Pseudomonas aeruginosa clinical blood isolates display significant phenotypic variability.

Pseudomonas aeruginosa is a significant threat in healthcare settings where it deploys a wide host of virulence factors to cause disease. Many virulence-related phenotypes such as pyocyanin production, biofilm formation, and twitching motility have been implicated in causing disease in a number of hosts. In this study, we investigate these three virulence factors in a collection of 22 clinical strains isolated from blood stream infections. Despite the fact that all 22 strains caused disease and came from the same body site of different patients, they show significant variability in assays for each of the three specific phenotypes examined. There was no significant correlation between the strength of the three phenotypes across our collection, suggesting that they can be independently modulated. Furthermore, strains deficient in each of the virulence-associated phenotypes examined could be identified. To understand the genetic basis of this variability we sequenced the genomes of the 22 strains. We found that the majority of genes responsible for pyocyanin production, biofilm formation, and twitching motility were highly conserved among the strains despite their phenotypic variability, suggesting that the phenotypic variability is likely due to regulatory changes. Our findings thus demonstrate that no one lab-assayed phenotype of pyocyanin production, biofilm production, and twitching motility is necessary for a P. aeruginosa strain to cause blood stream infection and that additional factors may be needed to fully predict what strains will lead to specific human diseases.

Introduction

In recent years, healthcare-associated infections have become an increasing burden on the American healthcare system [1]. One of the major pathogens that contributes to this problem is the Gram-negative rod-shaped bacteria Pseudomonas aeruginosa. While P. aeruginosa can be found in the environment, it is also known to infect a wide array of hosts from plants to humans [2-6]. The ability to infect such a diverse group of species is due in part to the large number of virulence factors P. aeruginosa utilizes. P. aeruginosa virulence regulation uses both chemical factors such as quorum sensing [7] and mechanical cues such as surface association [8,9] to modulate a large suite of virulence factors. These virulence factors include secreted factors like pyocyanin [10], which is regulated by quorum sensing; motility factors like those required for twitching motility [11], which are regulated by surface association; and structural factors like those involved in biofilm formation that are regulated both mechanically and chemically and notoriously difficult to treat with current antibiotics [12-14]. All these factors make P. aeruginosa a significant threat to human health.

In addition to possessing multiple virulence factors, P. aeruginosa can cause many different types of infections. For example, in humans P. aeruginosa infections can be found in sites including blood, skin, subcutaneous tissue, lungs, urinary tract, eyes, and ears. It is possible that P. aeruginosa has so many virulence factors because specific factors are specialized for specific host environments. For example, studies of P. aeruginosa strains isolated from the sputum of patients with cystic fibrosis suggested that there are specific pressures applied by this environment that drives mutations of P. aeruginosa, causing them to be niche specialists [15-21]. For body sites other than cystic fibrosis lungs, such as the blood, it remains unclear if P. aeruginosa acts as a niche specialist or if P. aeruginosa virulence factors act combinatorially. If strains act as niche specialists there should be little variability amongst a suite of virulence factors than can be assayed in the lab. Alternatively, clinical and environmental isolates can have genomic similarity while displaying phenotypic heterogeneity [22-24].

Here we addressed genomic and phenotypic heterogeneity of bloodstream infections by investigating a collection of 22 recent clinical strains from bloodstream infections of different patients. We measured pyocyanin, biofilm, and twitching assay to survey virulence traits regulated by different mechanisms. We were surprised to see a wide variability in all three phenotypes tested. To understand the genetic basis of this phenotypic variability, we performed whole genome sequencing on each of the strains. Relative to the substantial phenotypic variability displayed by the strains, the genes required for each behavior displayed significantly less variability. These results suggest that in blood infections, P. aeruginosa largely retains the structural genes required for multiple virulence-associated behaviors, that the strength of each of these behaviors is variable, and that no one behavior is necessary or strongly predictive of the ability to cause disease.

Materials and methods

Strains and growth conditions

All strains were grown at 37°C in liquid LB Miller (Difco) on a roller drum at 90 rpm. Solid media growth was done on LB 1.5% agar plates.

The RWJ strain collection contains de-identified clinical isolates from bloodstream infections and were obtained from Dr. Melvin Weinstein at the Rutgers Robert Wood Johnson Medical School.

Pyocyanin assay

Overnight cultures of the RWJ collection, as well as PA14 and PAO1 as pyocyanin positive controls and the pyocyanin-deficien phzS mutant as a negative control, were grown, and cell free supernatant was collected by pelleting cells at 8000 x g for 1 min. The supernatant was filtered through a 0.22 μm syringe filter to ensure the absence of cells. Aliquots of 50 μl of each supernatant was transferred to a well of a 96 well plate (Corning 3904) in triplicate. A spectrum ranging from 250–500 nm was taken for each well at 1nm increments on a microplate reader (Tecan). A blank of LB was also analyzed as a negative control and subtracted from the supernatant values. The integrated peak area from pyocyanin absorbance (306–326 nm) was calculated to determine the amount of redox active phenazines, primarily pyocyanin, which absorb light in this region.

Biofilm assay

The static biofilm assay protocol from [25] and [26] was used. Briefly, overnight cultures of the RWJ collection, as well as PA14 and PAO1 as biofilm positive controls, were diluted 1:100 into M63 minimal media supplemented with 1 mM magnesium sulfate, 0.2% glucose, and 0.5% casamino acids. Aliquots of 100 μl of each strain was placed into a round bottom 96 well plate (Corning 3788) in triplicate and grown overnight at 37°C. After overnight incubation, cultures were poured off and the plate was washed by submerging the plate into water. Each well was then stained with 125 μl of 0.1% (w/v) crystal violet solution in water. The plate was incubated for 10 minutes at room temperature after which the stain was poured off and the plate was rinsed by submersion in water. The remaining stain was solubilized with 125 μl of 30% acetic acid in water and incubated for 10 minutes at room temperature. The solubilized crystal violet was transferred to a new flat bottom clean 96 well plate (Corning 3370) and the absorbance of the wells was taken at 550 nm on a microplate reader (Tecan) for quantification.

Twitching assay

Individual colonies of cells of the RWJ collection, as well as PA14 and PAO1 as twitch positive controls, were picked with a 10 μl pipette tip and stabbed through the agar of a LB 1.5% agar plate and placed at 30°C for 4 days. After 4 days, the agar was gently removed from the dish and a sufficient volume of 0.5% (w/v) crystal violet in water was added to the plate until the surface covered. After 5 minutes of staining, the crystal violet solution was removed, and the plate was washed 3 times with water. The resulting crystal violet stained twitch rings were imaged using a Canon EOS Rebel T1i (Lake Success, NY) and measured in FIJI [27-29].

Statistical analysis

A unequal variance two-tailed Student’s T-test with the Bonferroni correction was used to determine statistical significance (padj <0.05 *, padj <0.01 **) using MATLAB (R2020a, MathWorks, Natwick MA) (Fig 1).

Fig 1

Virulence phenotype assays vary widely across 22 Pseudomonas aeruginosa bloodstream strains.

(A) Pyocyanin absorbance integrated from 306–326 nm. Mean and standard deviation shown from 3 biological replicates (B) Crystal violet staining of total biofilm production as determined by absorbance at 550 nm. Mean and standard deviation shown from 3 biological replicates. (C) Twitching radius determined after crystal violet staining, measured in mm. Mean and standard deviation shown from 3 biological replicates. A Student’s T-test with the Bonferroni correction was used to determine statistical significance (padj <0.05 *, padj <0.01 **). Strains were obtained from Robert Wood Johnson medical school and are listed as RWJ01-RWJ22 and colored in red. Reference lab strains show are PA14 (green) and PAO1 (purple).

The pairwise correlation coefficients were computed using MATLAB (R2020a, MathWorks, Natwick MA). When multiple hypothesis tests were performed (Fig 2) the pValue was corrected with the Bonferroni correction for multiple hypothesis tests. None of these comparisons failed to reject the null hypothesis (no correlation) in favor an alternative hypothesis (any non-zero correlation).

Fig 2

2D plots of normalized virulence data show no significant correlation.

(Top) Comparison of twitching distance and biofilm staining. Means of biological triplicates shown. Correlation padj = 1. (Middle) Comparison of biofilm staining and Pyocyanin absorbance. Means of biological triplicates shown. Correlation padj = 0.074 (Bottom) Comparison of twitching distance and pyocyanin production. Means of biological triplicates shown. Correlation padj = 1. RWJ strain data shown in red, PA14 data shown in green, PAO1 data shown in purple.

Whole genome sequencing and assembly

Overnight cultures of the RWJ collection, PA14, and PAO1 were grown and the genomic DNA of each strain was collected using the Qiagen DNeasy Blood and Tissue kit (cat no 6950). Genomic DNA was taken by the Princeton Genomics Core Facility and processed for next generation sequencing fragmenting and barcoding the samples for sequencing. A single lane of the MiSeq (600 nt) was used to sequence the strain collection.

After demultiplexing, raw reads were checked for quality control and barcode removal using the Trim Galore! 0.6.3. The resulting reads were then assembled into scaffolds using Unicycler 0.4.8.0 [30]. ORF calling in these scaffolded assemblies was performed by Prodigal 2.6.3 [31] and these protein and nucleotide databases were searched for relevant matches to a hand curated list of proteins of interest using blast+ [version 2.7.1]. Using blastp, we found the ORFs with the highest alignment score to the protein of interest (S2 Table). We then concatenated the nucleotide sequence for those entire ORFs to generate a super-gene. In order to generate the phylogenetic trees, multiple sequence alignment was performed by the combination of mafft [32], BMGE [33], and phy-ml [34]. As needed, MATLAB R2019a was used as general purpose scripting language to manage the semiautomated components of the pipeline.

For determining the similarity of proteins as seen in Fig 4 and S1 Fig we queried the assembled scaffold ORF databases for the virulence associated genes (S2 Table) using the published PAO1 reference genome for the best alignment protein. We then took the full ORF result of that search and used it to query back onto the PAO1 reference genome. If the reverse blast did not produce the starting gene query, we eliminated that gene from that strain in our table as we considered this to be an off target match. Similarity scores were calculated either as the alignment score divided by the perfect alignment of the template (S1 Fig) or of the resulting hit (Fig 4) using a BLOSUM62 scoring matrix. This method reduces the penalty for ORFs that run into the end of a contig and are therefore artificially short due to an assembly error instead of a biological difference.

Fig 4

Protein similarity amongst strains shows high homology amongst most genes.

Genes involved in the (A) pyocyanin and pyoverdine, (B) biofilm, (C) twitching were compared for similarity. (Left) Strains are ordered by their normalized phenotypic score from dark red indicating high to dark blue indicating low. (Right) Similarity scores between the gene in published PAO1 reference genome and ORF from our assembly. Dark purple on indicates 100 percent similarity with decreasing saturation indicating decreasing similarity. Gray boxes indicate that the gene was not found in the scaffolds of the strain and could indicate a loss of the gene or be the result of incomplete assembly of contigs. Checkerboard pattern indicates that the ORF was found running up to the edge of a contig and was scored based on the portion of the protein present. PA14 and PAO1 comparisons shown are based on our de novo assembly of these strains to provide a baseline for our sequencing and assembly pipeline. For scores calculated based on full length template see S1 Fig.

Results

Clinical strains show variability in multiple virulence-associated phenotypes

We obtained a collection of P. aeruginosa strains from bloodstream infections from the Robert Wood Johnson Hospital to investigate variability amongst strains from the same infection site. This collection contained 22 strains that had been previously isolated from the blood of 22 different patients. When the strains were plated on standard lab LB 1.5% agar plates there was a large amount of variability in colony morphology. 7 of the strains formed rugose colony biofilms while the other 15 formed more mucoid colony biofilms. In addition, 2 of the strains produced blue-green coloration of the agar near the plated bacteria indicating a large quantity of pyocyanin. When the strains were grown in standard LB liquid overnight, 5 of the cultures were a deep blue green, indicative of high pyocyanin production, while another 5 were a pale blue green color. The variability of both the colony morphology and the liquid growth culture colors led us to more rigorously quantify phenotypic variability amongst the strains. As positive control strains for each assay tested, we also measured the phenotypic response of PA14 and PAO1 as they have been well characterized by the field.

To measure pyocyanin production in a lab environment, we collected cell free supernatant and analyzed it by UV-Vis spectroscopy, measuring absorbance from 250–500 nm. The spectrum from a blank well containing only LB was subtracted from each of the sample spectra and the integrated peak area from pyocyanin absorbance (306–326 nm) was calculated. Confirming our initial qualitative observations, there were a wide range of pyocyanin levels ranging from ~2–13 integrated absorbance units (Fig 1A). PA14 showed pyocyanin levels similar to the highest pyocyanin producers of our strains, while PAO1 showed pyocyanin levels in the middle of the range produced by our collection. These results demonstrated that despite the fact that pyocyanin production has been previously correlated with virulence and pathogenicity in other infection sites (4, 10), P. aeruginosa can mount a successful blood infection even when the pyocyanin level, in a mono-cultured test tube setting, nears the bottom of our detection limit.

Another phenotype we sought to further quantify in our collection was biofilm formation. In order to measure the level of biofilm quantitatively we utilized a crystal-violet based microtiter plate biofilm formation assay [25,26]. This assay colorimetrically measures the amount of crystal-violet bound by the biofilm matrix. The amount of biofilm produced ranged from 0.06 to 0.88 (Fig 1B), with some strains making more biofilm than both PAO1 and PA14 and some making less than both PAO1 and PA14. Thus, biofilm production was also highly variable in the collection.

The final phenotype we investigated was the ability of P. aeruginosa to twitch, another behavior that has also been implicated as important for pathogenesis. Twitching is a surface associated motility phenotype where cells are able to move along the surface underneath soft agar. This phenotype is visualized by crystal violet staining of the resulting biofilm spread to show the furthest extent of motility. After 4 days of twitching at 30°C, we saw that while most of the strains twitched to a similar extent as PA14 and PAO1, there were 3 strains that appeared to be completely deficient in twitching and 4 strains that twitched further than PA14 (Fig 1C). Together with our pyocyanin and biofilm assay results, these findings reinforced the conclusion that individual strains from bloodstream infections are highly variable with respect to all lab phenotypes assayed despite the fact that all three phenotypes have been implicated as important for pathogenesis.

Virulence phenotypes show no significant covariance

Since each of the virulence phenotypes assayed showed significant variability, we sought to determine if any of them covary, which would indicate that they are coregulated, or if they vary independently, which would suggest that they might function combinatorially. Many virulence-related phenotypes can be co-regulated. For example, quorum sensing regulates both pyocyanin production and biofilm formation [35-37], while surface sensing regulated both biofilm formation and twitching motility [8,9,11,38]. However, in our analysis, pairwise comparisons of each of the phenotypes across the strains revealed no significant correlations (Fig 2, padj > 0.05). Twitching motility showed almost no correlation whatsoever with either pyocyanin production or biofilm formation (Fig 2). Meanwhile, pyocyanin production and biofilm formation displayed a slightly negative correlation, but this was not statistically significant (Fig 2 Middle, padj > 0.05). For this analysis, all data were normalized so that the phenotypes were on the same scale for each comparison (Fig 2).

While there was no significant covariance amongst the measured phenotypes, we also observed that there was no strain that was low across all the assays. Strains such as RWJ09 and RWJ12 were low in two of the assays, but high in a third (Fig 1). Other strains like RWJ01 and RWJ02 had no assays in which they were low (Fig 1). All of these data suggest that there is no “stereotypical” phenotype amongst the collection, but that each of the strains examined retained at least one of the pathogen-associated phenotypes tested.

Phylogenetic clustering and phenotypic clustering differ

The phenotypic variability of the collection in the biofilm, pyocyanin, and twitching assays, led us to investigate the genomes of all the strains by whole genome sequencing. In brief, we used Illumina 2x250 paired end nextgen sequencing. From these reads we assembled scaffolds using Unicycler 0.4.8.0 [30] and called ORFs using Prodigal 2.6.3 [31]. For details on assembly statistics see S1 Table. As a control for the efficacy of our bioinformatic pipeline, we also resequenced PA14 and PAO1 and used the same assembly pipeline to assemble the genomes de novo.

Following assembly of each strain’s genome, we used the housekeeping genes as a basis for phylogenetic analysis with the published PAO1 reference genome as a search template. Specifically, we generated a phylogenetic tree based on the distances for the combination of nine housekeeping genes (dnaE, guaA, gyrB, mreB, mutS, ppsA, recA, rpoB, and rpoD). Interestingly, we found several distinct phylogenetic clusters within our strains: RWJ05, RWJ15, RWJ11, and RWJ03 clustered with one another and with PAO1; RWJ02, RWJ10, RWJ14, RWJ08, and RWJ12 formed a distinct cluster also similar to PAO1; RWJ04 formed a distinct cluster with PA14; RWJ06, RWJ22, and RWJ18 form an additional distinct cluster; and RWJ13, RWJ21, RWJ16, RWJ19, RWJ09, RWJ17, RWJ20, RWJ07, and RWJ01 were dissimilar from both one another and from PAO1 and PA14 (Fig 3A).

Fig 3

Clustering of phylogenetic data and phenotypic data are not correlated.

(A) Phylogenetic tree of the RWJ samples, PA14, and PAO1 based on the genes dnaE, guaA, gyrB, mreB, mutS, ppsA, recA, rpoB, and rpoD. (B) Phenotypic clustering based on the 3 virulence assays tested. (C) The clustering order from the phylogenetic tree used to reorder the phenotypic data. White to dark green color used to display phylogenetic distance. White to dark pink color used to display phenotypic distance. (D) Pairwise phylogenetic distance compared with the pairwise phenotypic distances. No statistically significant correlation observed.

We noted that strains with similar genomes often exhibited different virulence-associated phenotypes. For example, RWJ04 phylogenetically clustered with PA14 but showed much lower twitching and pyocyanin production and higher biofilm formation than PA14 (Figs 1 and 3A). To examine the relationship between genomic and phenotypic similarity more systematically, we normalized the data from our phenotypic assays and used it to cluster the strains based on their phenotypes (Fig 3B). We observed that the clustering of the phenotypic data looked different from the phylogenetic clustering (Fig 3A and 3B).

To further compare the phenotypic and phylogenetic clusters we used the phylogenetic clustering order to reorder the phenotypic data. These reordered data were then put into a heat map (Fig 3C). The lower left portion of the heatmap is colored by the phylogenetic distance between two strains and the upper right is colored by the phenotypic distance. The lack of symmetry in this resulting heat map shows the lack of correlation between measuring the phenotypic and phylogenetic distances (Fig 3C). Since clustering adds an extra layer of factors into the data, we also plotted the distances between strains from both sets of data against each other and determined the correlation of these distances (Fig 3D). The plotted distances confirmed that there is no significant correlation between phylogenetic and phenotypic distances.

Most virulence phenotype-associated genes are highly conserved across the collection of strains

Our phylogenetic analysis was based on similarity across a set of housekeeping genes, but the virulence-associated phenotypes we assayed are known to be driven by specific sets of genes. We thus investigated whether there were specific genes that could explain some of the phenotypic variability in pyocyanin production, biofilm formation, and twitching motility. We first defined a distinct set of genes that have previously been associated with each of our three phenotypes. Using a custom MATLAB script, we determined the protein sequences of each of these genes of interest in each of the strains, as well as our assemblies of the reference strains PA14 and PAO1. This analysis included a total of 129 genes implicated in at least one of the three virulence-associated phenotypes. The search templates for these genes came from the published PAO1 reference genome [39]. Each of these homolog ORFs was given a similarity score compared to the reference protein (Figs 4, S1 and S2). Due to the read length and depth of coverage of our reads not every gene could be identified in every strain. If a gene was not identified in any strain, that gene was removed from further analysis. We also removed any ORFs that matched something other than the original template when realigned back against the reference PAO1 genome. Additionally, for ORFs that ran into the edge of the scaffolds, we computed similarity scores for the portion of the protein that was represented in our assembly.

We first examined the biosynthetic genes involved in pyocyanin production. We found that most of the pyocyanin genes searched for shared high homology amongst all the strains (Fig 4A). For example, a low pyocyanin production strain like RWJ09 exhibited similar homology in pyocyanin production genes to RWJ13, which had high pyocyanin production. Similarly, RWJ06, RWJ19, and RWJ07 each had nearly perfectly conserved pyocyanin genes yet exhibited high, moderate, and low pyocyanin production, respectively (Fig 4A). Thus, analysis of pyocyanin-related genes failed to identify a clear genetic signature associated with pyocyanin production (or lack thereof).

Analysis of biofilm production genes produced similar results to those of pyocyanin production, with high overall conservation of genes required for each phenotype and no clear genetic predictors of phenotypic strength (Fig 4B and 4C). For example, RWJ16 produced moderate biofilm while potentially lacking many biofilm genes whereas RWJ15 has high similarity to almost all biofilm associated genes but limited biofilm formation (Fig 4B). Meanwhile, FliC, the structural flagellin component, showed low homology in roughly half the strains, but this was not correlated to any particular level of biofilm formation (Fig 4B).

Similarly, when looking at genes involved in twitching, the variability of the few genes that displayed significant variability in their homology, like pilA and pilB, did not correlate with twitching motility distance (Fig 4C). We were able to identify strains that were lacking pilA or had low pilA homology but high twitching (RWJ18, RWJ15, RWJ11). Furthermore, RWJ03 and RWJ07 had similar homology profiles but differed in their twitch radii (Figs 1C and 4C).

The lack of correlation between the presence of genes required for a given phenotype and the strength of the phenotype could be the result of changes in regulatory proteins that modulate the levels of the genes responsible for the differences in phenotypes. We thus decided to examine quorum sensing genes, as these pathways are known to regulate genes involved in multiple virulence-associated pathways [7,35,36]. However, analysis of the quorum sensing gene sequences showed the same pattern as the other genes examined, with high overall homology and no clear correlation to any specific phenotype (S2 Fig).

Discussion

Due to P. aeruginosa’s ability to colonize and infect a wide array of hosts and body sites, we were interested if strains isolated from the same body site, in particular the bloodstream, of different patients would show variability in their phenotypes. There have been many studies on P. aeruginosa virulence factors implicating pyocyanin production, biofilm production, and twitching motility as important mediators of pathogenesis [4-6,40-45]. We observed high variability amongst the phenotypes we assayed, which is consistent with other studied that found high variability amongst isolates from other clinical sites such as urinary tracts and cystic fibrosis lungs [22-24]. It is possible that none of the three specific phenotypes we assayed are important for blood infections or are only induced in blood-like environments. Alternatively, we suggest that perhaps these virulence-promoting traits are partially redundant and can function synergistically when insufficient on their own. While one could continue to measure additional virulence factors, our survey of pyocyanin, biofilm, and twitching account for both mechanically- and chemically-regulated virulence factors. We note that we identified strains with low levels of each of the three individual phenotypes, but no strains in which all three phenotypes were low. Even strains such as RWJ16, RWJ21, and RWJ22, which were below average in all three normalized assay scores, had at least one phenotype at a moderate level. As such, our data support the combinatorial model as there is no one virulence-associated phenotype necessary to cause disease in the blood but all strains exhibited at least moderate levels of at least one phenotype tested.

To address possible genetic sources of the phenotypic variability observed, we performed whole genome sequencing of our bloodstream infection strains. We first established that the phenotypic and phylogenetic similarities across the strains were uncorrelated, as strains that were phylogenetically similar were phenotypically dissimilar. These results suggest that these phenotypes are not a product of any singular P. aeruginosa subtype amongst our collection from a particular body site.

Phenotypic variability can generally be attributed to structural or regulatory changes in the genome. We thus analyzed the conservation of >125 genes previously implicated in structural or regulatory functions associated with each of the three phenotypes investigated. Similar to other bloodstream isolate surveys [46,47], we found that most of these genes were highly conserved among our collection. In addition, the variability amongst genes that were not highly conserved did not correlate with the phenotypic variability, suggesting that these genetic changes could not explain the phenotypic changes. These data suggest that there may be additional factors regulating these behaviors such as the microenvironment of the bloodstream [46] or that their regulation is mediated by non-structural genetic changes, such as those associated with promoters, translation, or mRNA stability [48].

A possible confounding factor in interpreting the phenotypic variability we assayed is that growth conditions in the lab differ from those in the host. It is possible that the when the bacteria are in a bloodstream environment, virulence factor expression is modulated in a manner we did not detect in our assays. Future work to better mimic the complex environments bacteria encounter in a host will be needed to address this possibility [49-53]. As we do not have information regarding the patient’s outcome nor the specifics about their disease state in our deidentified samples we cannot assess how the phenotypic variability we saw in the lab assays compares to disease severity. There has been some success in identifying genomic signatures that are predictive of disease in mice [54-56].

Our work underscores the importance of multilevel investigations to determine the level of variability in strains and suggests that there might not be a “stereotypical” strain that defines a specific type of infection. While anti-virulence strategies have been proposed as a way to fight the rise in antibiotic resistance [57,58], these data suggest that in order to fight specific virulence factors we first need to know which ones are displayed by different strains at different infection sites. These findings underscore the need for quick diagnostic virulence assays to enable the success of anti-virulence strategies.

Protein similarity using full length template compared to ORFs.

Genes involved in the (A) pyocyanin and pyoverdine, (B) biofilm, (C) twitching, and (D) quorum sensing were compared for similarity. (Left) Strains are ordered by their normalized phenotypic score from dark red indicating high to dark blue indicating low. (Right) Similarity scores between the gene in published PAO1 reference genome and ORF from our assembly. Dark purple on indicates 100 percent similarity with decreasing saturation indicating decreasing similarity. Grey boxes indicate that the gene was not found in the scaffolds of the strain and could indicate a loss of the gene or be the result of incomplete assembly of contigs. Checkerboard pattern indicates that the ORF was found running up to the edge of a contig.

(PDF)

Click here for additional data file.

Protein similarity of quorum sensing genes.

Genes involved in the quorum sensing were compared for similarity. Dark purple on indicates 100 percent similarity with decreasing saturation indicating decreasing similarity. Grey boxes indicate that the gene was not found in the scaffolds of the strain and could indicate a loss of the gene or be the result of incomplete assembly of contigs. Checkerboard pattern indicates that the ORF was found running up to the edge of a contig and was scored based on the portion of the protein present.

(PDF)

Click here for additional data file.

Assembly statistics.

(PDF)

Click here for additional data file.

Genes of interest, their PAO1 reference name, and their associated biological activity category.

(PDF)

Click here for additional data file.

9 Mar 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Chih-Horng Kuo, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript titled “Pseudomonas aeruginosa clinical blood isolates display significant phenotypic variability” has been reviewed. While the study is important, there are substantial concern in the present form of the manuscript. Thus, the present manuscript should be revised.

The manuscript background should be amended, the rationale of the present study, what gap evident in the literature, how the present study mitigates those gaps, what are the specific objectives are to be clearly mentioned.

The authors mentioned that “We obtained 22 clinical blood isolates of P. aeruginosa from different patients from the Robert Wood Johnson Hospital to investigate variability amongst isolates from the same infection site” However, there are no specific details of the samples, The sample description should have proper details. Also, in the methodology, it is mentioned that “All strains were grown at 37 °C in liquid LB Miller (Difco) on a roller drum at 90 rpm.” Are you mentioning the isolates from the samples? No clear details.

What are the controls for each assay? These are not clear. Also, the statistical analysis should mention in the results and discussion. No clear idea about this.

The discussion should emphasize the novelty and uniqueness of the study. What gap mitigated in the literature also need mention, in comparison with latest papers.

The authors mentioned that "There was no significant correlation between the strength of the three phenotypes across isolates, suggesting that they can be independently modulated"". But in the results and discussion it is not easily comprehendible. The author should look into it and revise the manuscript accordingly.

Also, authors mentioned that “Our findings thus demonstrate that no one lab assayed phenotype of pyocyanin production, biofilm production, and twitching motility is necessary for a P. aeruginosa strain to cause blood stream infection and that additional factors may be needed to fully predict what strains will lead to specific human diseases” This is a major point according to this study. According to you, what are the probable reasons and causes, how the same enhances virulence of the organism? There should be separate discussion in these aspects in eth discussion section.

Reviewer #2: This is an interesting paper that describes an important phenomenon that has been a focus of study within the Pseudomonas aeruginosa research community for a number of years now. Genotypic and phenotypic heterogeneity has been described for several clinically relevant niches, none more so than the lungs of patients with Cystic Fibrosis. The current study reports on the phenotypic and genotypic heterogeneity of isolates from blood samples taken from patients and it focuses on three key virulence phenotypes, namely biofilm, twitching and pyocyanin production. The paper is focused and well written.

However, I have some questions relating to the experimental design and interpretation that should be dealt with by the authors. These relate to methodological queries, data availability and the scope of the introduction.

Comments:

The introduction should discuss more about the current state of the art with respect to phenotypic and genotypic heterogeneity with respect to microbial infections, and P. aeruginosa in particular. There are many important papers not cited and they would frame the current paper in a greater context were they to be so.

Page 4, Line 74: Why would the authors expect a ‘stereotypical’ phenotypic response when all the evidence relating to infection in the lungs would suggest that phenotypic and genotypic heterogeneity is common? The interesting finding from this study is that the same applies to bloodstream isolates.

Page 5, Line 95: Interesting that the bloodstream isolates were mucoid biofilms. Can the authors comment on the relevance of this?

Page 6, Lines 113-117: Can the authors be sure that PYO levels in the single isolates reflect the community in the bloodstream? If as they suggest there is significant heterogeneity, then multiple isolates from the same patients and same patient samples would be needed to determine whether or not PYO production was relevant to successful infection. I am not sure the data presented here is sufficient to make this assumption based on a single isolate, albeit from multiple patients. I also have questions relating to the measurement of PYO which I outline below. I am not familiar with this measurement methodology.

Page 7, Lines 154-157: Why would the authors expect the three phenotypes to co-vary? Are they regulated through common systems, is there previous data in the literature to suggest co-variation is likely?

Page 9, Lines193-204: The clustering described in the text is not reflective of what I see in the corresponding figure. Several strains described as clustering with PA14 do not (e.g. RWJ12 and RWJ14), it appears the text describes a different phylogenetic tree?

Page 12, Lines251-253: Can the authors clarify the issues with quality of the genomic assembly scaffolds? If there are issues, how do these impact on the robustness of the data presented?

Page 13, Line 284: Analysis of…………

Figure 1: Have the authors performed statistical analysis on these data and can they show which changes are statistically relevant?

Genome sequencing: Have the authors submitted their genomic data to the e.g. SRA database or other accessible system?

Genome comparison data: Gray boxes indicate that the gene was not found in the scaffolds of the strain and could indicate a loss of the gene or be the result of incomplete assembly of contigs. On this basis can the authors really infer any constructive insight from the analysis, given that the majority of genes were either present or at the end of a contig? Where differences exist, this could be a limitation to the data available? Also, given the comments relating to how regulation could be a factor rather than gene presence/absence, have the authors looked at the promoter regions upstream of the genes studied?

Figure S1: Perhaps I misinterpret the data here but why would PAO1 and PA14 have gray boxes for the phzA-G genes?

Pyocyanin assay: Can the authors provide a reference for the methods used for PYO quantification? The standard assay for pyocyanin analysis would be organic extraction in chloroform followed by 0.2 N HCl. Here the authors have used direct spectrophotometric analysis of the supernatant, which would contain more than one phenazine compound? Is there a reason the organic extraction was not performed, and can the authors clarify that their methodology detects PYO specifically?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Sinosh Skariyachan, PhD

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

27 May 2022

Reviewer #1: The manuscript titled “Pseudomonas aeruginosa clinical blood isolates display significant phenotypic variability” has been reviewed. While the study is important, there are substantial concern in the present form of the manuscript. Thus, the present manuscript should be revised.

Point: The manuscript background should be amended, the rationale of the present study, what gap evident in the literature, how the present study mitigates those gaps, what are the specific objectives are to be clearly mentioned.

Response: We thank the reviewer for pushing us to better clarify the rationale and background and have edited the introduction accordingly.

Point: The authors mentioned that “We obtained 22 clinical blood isolates of P. aeruginosa from different patients from the Robert Wood Johnson Hospital to investigate variability amongst isolates from the same infection site” However, there are no specific details of the samples, The sample description should have proper details. Also, in the methodology, it is mentioned that “All strains were grown at 37 °C in liquid LB Miller (Difco) on a roller drum at 90 rpm.” Are you mentioning the isolates from the samples? No clear details.

Response: We did not isolate the samples ourselves but have attempted to include as many details as possible about the samples. We have also been careful to use the term “strain” instead of “isolate” to make it clear that we did not isolate the strains ourselves in this study.

Point: What are the controls for each assay? These are not clear.

Response: We have used the well-characterized P. aeruginosa strains PA14 and PAO1, which are each known to be positive for twitching, biofilm formation, and pyocyanin production as positive controls and media as negative controls for our assays.

Point: Also, the statistical analysis should mention in the results and discussion. No clear idea about this.

Response: As suggested, we have added statistics for some of the other figures and a detailed statistical analysis section to the methods section.

Point: The discussion should emphasize the novelty and uniqueness of the study. What gap mitigated in the literature also need mention, in comparison with latest papers.

Response: We have expanded our discussion to emphasize these important points and contextualize them with other studies.

Point: The authors mentioned that "There was no significant correlation between the strength of the three phenotypes across isolates, suggesting that they can be independently modulated"". But in the results and discussion it is not easily comprehendible. The author should look into it and revise the manuscript accordingly.

Response: We have clarified this point in the text.

Point: Also, authors mentioned that “Our findings thus demonstrate that no one lab assayed phenotype of pyocyanin production, biofilm production, and twitching motility is necessary for a P. aeruginosa strain to cause blood stream infection and that additional factors may be needed to fully predict what strains will lead to specific human diseases” This is a major point according to this study. According to you, what are the probable reasons and causes, how the same enhances virulence of the organism? There should be separate discussion in these aspects in eth discussion section.

Response: This is an important point also touched upon by Reviewer #2. We have thus included in our discussion section a lengthy discussion of the causes, which may relate to different requirements for virulence factors in the blood, and more specifically highlighted the importance for future work beyond the scope of the current study to examine virulence in vivo.

Reviewer #2: This is an interesting paper that describes an important phenomenon that has been a focus of study within the Pseudomonas aeruginosa research community for a number of years now. Genotypic and phenotypic heterogeneity has been described for several clinically relevant niches, none more so than the lungs of patients with Cystic Fibrosis. The current study reports on the phenotypic and genotypic heterogeneity of isolates from blood samples taken from patients and it focuses on three key virulence phenotypes, namely biofilm, twitching and pyocyanin production. The paper is focused and well written.

However, I have some questions relating to the experimental design and interpretation that should be dealt with by the authors. These relate to methodological queries, data availability and the scope of the introduction.

Comments:

Point: The introduction should discuss more about the current state of the art with respect to phenotypic and genotypic heterogeneity with respect to microbial infections, and P. aeruginosa in particular. There are many important papers not cited and they would frame the current paper in a greater context were they to be so.

Response: As mentioned in response to Reviewer #1 we have now expanded the introduction and reframed it to emphasize that others have established virulence factor heterogeneity for P. aeruginosa in other infection sites but that isolates from blood have not been similarly examined.

Point: Page 4, Line 74: Why would the authors expect a ‘stereotypical’ phenotypic response when all the evidence relating to infection in the lungs would suggest that phenotypic and genotypic heterogeneity is common? The interesting finding from this study is that the same applies to bloodstream isolates.

Response: The reviewer raises a fair point and we have thus edited the text to remove this “straw man” and emphasize the novelty of our findings with respect to blood isolates (whose phenotypic heterogeneity had not been previously examined).

Point: Page 5, Line 95: Interesting that the bloodstream isolates were mucoid biofilms. Can the authors comment on the relevance of this?

Response: This is an interesting question but unfortunately we do not understand its relevance at this point. We thus note that our findings suggest that blood isolates retain mucoid characteristics and that further investigation will be needed to understand its relevance.

Point: Page 6, Lines 113-117: Can the authors be sure that PYO levels in the single isolates reflect the community in the bloodstream? If as they suggest there is significant heterogeneity, then multiple isolates from the same patients and same patient samples would be needed to determine whether or not PYO production was relevant to successful infection. I am not sure the data presented here is sufficient to make this assumption based on a single isolate, albeit from multiple patients. I also have questions relating to the measurement of PYO which I outline below. I am not familiar with this measurement methodology.

Response: We agree that we are measuring PYO levels in a different environment than the blood and from single isolates. We now clarify the caveat that PYO regulation may be different in the bloodstream, but also note that our assay focuses on the capacity of the strains to make this virulence factor in a context in which P. aeruginosa is known to make PYO. We hope that this clarification helps the readers to appreciate what we did, the importance of our findings, and recognize their limitations and importance of future experiments.

Point: Page 7, Lines 154-157: Why would the authors expect the three phenotypes to co-vary? Are they regulated through common systems, is there previous data in the literature to suggest co-variation is likely?

Response: Several virulence factors are indeed thought to be co-regulated. For example, both biofilm formation and PYO production are regulated by quorum sensing and both biofilm formation and twitching motility are regulated by surface sensing. We have better explained this background in the revision.

Point: Page 9, Lines193-204: The clustering described in the text is not reflective of what I see in the corresponding figure. Several strains described as clustering with PA14 do not (e.g. RWJ12 and RWJ14), it appears the text describes a different phylogenetic tree?

Response: There was an accidental line from a previous version that has been removed and has been double checked against the data presented. We thank the reviewer for catching this error.

Point: Page 12, Lines251-253: Can the authors clarify the issues with quality of the genomic assembly scaffolds? If there are issues, how do these impact on the robustness of the data presented?

Response: Like most genomic assemblies, our assemblies are incomplete due to the length of our reads and the depth of coverage. This does not affect the data presented as we accounted for representation issues in our analysis, as explained in the methods section and the caption to Fig. 4.

Point: Page 13, Line 284: Analysis of…………

Response: We forget the word “of” and fixed that issue.

Point: Figure 1: Have the authors performed statistical analysis on these data and can they show which changes are statistically relevant?

Response: As suggested, we have statistically compared all of our isolates to our PAO1 control and now show that some are statistically similar and some significantly distinguishable from PAO1.

Point: Genome sequencing: Have the authors submitted their genomic data to the e.g. SRA database or other accessible system?

Response: The data have been submitted to the NIH Genbank, we included the accession in the Assembly Statistics Table and have included it in the revised text.

Point: Genome comparison data: Gray boxes indicate that the gene was not found in the scaffolds of the strain and could indicate a loss of the gene or be the result of incomplete assembly of contigs. On this basis can the authors really infer any constructive insight from the analysis, given that the majority of genes were either present or at the end of a contig? Where differences exist, this could be a limitation to the data available? Also, given the comments relating to how regulation could be a factor rather than gene presence/absence, have the authors looked at the promoter regions upstream of the genes studied?

Response: First, it is important to clarify that the fully Gray boxes are the only genes that were not found. Checkerboard patterns reflect genes that were found but were at the end of contigs and thus partial sequences, but we still were able to determine the extent of gene similarity to these partial reads. Thus, for the majority of genes we can indeed provide insight. We agree with the reviewer that one should be cautious to draw strong conclusions about a lack of finding a particular protein and why we focus on the high level of protein similarity we find between the strains.

We have not looked at promoters as our analysis of protein similarity within the assay associated pathways was performed at the level of protein sequence instead of DNA, and not all of these genes have clearly defined promoters. Additionally, given that some proteins were at the ends of the contigs, these would be entirely lacking promoters, even when a portion of the coding sequence was identified.

Point: Figure S1: Perhaps I misinterpret the data here but why would PAO1 and PA14 have gray boxes for the phzA-G genes?

Response: To provide a more fair comparison we included PAO1 and PA14 in our genomic sequencing and assembly pipeline and for our analysis compared our de novo assembly to the published annotated genomes. This gives us a baseline for understanding the frequency of missing genes. We have clarified this point in the revised text.

Point: Pyocyanin assay: Can the authors provide a reference for the methods used for PYO quantification? The standard assay for pyocyanin analysis would be organic extraction in chloroform followed by 0.2 N HCl. Here the authors have used direct spectrophotometric analysis of the supernatant, which would contain more than one phenazine compound? Is there a reason the organic extraction was not performed, and can the authors clarify that their methodology detects PYO specifically?

Response: In P. aeruginosa, PhzS is responsible for making pyocyanin and 1-hydroxyphenazine. By comparing the UV-VIS spectra from WT and Tn::phzS, strains we saw a reduction in the absorbance around 320 nm indicating that the peak measured is pyocyanin as phzS is responsible for making pyocyanin. Thus, we used the absorbance in this region as a correlate for determining pyocyanin levels without performing a full extraction on all strains tested.

Submitted filename: Response to Reviewers PLoSOne.docx

Click here for additional data file.

14 Jun 2022

Pseudomonas aeruginosa clinical blood isolates display significant phenotypic variability

PONE-D-22-01014R1

Dear Dr. Gitai,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Chih-Horng Kuo, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Congratulations on the successful revision. The reviewers have some more minor suggestions, which I consider to be optional. Also, there are some minor mistakes, such as extra line breaks in lines 204/229/391. Please make all necessary changes and send the final version to editorial office for production, as well as arrange for data release in GenBank.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: The authors have carefully considered all points made in the initial peer review, have provided a rational and coherent response to all points made, and have revised the manuscript accordingly.

I just have two points to complete the review: Firstly, the comment on mucoidy of blood isolates refers to the fact that mucoid Pa strains are often an indication of a chronic persistent infection. One would imagine that the blood stream isolates would typically be more associated with an acute infection and therefore would not typically possess the mucoid phenotype. An interesting observation that does not need to be addressed, just a point of note. Secondly, the authors provide some good evidence for the PYO measurements, and the increased PYO levels in PA14 compared to PAO1 appear to support their methodology. It would be good for the authors to consider a methods paper that validates their detection method using (i) PYO (commercially available) and/or (ii) extracts from Pa isolates. Their methodology would enable a higher throughput analysis of PYO production in Pa, but further validation would be needed to justify its adoption as a core method.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

27 Jun 2022

PONE-D-22-01014R1

Pseudomonas aeruginosa clinical blood isolates display significant phenotypic variability

Dear Dr. Gitai:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Chih-Horng Kuo

Academic Editor

PLOS ONE