Application of mini-MLST and whole genome sequencing in low diversity hospital extended-spectrum beta-lactamase producing Klebsiella pneumoniae population.

Studying bacterial population diversity is important to understand healthcare associated infections' epidemiology and has a significant impact on dealing with multidrug resistant bacterial outbreaks. We characterised the extended-spectrum beta-lactamase producing K. pneumoniae (ESBLp KPN) population in our hospital using mini-MLST. Then we used whole genome sequencing (WGS) to compare selected isolates belonging to the most prevalent melting types (MelTs) and the colonization/infection pair isolates collected from one patient to study the ESBLp KPN population's genetic diversity. A total of 922 ESBLp KPN isolates collected between 7/2016 and 5/2018 were divided into 38 MelTs using mini-MLST with only 6 MelTs forming 82.8% of all isolates. For WGS, 14 isolates from the most prominent MelTs collected in the monitored period and 10 isolates belonging to the same MelTs collected in our hospital in 2014 were randomly selected. Resistome, virulome and ST were MelT specific and stable over time. A maximum of 23 SNV per core genome and 58 SNV per core and accessory genome were found. To determine the SNV relatedness cut-off values, 22 isolates representing colonization/infection pair samples obtained from 11 different patients were analysed by WGS with a maximum of 22 SNV in the core genome and 40 SNV in the core and accessory genome within pairs. The mini-MLST showed its potential for real-time epidemiology in clinical practice. However, for outbreak evaluation in a low diversity bacterial population, mini-MLST should be combined with more sensitive methods like WGS. Our findings showed there were only minimal differences within the core and accessory genome in the low diversity hospital population and gene based SNV analysis does not have enough discriminatory power to differentiate isolate relatedness. Thus, intergenic regions and mobile elements should be incorporated into the analysis scheme to increase discriminatory power.

Introduction

Klebsiella pneumoniae (KPN) frequently causes community and hospital acquired infections including pneumonia, urinary tract infections and pyogenic liver abscesses [1, 2]. The main KPN transmission reservoirs are the gastrointestinal tract and the hands of hospital personnel and patients. Nosocomial KPN isolates often display highly resistant phenotypes with an extended-spectrum beta-lactamases producing (ESBLp) KPN prevalence between 2% and 55% [3-5].

Typing methods to discriminate different bacterial isolates from the same species are essential epidemiological tools. In populations with a high prevalence of ESBL, the knowledge of bacterial population structure and dynamics is especially important in outbreak detection and intervention. To monitor the bacterial population, cheap, rapid and robust methods are needed. In our previous study, we proved that mini-MLST, a method derived from multi locus sequence typing (MLST) in which costly and time-consuming sequencing is replaced with high resolution melting analysis, is suitable for long term prospective KPN population screening. Currently, besides KPN [6], mini-MLST has been established for Staphyloccoccus aureus [7], Enterococcus faecium [8] and Streptococcus pyogenes [9]. However, its lower discriminatory power makes mini-MLST insufficient to identify outbreak strains and it needs to be combined with more sensitive methods.

Recently, whole genome sequencing (WGS) has revolutionized our ability to differentiate between bacterial strains at the entire genome’s DNA sequence level. For bacterial typing, WGS has two major approaches–core genome multilocus sequence typing (cgMLST) and single nucleotide variant analysis (SNV) [10]. CgMLST is based on an allele numbering system of a pre-determined set of genes. An advantage of cgMLST’s approach is its inter-laboratory portability and the existence of public databases e.g. https://www.cgmlst.org/ncs, http://enterobase.warwick.ac.uk/ or https://pubmlst.org/. SNV analysis is based on mapping raw sequence reads against a reference genome and detecting nucleotides that vary within the dataset. This approach provides an even higher resolution power than cgMLST, but is far more computationally intensive than cgMLST analysis and interpreting the results is more complex [11, 12].

As generating WGS data become more accessible, rapid and cheap, bottlenecks remain in proper pre-sequencing sample selection and post-sequencing data analysis [11]. The main bottlenecks include the need to critically evaluate the raw sequencing data, some knowledge and skills in programming and improvements in data analysis to translate the enormous amount of obtained data into understandable results for health professionals [13]. Knowledge of the local bacterial population’s genetics characterisation is also crucial for the results to be correctly interpreted, as there are no general thresholds of relatedness [10].

The main objectives of this study were i) to characterise the ESBLp KPN population in our hospital using mini-MLST prospective typing ii) to evaluate core and accessory genome single nucleotide variant analysis contribution in possible outbreak detection within the low diversity ESBLp KPN hospital population.

Material and methods

Clinical isolates

The study was conducted at the University Hospital Brno (Brno, Czech Republic), a tertiary care hospital with more than 2,000 beds and 5,000 employees. There are more than 1,000,000 people treated in out-patient clinics and over 70,000 patients hospitalized every year. During the systematic strain collection between 7/2016 and 5/2018, we collected all ESBLp KPN isolates from high risk departments (Department of Internal Medicine–Hematology and Oncology, Department of Internal Medicine, Geriatrics and Practical Medicine, Department of Anesthesiology and Intensive Care Medicine) and all isolates from neonates, new-borns and children. In total, we collected 922 ESBLp KPN isolates (Table 1). For the purposes of this study, 24 ESBLp KPN isolates from our previous study collected between 1/2014 and 10/2014 were also included [14]. All isolates were collected during routine practice and were made completely anonymous.

Table 1

Epidemiological data; n = 922.

Gender	no. of isolates (%)
Male	510 (55.3)
Female	412 (44.7)
Patient age	mean (range)
	56 years (0 days-99 years)
Hospitalisation/outpatient	no. of isolates (%)
Hospitalisation	863 (93.6)
Outpatient	59 (6.4)
	mean (range)
Hospitalisation length	27 days (0–724 days)
Hospitalisation length to collection	11 days (0–369 days)
Source	no. of isolates (%)
Urine	269 (29.2)
Rectum	226 (24.5)
Blood culture	96 (10.4)
Oral cavity swab	84 (9.1)
Throat swab	49 (5.3)
Sputum	43 (4.7)
Perianal swab	39 (4.2)
Wound swab	30 (3.3)
Urinary catheter	24 (2.6)
Skin swab	13 (1.4)
Handprint	11 (1.2)
Venous catheter	10 (1.1)
Nose swab	7 (0.8)
Stool	6 (0.7)
Bronchoalveolar lavage fluid	4 (0.4)
Vagina	3 (0.3)
Ascites	2 (0.2)
Suction catheter	2 (0.2)
Ear swab	1 (0.1)
Eye swab	1 (0.1)
Stoma swab	1 (0.1)
Urethra swab	1 (0.1)

DNA isolation

For mini-MLST, bacterial genomic DNA (gDNA) was isolated using Chelex 100 Resin (Bio-Rad, USA). The Bacterial culture was homogenised in 100 μ of 5% w/v Chelex 100 Resin with a vortex. The suspension was incubated for 10 min at 100°C and then centrifuged for 2 min at 15,500 rcf. A supernatant containing gDNA was transferred into a clean microtube. For WGS, gDNA was purified using DNeasy Blood & Tissue Kit (Qiagen GmbH). The DNA concentration was measured using NanoDrop (Thermo Scientific, USA).

Mini-MLST

Mini-MLST was performed with primers described by Andersson, Tong [6] on a RotorGene 6000 platform (Corbett Research, Australia). The 20 μL reaction volume contained 10 μL 2× SensiFAST HRM mix (Bioline Reagents, UK), 0.4 μM of each primer, 1 μL of extracted genomic DNA (30 ng) and deionized water to a final volume of 20 μL. Thermo cycling parameters were: 95°C for 3 min, 40 cycles of 95°C for 5 s, 65°C for 10 s and 72°C for 20 s, then one cycle of 95°C for 2 min and 50°C for 20 s, followed by HRM ramping from 70 to 95°C, increasing by 0.2°C at each step. The results were interpreted using the current version of our conversion key, which is available for free download at http://www.cmbgt.cz/mini-mlst/t6353.

Whole genome sequencing and analysis

Purified gDNA was shredded using S220 Focused-ultrasonicator (Covaris, USA). WGS libraries were prepared with KAPA HyperPrep Kits (Roche, Switzerland) and a quality check was performed using 2100 Bioanalyzer (Agilent Technologies, USA). The Illumina MiSeq platform was used for WGS and 250-bp paired-end sequencing was performed. The raw sequencing data were deposited in the NCBI BioProject database under the project ID PRJNA515630 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA515630). The obtained reads were quality checked using FastQC (Babraham Bioinformatics, UK) and assembled using Burrows-Wheeler Aligner [15]. Ridom SeqSphere+ (Ridom, DE) seed genome Klebsiella pneumoniae subsp. pneumoniae NTUH-K2044 (NC_012731.1) was used as the reference genome. To remove unmapped reads, reads with poor quality and duplicates, SAMtools were used [16]. After reference mapping, all positions with less than 10× coverage and all ambiguous positions (less common base represented at least 10% of bases in the target position) were removed from further analysis. UGENE software was used to obtain consensus sequences [17]. Resistome and virulome analysis was carried out using public online databases (http://www.genomicepidemiology.org/, http://bigsdb.pasteur.fr/). The cgMLST was performed using Ridom SeqSphere+ (Ridom, DE) with an incorporated KPN cgMLST scheme. This included 2,358 target genes whose alleles were used to generate a cgMLST dendogram. For the core genome SNV minimum spanning tree 2,042,000 aligned nucleotide sites were analysed. In total, 4,179,387 aligned nucleotide sites were accounted for by the core and accessory genome SNV minimum spanning tree analysis. The WGS data were used to determine the MLST type of all sequenced strains.

Results

For hospital population characterisation, we used mini-MLST to type all collected isolates. Overall, 38 different MelTs were identified among 922 ESBLp KPN isolates collected between 7/2016 and 5/2018. MelT145 (30.8%, n = 284), MelT26 (25.9%, n = 239), MelT132 (8.0%, n = 74), MelT139 (7.7%, n = 71), MelT269 (5.4%, n = 50), MelT281 (5.0%, n = 46), MelT266 (2.6%, n = 24) and MelT20 (2.3%, n = 21) were the most predominant (Fig 1). The remaining 30 MelTs were present in less than 2% of all isolates. Fig 1 shows the proportions of certain MelTs in our hospital population are quite stable (MelT145, MelT26), while an increase (MelT132) or decrease (MelT139) can be observed in others. Mini-MLST’s discriminatory power was D = 0.8189. The mini-MLST results were used for basic population sorting and WGS sample selection.

Fig 1

Comparison of MelTs distribution in monitored period (7/2016-5/2018).

Whole genome sequence analysis

Whole genome sequence sample selection

We randomly selected 14 isolates from the four most predominant MelTs from 7/2016-5/2018 (3 MelT145 isolates, 5 MelT26 isolates, 3 MelT132 isolates and 3 MelT139 isolates). Because these MelTs were also present in our hospital during our previous study [14], we added 10 isolates isolated in our hospital in 2014 (3 MelT145 isolates, 1 MelT26 isolate, 3 MelT132 isolates and 3 MelT139 isolates).

To set SNV relatedness cut-off values, 22 isolates representing colonization (rectal swab)/infection (blood culture) pair samples obtained from 11 different patients were analysed using WGS. Pair sample isolates belonged to 6 different MelTs (2 MelT26 isolates, 2 MelT130 isolates, 6 MelT132 isolates, 6 MelT145 isolates, 4 MelT266 isolates and 2 MelT281 isolates). For all 46 sequenced isolates, we performed in silico MLST, resistome and virulome analysis, cgMLST and SNV analysis of the core genome and accessory genome.

The MLST, resistome and virulome analysis

First, in silico STs were determined and resistome and virulome were compared for all 46 samples (Fig 2). MLST analysis (based on seven house-keeping genes rpoB, gapA, mhd, pgi, phoE, infB and tonB) showed ST uniformity within individual MelT groups, except isolate S13 from MelT26 which, unlike the other isolates in this MelT group, was ST29. This isolate was the only MelT26 detected in 2014. All other sequenced MelT26 isolates from 2016 and 2017 were ST1271. ST1271 and ST29 only differ in the MLST scheme’s phoE allele (allele phoE 4 in ST1271 against allele phoE 6 in ST26) (S1 Table).

Fig 2

WGS results of 46 selected ESBLp KPN isolates.

The cgMLST dendogram was made using Ridom Seqsphere+ software and is based on sequence similarity in the 2,251 core genome targets shared by all 46 isolates. Mini-MLST was done experimentally and MLST was done in silico using WGS data. Mini-MLST, MLST, cgMLST and SNV analysis of pair samples (blood culture/rectal swab pairs are highlighted in the same colour). The time lapse indicates time between collecting the colonizing and infecting isolates. * Samples obtained from one patient, but with no rectal swab.

Comparing resistome and virulome correlated with the MLST analysis results and confirmed a high level of similarity inside the MelTs (see detailed results in S1 Table). Isolates clustered together, as in the in silico MLST analysis (including separation of isolate S13 from other isolates) with one exception in MelT132, which was divided into two lineages, A and B, according to the different resistome and virulome composition. Lineage A included S19, S20, S21 and S44; lineage B included S22, S24, S25, S26, S35 and S36.

cgMLST and SNV analysis

CgMLST analysis provides more discriminatory power than traditional MLST. In this study, we used a cgMLST scheme consisting of 2,358 targets processed by Ridom SeqSphere+ software (Ridom, DE). A cgMLST dendrogram based on 2,251 targets present in all analysed strains was constructed (Fig 2). The other 107 targets from the original 2,358 targets were absent in some or all of the strains and were excluded from all further analyses. The cgMLST divided 46 isolates into 7 clearly distinguishable clusters corresponding to their MelTs groups, with two exceptions. The S13 isolate differed from the other MelT26 isolates in 584 targets, while the average number of different alleles within the other MelTs ranged from 0 to 19 differences between two samples of the same MelT. The second exception was MelT132, which was divided into two subpopulations with 127 different alleles between them. Both exceptions correlated with the previous in silico MLST and resistome and virulome analysis. The average number of different alleles between individual MelTs was 1827 (range from 1809 to 1844 differences).

SNV analysis was carried out for both a core genome (n = 2,251 genes present in all analysed strains) and core with accessory genome (n = 4,084 genes present in all analysed strains). In the core genome, 29,535 SNV positions were identified in total. In general, the core genome SNV number inside each MelT varies from 0 to 23 SNV (MelT145 0–23 SNV, n = 12; MelT139 0–21 SNV, n = 8; MelT26 0–20 SNV, n = 8; MelT266 0–14 SNV, n = 4; MelT132 cluster A 0–22 SNV, n = 5; MelT132 cluster B 0–4 SNV, n = 6; MelT281 0 SNV, n = 2 and MelT130 0 SNV, n = 2) (Fig 3A). The S13 isolate differed by 2,311 SNV from other MelT26 isolates, while between individual MelTs, the number of SNV ranged from 9,595 to 9,821. The MelT132 isolates were divided into two lineages, A and B, similar to the cgMLST analysis. Both MelT132 lineages differed with 137 SNV between them. The complete distance matrix for the core genome SNV analysis is showed in S2 Table.

Fig 3

Core genome SNV (A) and core with accessory genome SNV (B) analysis of 46 selected ESBLp KPN isolates. SNV analysis was performed on sequences of 2,251 core genome targets and 4,084 core and accessory genome targets respectively. We found 29, 535 SNV positions in the core genome targets (from a total of 2,042,000 positions) and 61,343 SNV positions in the core and accessory targets (from a total of 4,179,387 positions).

The core and accessory genome SNV analysis was done with 4,084 gene targets present in all 48 isolates (Fig 3B). 61,343 SNV positions were identified with the SNV count within each MelT ranging from 0 to 58 (MelT145 0–58 SNV, n = 12; MelT139 0–53 SNV, n = 8; MelT26 3–32 SNV, n = 8; MelT266 2–50 SNV, n = 4; MelT132 lineage A 0–36 SNV, n = 5; MelT132 lineage B 1–6 SNV, n = 6; MelT281 1 SNV, n = 2 and MelT130 1 SNV, n = 2). The average SNV number between individual MelTs was 20,196 SNV (range from 19,936 to 20,358 SNV). The distance matrix for the core and accessory genome’s SNV analysis is showed in S3 Table. The major topology difference between the core (A) and the core and accessory (B) genome panels was the position exchange between the MelT281 and the MelT132 clusters. Additionally, using the core and accessory genome panel (B), the MelT26 and the MelT132 clusters were linked with the S13 isolate that was only linked to MelT26 cluster when using the core genome panel (A).

Infection and colonisation pair isolates analysis

We analysed 11 pairs (each pair included a rectal swab and a blood culture) of ESBLp KPN isolates obtained from 11 patients to set the SNV cut-off, which determined the relatedness of isolates (Fig 2). From 22 ESBLp KPN isolates, 6 distinct STs were identified: ST405 (n = 6), ST433 (n = 6), ST323 (n = 4), ST1271 (n = 2), ST458 (n = 2) and ST23 (n = 2). The isolates from each patient share the same ST within the pair. The core genome’s SNV number within each pair varies in range from 0 to 22 SNV, while together with the accessory genome’s SNV number, varies in range from 0 to 40 SNV. Based on paired sample analysis, we established relatedness cut-off values for our ESBLp KPN population as the highest SNV number within pair isolates, 22 SNV for the core genome and 40 SNV for the core with the accessory genome. Both the highest SNV values were for pair S39/S40 that had time lapses between samples of 47 days. The next highest SNV values were only 3 SNVs for the core genome and 5 SNVs for the core and accessory genome, despite time lapses of up to 90 days."

Discussion

We are currently using the following protocol in routine practice. ESBLp KPN collected from high-risk departments are prospectively tested with mini-MLST to determine the MelT. When the strains MelT differ, transmission is unlikely. When we observe an increased incidence of one MelT, we investigate the potential epidemiological linkages and then we decide if there is a possible outbreak and need for WGS analysis. Meanwhile, early epidemiological measures can be implemented to prevent further spread of infection.

KPN mini-MLST is a cheap, rapid and robust method for epidemiological strain typing introduced by Andersson, Tong [6], with advantages in its robustness, reproducibility and portability between laboratories as it is based on a well-established MLST method. In studies with a large number of samples, mini-MLST also can be used for sample sorting and pre-selecting for more detailed analysis. We evaluated this method for prospective ESBLp KPN population screening and to solve healthcare associated infection outbreaks in our previous study [14]. During the mini-MLST results evaluation, we were unable to determine the MelT of several isolates as their mini-MLST allele combination did not corresponded to any MelT in the published conversion key. On the date of the original article publication there were 863 STs, while in 7/2019 there were 4126 STs, from which it was not possible to determine some of the newly described STs’ MelT according to the original conversion key. Following this, we constructed a new automatic algorithm (applicable to already existing and future-designed mini-MLST schemes) and created a new conversion key that will be regularly updated. The latest version of the key is available on http://www.cmbgt.cz/mini-mlst/t6353.

Based on our previous study, we started prospective screening in certain departments of our hospital in 7/2016. Since then, we have observed six MelTs account for 82.8% of isolates, from a total of 38 MelTs identified between 7/2016 and 5/2018 (Fig 1). The most pronounced changes in proportional representation were observed in MelT145 (from 38.1% in 2016 to 29.5% in 2018), MelT139 (from 10.9% in 2016 to 3.6% in 2018), MelT132 (from 2.5% in 2016 to 4.7% in 2018) and MelT266 (from 1.7% in 2016 to 5.2% in 2018). We did not detect any emerging MelT which spread to more than 1% of the study population. This suggests a relatively stable low diversity bacterial population which cannot be divided more by using the mini-MLST method only and needs methods with more discriminatory power.

WGS has become a powerful method for most epidemiological studies and hospital outbreak investigations, as it provides multiple analyses from a single technique, including data on MLST, resistant genes, virulence genes, plasmids and genome comparison [18]. Currently, epidemiological investigations have mainly focused on an allele-based approach (cgMLST, wgMLST) and SNV analysis. While the allele-based approach is particularly suitable for multicentre studies and clustering large populations of bacteria [19], SNV analysis is especially useful when dealing with outbreak episodes where we are expecting the analysed strains to be similar. The number of differences in alleles or SNV that still identify isolates as similar or related and include them in outbreak episodes differs in various publications [11]. According to our best knowledge, there are no general standards or rules to set a sufficient threshold value. For this reason, it is not possible to simply compare the results of individual studies and freely use previously published cut-off values, the cut-off being typically determined based on the specific results from those publications and mostly varying between 0 and several dozen SNV [10]. Also, comparing outbreak strains and the local bacterial population is not commonly done in most studies, which may give important information to determine relatedness and assess outbreaks.

In order to evaluate the role of WGS in low diversity population outbreak analysis, we sequenced 46 ESBLp KPN isolates belonging to the four mostly spread MelTs from the monitored period from 7/2016 to 5/2018, isolates from the same MelTs originally collected for our mini-MLST pilot study in 2014 [14] and colonization/infection pair samples. First, we performed in silico MLST to correlate the STs with MelTs and characterise our population in the worldwide KPN population context. All predominant STs in our population have previously been described in literature. Serotype K1 strains belonging to ST23 (MelT281) were described as a frequent cause of invasive infections and liver abscesses [20]. ST321 (MelT139), ST323 (MelT266), ST1271 (MelT26) and ST29 (MelT26) were described as long-term, persistent MDR strains in hospital facilities [21-23]. ST405 (MelT132) was often described as an OXA-48 producer [24] and ST433 (MelT145) as hyper virulent biofilm producing strains [25]. The only discrepancy between MLST and mini-MLST was in sample 13, which belonged to ST29, in contrast with the other MelT26 isolates, which belonged to ST1271. S13 was the only MelT26 isolate collected in 2014.

Second, we used public online databases to determine the profile of resistance genes, virulence factor genes and efflux pump genes (S1 Table). Compared to the mini-MLST results, resistome, virulome and efflux pump gene analysis only separated S13 from others belonging to MelT26 (as well as MLST) and divided MelT132 into two clearly separate clusters (Isolates from clusters A and B from MelT132 both belonged to ST405). In summary, these analyses do not have sufficient discriminatory power for the epidemiology of low diversity hospital bacterial populations. However, they provide information on the resistance and virulence of the examined strains, which can help manage the spread of hyper-resistant and hyper-virulent strains and may be useful in patients’ treatment.

The cgMLST provides more discriminatory power than previously mentioned analyses, since it is based on hundreds or thousands KPN genes’ allelic similarities depending on the study’s design [26-28]. Despite analysing 2,251 genes, we did not find enough allelic differences between isolates belonging to the same MelT to draw a conclusion about isolate relatedness.

To analyse the SNV analysis results, we have to set the SNV cut-off values to determine ESBLp KPN isolate similarity. We used the general premise that the colonization strain is in most cases the cause of the subsequent infection [26, 29, 30]. Therefore, the number of SNV between colonization/infection pair isolates defines the value of the difference level at which the isolates can still be considered similar. Based on our results, we established two sets of cut-off values. The first set was based on all pairs and cut-off values were 22 SNV for the core genome and 40 SNV for the core and accessory genome (Fig 2). The stricter cut-offs were set with excluded SNV values for pair S39/S40 and values were 3 SNV for the core genome and 5 SNV for the core and accessory genome. Both cut-off values set reflected the low diversity in the highly selective ESBLp KPN hospital population. Using both of our cut-off values on our population, even isolates without evident epidemiological associations were clustered together, which makes evaluating outbreaks difficult and may lead to erroneous conclusions.

We are aware that our study is a single-centre study with a small number of isolates sequenced. Especially to determine cut-off, more paired isolates should be analysed. However, due to the specific hospital’s environment, we can expect highly-selected bacterial populations to also be found in other hospitals.

To our best knowledge, this is the first study where prospective molecular typing is combined with WGS to define the epidemiological background and the genetic structure of the hospital’s bacterial population. We proved that mini-MLST is a cost effective means of ruling out epidemiological linkage, but only complete genome analysis can provide strong evidence in favour of epidemiological linkage. Our findings showed there were only minimal differences within the core/accessory genome in the low diversity hospital population and gene based SNV analysis does not have enough discriminatory power to differentiate isolates’ relatedness or evaluate whether it is an outbreak or not. Thus, intergenic regions and mobile elements should be incorporated to the analysis scheme to increase the discriminatory power. Therefore, when evaluating any molecular biological data, it is necessary to analyse them to concord with the epidemiological background.

Resistome, virulome and efflux genes profiles of 46 ESBLp KPN isolates.

(XLSX)

Click here for additional data file.

Core genome SNV distance matrix.

(XLSX)

Click here for additional data file.

Core and accessory genome SNV distance matrix.

(XLSX)

Click here for additional data file.

3 Jul 2019

PONE-D-19-14966

Application of whole genome sequencing in low diversity hospital extended-spectrum beta-lactamases producing Klebsiella pneumoniae population

PLOS ONE

Dear Dr. Lengerova,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that carefully addresses each of the points raised during the review process.

We would appreciate receiving your revised manuscript by Aug 17 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

D. Ashley Robinson, 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

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Our editorial staff has assessed your submission, and we have concerns about the grammar, usage, and overall readability of the manuscript.  We therefore request that you revise the text to fix the grammatical errors and improve the overall readability of the text before we send it for review. We suggest you have a fluent, preferably native, English-language speaker thoroughly copyedit your manuscript for language usage, spelling, and grammar.

If you do not know anyone who can do this, you may wish to consider employing a professional scientific editing service.

Whilst you may use any professional scientific editing service of your choice, PLOS has partnered with both American Journal Experts (AJE) and Editage to provide discounted services to PLOS authors. Both organizations have experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. To take advantage of our partnership with AJE, visit the AJE website (http://learn.aje.com/plos/) and enter referral code PLOS15 for a 15% discount off AJE services. To take advantage of our partnership with Editage, visit the Editage website (www.editage.com) and enter referral code PLOSEDIT for a 15% discount off Editage services. If the PLOS editorial team finds any language issues in text that either AJE or Editage has edited, the service provider will re-edit the text for free.

Upon resubmission, please provide the following:

•    The name of the colleague or the details of the professional service that edited your manuscript

•    A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

•    A clean copy of the edited manuscript (uploaded as the new *manuscript* file)

Please note that PLOS ONE does not copyedit accepted manuscripts and that one of our criteria for publication is that articles must be presented in an intelligible fashion and written in clear, correct, and unambiguous English (http://www.plosone.org/static/publication#language). If the language is not sufficiently improved, we may have no choice but to reject the manuscript without review.

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: Yes

Reviewer #2: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: N/A

**********

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: No

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: Yes

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: This manuscript is particularly well written and describes an interesting and well-designed study. My comments are quite minor.

• Not sure where the sequence data/sequence reads are deposited

• The figure legends are short and uninformative. I know they should not repeat was in the text, but I was having trouble understanding exactly how each figure was put together. Perhaps just a little more information is necessary. One thing that was specifically unclear was whether the dendrogram in Fig 3 was on the basis of allele number similarity, or overall sequence similarity. Similarly, it appears from the Methods that allele numbers were used to create the minimum spanning trees in Fig 4 – which seems odd given that the Results implies it is from actual sequence similarity. Can this be clarified?

• Can the authors give some thought to shortening the Discussion? The main points conclusions could possibly be stated more clearly and succinctly e.g. “miniMLST is a cost effective means of ruling out epidemiological linkage, but only complete genome analysis can provide strong evidence in favour of epidemiological linkage”.

• Line 61: should be “bacterial”

Reviewer #2: Title: Application of whole genome sequencing in low diversity hospital extended-spectrum beta-lactamases producing Klebsiella pneumoniae population

This manuscript presents a well written description of the application of mini-MLST in the surveillance of ESBL-KPN within a single hospital in the Czech Republic. Additionally, the authors discuss the limitations of mini-MLST in resolving closely related isolates, in particular outbreak strains, and the need for WGS in these situations. The authors then show the potential application of WGS in a limited sample set.

Minor comments:

Line 52 – the word “solution” should likely be “intervention”.

Line 237-238 – The authors indicate they have generated a new automatic algorithm that acts as a conversion key. Is it possible to utilize this conversion key with the existing MLST schema to generate a conversion list of MSLT to melT groups? This would allow for direct comparison of the large existing studies that use MLST with the use of melT groups and will also clarify the resolution of melT groups. I think this would help clarify the potential role of melT groups among a growing list of molecular epidemiology markers of bacterial diversity (e.g. PFGE, MLST, mini-MLST, cgMLST, WGS).

Line 63-65 – The authors define CgMLST and SNV analyses, but the descriptions are confusing. In particular, the authors indicate SNV analysis provides a higher resolution power as compared to cgMLST, but then use SNV as a precursor to CgMLST. It would be good to clarify that both are based upon the detection of variants, but while CgMLST is looking at variants in a specific pre-determined set of genes, true whole genome alignment allows for assessment of variants in all regions that are not masked.

Table 1.- Can the authors please include the percentage of each subgrouping with the count? This will help the reader understand the proportion of isolates, especially within the source section of the table.

Lines 95-97 – the methods listed for use with mini-MLST differ from the methods used in the previous reference within Diagnostic Microbiology and Infectious Disease (citation 14). Can the authors comment on why the methods were changed from the UltraClean Microbial DNA Isolation Kit to the Chelex based extraction?

Line 122 – Can the authors indicate how they defined ambiguous positions? Was this based upon a percent threshold for variant detection?

Line 235 – please update the number of ST types as it now appears there are 4,054 different MSLT profiles within the Pasteur MLST database.

Line 249 – Please update the sentence to say “WGS has become a powerful method”

Major comments:

Title – As this manuscript focuses heavily on mini-MLST, please add mini-MLST to the title.

Lines 35-37, 57-59, 318-323 – The authors describe the potential for mini-MLST in real-time epidemiology, but point out the need for mini-MLST to be combined with more discriminatory analyses in outbreak/low genetic diversity settings. Can the authors describe in more detail how this progressive application of mini-MLST and WGS would work in active surveillance? In specific, it would be good to identify how mini-MLST could be used to select isolates for WGS and how mini-MLST would be able to identify outbreaks to focus on.

Lines 146-156 – Do the authors intend to submit the WGS sequencing data to a repository such as the European Nucleotide Archive? If so, can the authors include the project ID within the manuscript?

Line 215-217 – What was the order of collection for the colonization and infection strains? Were colonization strains always collected before outbreak strains? This may be good information to include in a supplemental table. Additionally, can the authors discuss the number of remaining positions within the genome that are accounted for in calculating the number of SNVs between each pair of isolates?

Line 271-275 – Can the authors make it more clear if this discussion is referring to the 14 isolates described on lines 146-147, the 24 isolates described from lines 146-150, or does this refer to the entire set of 46 isolates (line 155)? If this is either the 14 or 24 isolates, this represents a small subset of the overall 922 isolates. Can the authors discuss the possibility that this is not a broad enough subsample of the total set of ESBL-KPN isolates to truly evaluate the role of WGS in a low diversity population?

Line 301-310 – I am confused about the application of the premise that “the colonization strain is the cause of the subsequent infection in most cases” to the idea of evaluating a hospital outbreak. In particular, hospital outbreaks are characterized by person-to-person transmission, or are mediated by fomites in the environment. Can the authors describe the potential for misclassification of colonizing and infecting strains if the infecting strain was acquired within the hospital environment?

In a time where there is more focus on acquired resistance genes, and in particular how these acquired genes can change the required antimicrobial treatment choices, how can a technique like mini-MLST be used to help target hospital infection control in the case of an outbreak?

Figure 2 – This figure is very data dense, but does not display well on a single printed page. Additionally, as the discussion of virulence factors is limited within the manuscript, it may be good to reformat this figure or move it to the supplemental files.

**********

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: No

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 to be viewed.]

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 us at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Jul 2019

Journal Requirements

• Language corrections

Our manuscript was reviewed by a native English-speaker. His name is Matthew Smith at Brno Editing, Proofreading & Translation Services (www.translationandproofreadingservices.com) and his contact e-mail is matsmithj@yahoo.com.

Editor Comments to Author

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

Line 119-121 - Data accession for our project was updated in the manuscript. All raw sequencing data are available under the accession number PRJNA515630 in the BioProject database (direct link was also added to the revised manuscript https://www.ncbi.nlm.nih.gov/bioproject/PRJNA515630.)

Reviewers' Comments to Author

Reviewer: 1

• Not sure where the sequence data/sequence reads are deposited

Line 119-121 - Data accession for our project was updated in the manuscript. All raw sequencing data are available under the accession number PRJNA515630 in the BioProject database (direct link was also added to the revised manuscript https://www.ncbi.nlm.nih.gov/bioproject/PRJNA515630.)

• The figure legends are short and uninformative. I know they should not repeat was in the text, but I was having trouble understanding exactly how each figure was put together. Perhaps just a little more information is necessary. One thing that was specifically unclear was whether the dendrogram in Fig 3 was on the basis of allele number similarity, or overall sequence similarity. Similarly, it appears from the Methods that allele numbers were used to create the minimum spanning trees in Fig 4 – which seems odd given that the Results implies it is from actual sequence similarity. Can this be clarified?

Line 193-199; Line 213-217 - We added short legends to Fig 2 and Fig 3 (formerly Fig 3 and Fig 4). We also clarified allele/sequence similarity in the Methods and Results sections. We transferred Fig 2 to the Table and moved it to the supplemental files (S1 Table).

• Can the authors give some thought to shortening the Discussion? The main point’s conclusions could possibly be stated more clearly and succinctly e.g. “miniMLST is a cost effective means of ruling out epidemiological linkage, but only complete genome analysis can provide strong evidence in favour of epidemiological linkage”.

Line 241-341 - We rewrote the discussion to highlight the main conclusions

• Line 61: should be “bacterial”

Line 62 - Text corrected

Reviewer: 2

Minor comments :

• Line 52 – the word “solution” should likely be “intervention”.

Line 53 - Text corrected

• Line 237-238 – The authors indicate they have generated a new automatic algorithm that acts as a conversion key. Is it possible to utilize this conversion key with the existing MLST schema to generate a conversion list of MSLT to melT groups? This would allow for direct comparison of the large existing studies that use MLST with the use of melT groups and will also clarify the resolution of melT groups. I think this would help clarify the potential role of melT groups among a growing list of molecular epidemiology markers of bacterial diversity (e.g. PFGE, MLST, mini-MLST, cgMLST, WGS).

Line 259-260 - Once you have designed the mini-MLST scheme the algorithm is applicable to any MLST scheme. To run the algorithm, you need a list of ST including allele variants; FASTA file of all genes used in mini-MLST scheme and mini-MLST primers sequences. Added to the manuscript.

• Line 63-65 – The authors define CgMLST and SNV analyses, but the descriptions are confusing. In particular, the authors indicate SNV analysis provides a higher resolution power as compared to cgMLST, but then use SNV as a precursor to CgMLST. It would be good to clarify that both are based upon the detection of variants, but while CgMLST is looking at variants in a specific pre-determined set of genes, true whole genome alignment allows for assessment of variants in all regions that are not masked.

Line 64-71 - We rewrote the definition of cgMLST and SNV analysis to clarify the differences between those two approaches.

• Table 1.- Can the authors please include the percentage of each subgrouping with the count? This will help the reader understand the proportion of isolates, especially within the source section of the table.

Table 1, Line 96 - Subgroups percentages were added to the Table 1.

• Lines 95-97 – the methods listed for use with mini-MLST differ from the methods used in the previous reference within Diagnostic Microbiology and Infectious Disease (citation 14). Can the authors comment on why the methods were changed from the UltraClean Microbial DNA Isolation Kit to the Chelex based extraction?

As we process a large number of isolates we were looking for a more efficient DNA extraction method. Chelex based extraction costs $ 0.5 per sample and takes less than 30 minutes with minimum hands-on time. UltraClean Microbial DNA Isolation Kit costs $ 5 per sample, takes about 60 minutes and requires more hands-on time. The quality and concentration of DNA extracted with Chelex is suitable for use in PCR methods.

Line 122 – Can the authors indicate how they defined ambiguous positions? Was this based upon a percent threshold for variant detection? Line 127-128 - The ambiguous base was added to a position when there were at least two different bases and the less common base represented at least 10 % of bases in the target position. We put a note in the manuscript.

Line 235 – please update the number of ST types as it now appears there are 4,054 different MSLT profiles within the Pasteur MLST database. Line 257 - The number of MLST profiles was updated in manuscript on 9/7/2019.

Line 249 – Please update the sentence to say “WGS has become a powerful method” Line 271 - Text corrected

Major comments :

Title – As this manuscript focuses heavily on mini-MLST, please add mini-MLST to the title. Title, line 1 - Title updated to “Application of mini-MLST and whole genome sequencing in low diversity hospital extended-spectrum beta-lactamases producing Klebsiella pneumoniae population”

Lines 35-37, 57-59, 318-323 – The authors describe the potential for mini-MLST in real-time epidemiology, but point out the need for mini-MLST to be combined with more discriminatory analyses in outbreak/low genetic diversity settings. Can the authors describe in more detail how this progressive application of mini-MLST and WGS would work in active surveillance? In specific, it would be good to identify how mini-MLST could be used to select isolates for WGS and how mini-MLST would be able to identify outbreaks to focus on. Line 242-247 - The following paragraph was added to the discussion.

We are currently using the following protocol in routine practice. ESBLp KPN collected from high-risk departments are prospectively tested with mini-MLST to determine MelT. When the strains’ MelT differ, transmission is unlikely. When we observe an increased incidence of one MelT, we check the potential epidemiological linkages and then we decide if there is a possible outbreak and need of WGS analysis. Meanwhile, early epidemiological measures can be implemented to prevent further spread of infection.

Lines 146-156 – Do the authors intend to submit the WGS sequencing data to a repository such as the European Nucleotide Archive? If so, can the authors include the project ID within the manuscript? Line 119-121 - Data accession for our project was updated in the manuscript. All raw sequencing data are available under the accession number PRJNA515630 in the BioProject database (direct link was also added to the revised manuscript https://www.ncbi.nlm.nih.gov/bioproject/PRJNA515630.)

• Line 215-217 – What was the order of collection for the colonization and infection strains? Were colonization strains always collected before outbreak strains? This may be good information to include in a supplemental table. Additionally, can the authors discuss the number of remaining positions within the genome that are accounted for in calculating the number of SNVs between each pair of isolates? Fig 2, Line 193 - Collection dates are listed in Fig 2 and also in S1 Table. In one case the rectum isolate was collected three days before the blood culture isolate and in one case the rectum isolate was collected sixteen days before the blood culture isolate. All other colonization isolates were collected before blood culture isolates.

In total 4,179,387 positions were used in SNV analysis. For core genome SNV analysis only, 2,042,000 positions were analysed. We added S2 and S3 Tables to supplementary files, which show SNV distance matrix for all 46 samples for core genome SNV and core and accessory genome SNV respectively.

• Line 271-275 – Can the authors make it more clear if this discussion is referring to the 14 isolates described on lines 146-147, the 24 isolates described from lines 146-150, or does this refer to the entire set of 46 isolates (line 155)? If this is either the 14 or 24 isolates, this represents a small subset of the overall 922 isolates. Can the authors discuss the possibility that this is not a broad enough subsample of the total set of ESBL-KPN isolates to truly evaluate the role of WGS in a low diversity population?

Line 287-290 - This part of the discussion is referring to all 46 sequenced isolates. We clarified this in the relevant part of the discussion.

Our further project will focus on more detailed analysis of the most predominant MelTs, including sequencing of a larger isolate set, long reads sequencing and also plasmid analysis.

• Line 301-310 – I am confused about the application of the premise that “the colonization strain is the cause of the subsequent infection in most cases” to the idea of evaluating a hospital outbreak. In particular, hospital outbreaks are characterized by person-to-person transmission, or are mediated by fomites in the environment. Can the authors describe the potential for misclassification of colonizing and infecting strains if the infecting strain was acquired within the hospital environment?

Line 314-318 - The number of SNV between colonization/infection pair isolates defines the value of the difference level at which the isolates can still be considered.

We clarified this in the relevant part of the discussion.

In our study (and hospital), the infecting strain is collected from blood culture, urine or other primary sterile material. If the patient has an infection with his own colonizing strain, it is possible to tell only when we also have an isolate from this patient from a previous collection.

• In a time where there is more focus on acquired resistance genes, and in particular how these acquired genes can change the required antimicrobial treatment choices, how can a technique like mini-MLST be used to help target hospital infection control in the case of an outbreak?

Using knowledge of the local population, its stability and epidemiological linkages along with rapid prospective screening can help implement initial epidemiological measures or find and eliminate a potential outbreak source. As such, the Mini-MLST does not provide information on antibiotic resistance genes. However, information on the resistant genes’ presence in certain MelT acquired by the WGS along with the fact that the hospital population seems to be relatively stable over the long term can be used to predict the resistome of the examined strains. However, to confirm, the strains resistome needs to be examined by specific PCR or WGS.

• Figure 2 – This figure is very data dense, but does not display well on a single printed page. Additionally, as the discussion of virulence factors is limited within the manuscript, it may be good to reformat this figure or move it to the supplemental files.

S1 Table - We changed the format to Table and move it to the supplemental files

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

22 Jul 2019

PONE-D-19-14966R1

Application of mini-MLST and whole genome sequencing in low diversity hospital extended-spectrum beta-lactamases producing Klebsiella pneumoniae population

PLOS ONE

Dear Dr. Lengerova,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. The Academic Editor believes that the manuscript's remaining issues with English grammar and use can be addressed by the authors. Therefore, we invite you to submit a revised version of the manuscript that addresses the Editors comments below.

We would appreciate receiving your revised manuscript by Sep 05 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

D. Ashley Robinson, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

Title

*should read "beta-lactamase producing", not plural "beta-lactamases producing"

Abstract

*lines 32, 35, 39, referring to the analysis of core and accessory genes as "core/accessory" is ambiguous in the abstract, please state as "core and accessory" as done elsewhere in the text

Intro

*first 2 sentences are very poorly constructed, please revise

*line 65, suggested wording "of a pre-determined set of.."

Methods

*lines 133, 134 suggested wording of "aligned nucleotide sites" instead of ambiguous term "positions"

Results

*where is Fig 1? I did not see it in the files attached to this paper

*line 175 previous wording is better, suggested wording "(including separation of isolate S13 from other isolates)"

*lines 187 and 190 "average number of different alleles" is more precise language, please revise as appropriate

*Fig3 shows a major topology difference between panels A (core) and B (core and accessory) in the switching of the position of the MelT281, ST23 isolates and MelT132, ST405 isolates. This should be mentioned.

*line 234 suggested wording "Based on paired sample analysis,..."

*lines 237 to 240 wording is unclear. Do the authors mean "Both the highest SNV values were for pair S39/S40 that had time lapses between samples of 47 days. The next highest SNV values were only 3 SNVs for the core and 5 SNVs for the core and accessory, despite time lapses of up to 90 days." Please revise as appropriate.

Discussion

*line 263 suggested wording "Since then, we have observed six MelTs that account for 82.8% of isolates, from a total of 38 MelTs..."

*line 276 instead of "big bacterial populations", suggested wording "large populations of bacteria"

Overall

*Fig2 legend, Same issue as raised by Reviewer1, is the tree based on number of shared alleles or the sequence similarity in the shared alleles? Mostly likely, this legend should include "is based on sharing of 2,251 core genome targets." Please revise as appropriate.

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

[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 to be viewed.]

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 us at figures@plos.org. Please note that Supporting Information files do not need this step.

30 Jul 2019

Title

• Should read "beta-lactamase producing", not plural "beta-lactamases producing"

o Line 1 - Title corrected

Abstract

• Lines 32, 35, 39, referring to the analysis of core and accessory genes as "core/accessory" is ambiguous in the abstract, please state as "core and accessory" as done elsewhere in the text

o Lines 32, 35 and 39 - Text corrected

Introduction

• First 2 sentences are very poorly constructed, please revise

o Line 44-49 – Text corrected

• Line 65, suggested wording "of a pre-determined set of.."

o Line 65 – Text corrected

Methods

• Lines 133, 134 suggested wording of "aligned nucleotide sites" instead of ambiguous term "positions"

o Line 133 and 134 – Text corrected

Results

• Where is Fig 1? I did not see it in the files attached to this paper

o Fig 1 is uploaded in the attachment files

• Line 175 previous wording is better, suggested wording "(including separation of isolate S13 from other isolates)"

o Line 175-176 – Text corrected

• Lines 187 and 190 "average number of different alleles" is more precise language, please revise as appropriate

o Lines 187 and 190 – Text corrected

• Fig3 shows a major topology difference between panels A (core) and B (core and accessory) in the switching of the position of the MelT281, ST23 isolates and MelT132, ST405 isolates. This should be mentioned.

o Lines 229-233 – This information was added to the manuscript

• Line 234 suggested wording "Based on paired sample analysis,..."

o Line 239 – Text corrected

• Lines 237 to 240 wording is unclear. Do the authors mean "Both the highest SNV values were for pair S39/S40 that had time lapses between samples of 47 days. The next highest SNV values were only 3 SNVs for the core and 5 SNVs for the core and accessory, despite time lapses of up to 90 days." Please revise as appropriate.

o Lines 242 to 244 – Text corrected

Discussion

• Line 263 suggested wording "Since then, we have observed six MelTs that account for 82.8% of isolates, from a total of 38 MelTs..."

o Line 267-268 – Text corrected

• Line 276 instead of "big bacterial populations", suggested wording "large populations of bacteria"

o Line 280 – Text corrected

Overall

• Fig2 legend, Same issue as raised by Reviewer1, is the tree based on number of shared alleles or the sequence similarity in the shared alleles? Mostly likely, this legend should include "is based on sharing of 2,251 core genome targets." Please revise as appropriate.

o Line 194 – Fig 2 legend revised

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

1 Aug 2019

Application of mini-MLST and whole genome sequencing in low diversity hospital extended-spectrum beta-lactamases producing Klebsiella pneumoniae population

PONE-D-19-14966R2

Dear Dr. Lengerova,

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

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. 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 enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and 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.

With kind regards,

D. Ashley Robinson, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

5 Aug 2019

PONE-D-19-14966R2

Application of mini-MLST and whole genome sequencing in low diversity hospital extended-spectrum beta-lactamase producing Klebsiella pneumoniae population

Dear Dr. Lengerova:

I am 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 notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, 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.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. D. Ashley Robinson

Academic Editor

PLOS ONE