date as 06 April 2022, while the IRB statement says 06/04/2023. I would suggest clarifying which is correct.

Author’s answer: Thank you very much for your observation. The information was verified and the approval date was corrected to maintain consistency between Section 4.10 and the ethics committee (IRB) statement. The correct date is April 6, 2023, so Section 4.10 was updated accordingly. We apologize for this confusion.

Comment 5: Section 4.8 lists accessions for 12 isolates, but the Data Availability Statement lists only 9 (PA_02, PA_11, and PA_14 appear to be missing from the latter).

Author’s answer: Thank you for your comment. The consistency between Section 4.8 and the data availability statement was reviewed, and the latter was corrected to include the 12 corresponding accesses, incorporating the missing isolates (PA_02, PA_11, and PA_14).

Comment 6: The resistome analysis specifies AMRFinderPlus, but the virulome analysis is described only as “targeted gene searches” without naming the database or thresholds used. I would suggest specifying the database and the identity/coverage parameters for reproducibility.

Author’s answer: We appreciate the reviewer's comment. The description of the virulome analysis was strengthened to ensure reproducibility. In this study, the virulome was not determined by searching an external database, but rather through genomic annotation with Bakta (including its Bakta DB database) and subsequent targeted extraction of virulent genes from the annotation results. We have added to the Materials and Methods section the tool used (Bakta) and the database used (Bakta DB) (lines 448-449, 452-453). Thank you for improving the procedure description.

Comment 7: Terms such as “macro-groups,” “clusters,” and “internal groupings” are used somewhat interchangeably. I would suggest defining these consistently upon first use.

Author’s answer: Thank you very much for your comment. The terminology was standardized throughout the manuscript to avoid the interchangeable use of terms. The term “clusters” was used to refer to the main groupings and “subclusters” for the internal groupings, maintaining this nomenclature consistently throughout the text. We apologize for this inconsistent description in the original version of the manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

The article Integrated phenotypic, proteomic (MALDI-TOF MS), and genomic (WGS) investigation of a prolonged hospital outbreak of Pseudomonas aeruginosa with high biofilm-forming capacity presents interesting results demonstrating carbapenemases, biofilm-forming ability, and determinants of Pseudomonas aeruginosa isolates. I recommend the article for publication after minor changes, which I believe will improve the article.

The points for changes are indicated below:

Introduction

There are a few articles discussing the phenotypic diversity of Pseudomonas aeruginosa clinical and environmental isolates. I think these findings should at least be mentioned in the Introduction.

Results

1. It is hard to read the section due to the confusion with the number of isolates:

110 Thirty-eight clinical strains of P. aeruginosa causing nosocomial infections ..

In the same paragraph, the authors discuss Table 1. However, the Table shows 39 isolates.

119 In addition, one environmental P. aeruginosa isolate

Most probably, 39 is the result of adding one environmental isolate, which is quite confusing for the reader who already looked at the Table 1 described before the environmental isolate was mentioned and saw 39 isolates rather than 38.

2. 128 The 38 clinical isolates showed high resistance to commonly used antibiotics such as ..

In the same paragraph, the authors discuss Table 2. However, the Table 2 again shows 39 isolates.

3. Is resistance to antibiotics not tested for the environmental isolate?

 What is the reason for it? In the next section, the ability to form biofilm was tested for all 39 isolates.

4. Abbreviations should be listed in alphabetical order.

Author Response

Based on the comments and suggestions made by Reviewer 2, we have made respective modifications to the original manuscript above.

Comment 1: Introduction

There are a few articles discussing the phenotypic diversity of Pseudomonas aeruginosa clinical and environmental isolates. I think these findings should at least be mentioned in the Introduction.

Author’s answer: We would like to thank reviewer 2 for his positive and constructive comments that have helped us to considerably improve our original manuscript. In response to his first suggestion, we have included in the Introduction a brief mention of the phenotypic diversity described for clinical and environmental isolates of Pseudomonas aeruginosa, supported by additional and relevant articles (lines 62-65).

Comment 2: Results

1. It is hard to read the section due to the confusion with the number of isolates:

110 Thirty-eight clinical strains of P. aeruginosa causing nosocomial infections ..

In the same paragraph, the authors discuss Table 1. However, the Table shows 39 isolates.

119 In addition, one environmental P. aeruginosa isolate

Most probably, 39 is the result of adding one environmental isolate, which is quite confusing for the reader who already looked at the Table 1 described before the environmental isolate was mentioned and saw 39 isolates rather than 38.

Author’s answer: Thank you very much for your comment. We have improved this section to avoid confusion between the number of clinical isolates and the total number in the study. We have specified from the first sentence that a total of 39 isolates were included (38 clinical and 1 environmental recovered from the ICU sink, PA_35) (lines 107-108), and the clinical and epidemiological description corresponding to the 38 clinical isolates is maintained, with reference to Table 1. This improves clarity for the reader without altering the structure of the text. We apologize for this confusion.

Comment 3:

1. 128 The 38 clinical isolates showed high resistance to commonly used antibiotics such as ..

In the same paragraph, the authors discuss Table 2. However, the Table 2 again shows 39 isolates.

Author’s answer: We appreciate the reviewer's comment. To clarify, antimicrobial susceptibility testing was performed only on the 38 clinical isolates; the environmental isolate (PA_35) was not included in the susceptibility panel and is therefore not included in the resistance counts described in that paragraph. Consequently, Table 2 presents only the 38 clinical isolates (PA_35 is not included). Additionally, for the environmental isolate, only the VIM determination was reported in the text. To avoid further confusion, we have added a clarifying sentence in Section 2.2 explicitly stating that phenotypic AST was performed only on the 38 clinical isolates. Thank you for calling our attention to this matter.

Comment 4:

1. Is resistance to antibiotics not tested for the environmental isolate?

 What is the reason for it? In the next section, the ability to form biofilm was tested for all 39 isolates.

Author’s answer: We thank the reviewer´s comment. Phenotypic antimicrobial susceptibility was evaluated only for the 38 clinical isolates, since AST processing was performed within the hospital laboratory workflow and the environmental isolate (PA_35) was not associated with a patient (no code/clinical order), which limited its execution in the routine automated testing system. However, for PA_35, the resistance determinant profile was reported through genomic analysis of the resistome along with the other sequenced genomes. Regarding biofilm formation, this assay was performed for all 39 isolates because it corresponds to a phenotypic research evaluation that can be applied to clinical and environmental isolates available in the collection and is particularly relevant for interpreting persistence and the possible role of the environment in the context of the outbreak.

Comment 4:

1. Abbreviations should be listed in alphabetical order.

Author’s answer: We appreciate the reviewer's comment. The list of abbreviations has been reorganized so that it is presented in alphabetical order.