Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Oct 1.
Published in final edited form as: J Infect. 2019 Jul 30;79(4):312–321. doi: 10.1016/j.jinf.2019.07.009

Multifaceted mechanisms of colistin resistance revealed by genomic analysis of multidrug-resistant Klebsiella pneumoniae isolates from individual patients before and after colistin treatment

Yan Zhu a, Irene Galani b, Ilias Karaiskos c, Jing Lu a, Su Mon Aye a, Jiayuan Huang a, Heidi H Yu a, Tony Velkov d, Eleni Giamarellou c,*, Jian Li a,*
PMCID: PMC7264071  NIHMSID: NIHMS1590989  PMID: 31374222

Abstract

Objectives

Polymyxins (i.e. polymyxin B and colistin) are used as a last-line therapy to combat multidrug-resistant (MDR) Klebsiella pneumoniae. Worryingly, polymyxin resistance in K. pneumoniae is increasingly reported worldwide. This study identified the genetic variations responsible for high-level colistin resistance in MDR K. pneumoniae clinical isolates.

Methods

Sixteen MDR K. pneumoniae isolates were obtained from stool samples of 8 patients before and after colistin treatment. Their genomes were sequenced on Illumina MiSeq to determine genetic variations.

Results

Fifteen of 16 isolates harbored ISKpn26-like element insertion at nucleotide position 75 of mgrB, abolishing its negative regulation on phoPQ; while colistin-susceptible ATH7 contained intact mgrB and phoQ. Interestingly, each of the 7 mgrB-disrupted, colistin-susceptible isolates contained a nonsynonymous substitution in PhoQ (G39S, L239P, N253T or V446G), potentially impairing its function and intergenically suppressing the effect caused by mgrB inactivation. Additionally, three of the 7 corresponding mgrB-disrupted, colistin-resistant isolates harboured a secondary nonsynonymous substitution in PhoQ (N253P, D438H or T439P).

Conclusions

This is the first report of phoQ mutations in mgrB-disrupted, colistin-susceptible K. pneumoniae clinical isolates. We also discovered multiple phoQ mutations in mgrB-disrupted, colistin-resistant strains. Our findings highlight the multifaceted molecular mechanisms of colistin resistance in K. pneumoniae.

Keywords: Klebsiella pneumoniae, genomics, colistin, mgrB, phoQ

Introduction

Multidrug-resistant (MDR) Klebsiella pneumoniae is a severe threat to human health worldwide and urgently requires novel treatment options.1 K. pneumoniae is a major cause of hospital-acquired infections including pneumonia, bloodstream infections, and infections in newborns and critically-ill patients.2–4 The increasing reports of MDR K. pneumoniae isolates that are resistant to all classes of antibiotics, including the last-line polymyxins, are most disconcerting.3, 5–7

Polymyxins were originally discovered in the 1940s and abandoned in the 1970s due to potential nephrotoxicity and neurotoxicity; however, over the last decade they have been revived as a last resort to treat severe infections caused by Gram-negative ‘superbugs’.8–10 Of the major polymyxin families (e.g. polymyxins A to E), only polymyxin B and colistin (i.e. polymyxin E) are used in the clinic.10 Polymyxins are a class of polycationic lipopeptide antibiotics with a narrow spectrum of activity against Gram-negative bacteria.11 The exact mode of action of polymyxins is not clear, but involves initial electrostatic and hydrophobic interactions between the cationic polymyxin molecule and the anionic lipid A of lipopolysaccharides (LPS) in bacterial outer membrane (OM), followed by the replacement of divalent cationic ions (e.g. Mg2+ and Ca2+), disorganisation of cell envelope, cellular component leakage, and eventually cell death.12 Alternative mechanisms were also reported, including inhibition of membrane-bound type II NADH-quinone oxidoreductase and generation of cytotoxic hydroxyl radicals.13, 14

In K. pneumoniae, polymyxin resistance is mostly mediated by modifications of lipid A with 4-amino-4-deoxy-L-arabinose (L-Ara4N, arnBCADTEF-ugd loci) and/or phosphoethanolamine (pEtN, pmrC or plasmid-borne mcr-1).15–17 The chromosomally encoded lipid A modifications can be induced by several two-component systems (TCSs), including PhoPQ and PmrAB.18, 19 MgrB is a small transmembrane protein and exerts negative feedback regulation on phoPQ.20 Loss-of-function mutations of mgrB invoked constitutive induction of phoPQ, thereby upregulating the lipid A modification and conferring polymyxin resistance.20–26 Moreover, mgrB mutations were not necessarily associated with a significant fitness cost and could be stably maintained in the absence of selective pressure.27 Interestingly, a recent study showed the presence of partial suppressor mutations of phoQ (N253T and V446G) in mgrB-disrupted, colistin-resistant K. pneumoniae clinical isolates, indicating that these additional phoQ mutations could constitute a fitness advantage.23 However, it is unclear whether these varations are prevalent in mgrB-inactivated MDR K. pneumoniae clinical isolates.

In this study, we report the comparative genomic analysis of the 16 colistin-susceptible and -resistant carbapenemase-producing K. pneumoniae isolates obtained from 8 patients before and after colistin treatment. The results provide key insights into the acquisition of polymyxin resistance in this problematic ‘superbug’.

Materials and methods

Bacterial strains, media and antimicrobial susceptibility test

Sixteen isolates were obtained from stool samples of eight patients with an infection of carbapenemase-producing K. pneumoniae before and after colistin treatment in Athens, Greece. Antimicrobial compounds tested in this study are listed in Table S1. Colistin MICs were determined using broth microdilution according to the Clinical and Laboratory Standards Institute guideline.28

Genome sequencing and assembly

Bacterial genomic DNA was extracted from log-phase cultures grown in cation-adjusted Muller-Hinton broth (CAMHB, Oxoid) using a DNeasy Blood and Tissue Kit (QIAGEN, Dusseldorf, Germany). Electrophoresis and Qubit (Life Technologies, USA) were used to assess the quality and quantity of DNA samples, respectively. DNA libraries were constructed using a Nextera XT DNA Library Preparation Kit (Illumina, San Diego, CA, USA), and indexed with TruSeq Dual Indexing Reagent (Illumina). Single-end 75 bp DNA sequencing was performed on an Illumina MiSeq (Micromon, Clayton, Australia). Quality-trimmed reads were employed for de novo genome assembly with SPAdes,29, 30 and the results were evaluated using QUAST 4.5.31 With Mauve Contigs Mover,32 the assembled contigs were ordered with complete reference genomes of TGH13 and AUSMDU00008079, the two closest isolates based on phylogenetic analysis (below). Genome annotation was conducted with the ordered contigs using Prokka,33 and insertion sequences (ISs) were predicted using ISEScan.34 The assembly of each K. pneumonia genome was uploaded to the Resistance Gene Identifier (RGI) 3.2.1 web portal from the Comprehensive Antimicrobial Resistance Database (CARD) to predict antimicrobial resistance genes using Strict algorithm.35 Specifically, the Strict algorithm detects the homologs of the known antimicrobial resistance genes in CARD database.35 For each genome, the predicted antimicrobial resistance genes were grouped according to the Drug Class categorisation. Heatmap was generate to show the presence and absence of classes of antimicrobial resistance genes. IslandViewer 4 was employed to predict genomic islands.36

Sequence typing and phylogenetic analysis

The 16 assemblies and 143 K. pneumoniae complete genomes from GenBank (Table S2) were submitted to Multilocus Sequencing Typing (MLST) database to determine sequence types (STs),37 and were compared to determine the core genome using Harvest tool.38 The derived core genome alignment was employed to estimate Maximum-Likelihood (ML) tree using RAxML 8.2.9 with GTRGAMMA model (150 bootstrap replicates, MRE-based bootstrapping criterion).39 The tree was visualised using iTOL v4.4.2.40

Genetic variation detection

For each of the 16 isolates, quality-trimmed reads were aligned to the reference (TGH13 or AUSMDU00008079) using SubRead.41 Nesoni was used to identify, filter and annotate single nucleotide polymorphism (SNP).42 Specifically, SNPs were identified using a diploid analysis with the predominant variants showing in ≥66.7% of the mapped reads and having a quality score >20. Structure variations were initially identified using GRIDSS (quality score ≥ 1,000 and having assemblies from both sides of the breakpoint) with the reference genomes and further checked by Artemis bamview.43, 44 Gene presence/absence variation was analysed using Roary with a sequence identity of ≥99%.45 Mutations in mgrB and phoQ were validated by PCR amplification and Sanger sequencing using the primers in Table S3. PROVEAN (PRotein Variation Effect ANalyzer) was employed to predict the function impact of an amino acid substitution in PhoQ.46

Results

Antimicrobial susceptibility

For 36 antibiotics of 14 major classes, 15 isolates showed resistance to 35 antibiotics except ATH21, and were thus classified as extensively drug-resistant (XDR) according to the breakpoints from the European Committee on Antimicrobial Susceptibility Testing (Table S1).47 Strain ATH21 was susceptible to imipenem (MIC = 0.25 mg/L) and gentamicin (MIC = 1 mg/L). Our broth microdilution results showed that all the 8 colistin-susceptible strains had colistin MICs ≤ 0.5 mg/L; whereas the other 8 colistin-resistant isolates had colistin MICs ≥ 64 mg/L (Table 1), representing 128 to 1,024-fold increase.

Table 1.

MICs of polymyxin B and colistin against MDR K. pneumoniae isolates.

Patient Strain MIC (mg/L)
Colistin Polymyxin B
A ATH7 0.25 <0.125
ATH8 >128 128
B ATH9 0.25 0.5
ATH10 64 32
C ATH15 <0.125 <0.125
ATH16 128 128
D ATH17 <0.125 0.25
ATH18 128 128
E ATH21 0.25 0.5
ATH22 64 32
F ATH23 0.5 0.5
ATH24 64 64
G ATH25 0.5 0.5
ATH26 128 64
H ATH30 0.5 0.5
ATH29 64 32
Open in a new tab

Genome sequencing and assembly

Genome sequencing of the 16 K. pneumoniae isolates yielded 2.85–3.10 million reads per sample, equivalent to 181 to 241 MB data per isolate (Table S4). After filtering, 2.08–2.78 million quality-trimmed reads were obtained for each sample with the length of 67–72 bp, representing approximately 30-fold coverage of the genomes (Table S4). De novo assembly yielded 238–399 contigs (≥500 bp) for each isolate with the total length ranging from 5,364,335 to 5,748,458 bp, GC content in 56.9–57.3% and N50 of 33,800–66,326 bp (Table 2). Overall, 5,079–5,524 genes and 5,017–5,466 protein-encoding sequences were annotated from the 16 draft genomes (Table 2). The genomes were deposited in DDBJ/ENA/GenBank under the accession numbers shown in Table 2.

Table 2.

Draft assemblies of the 16 K. pneumoniae isolates.

Isolate Accession No. Assembly size (bp) No. of Contigs (≥500 bp) GC content (%) No. of Plasmids No. of Genes MLST type
ATH7 VJXP00000000 5,708,072 287 57.04 6 5,485 ST258
ATH8 VJXO00000000 5,710,068 320 57.03 4 5,484 ST258
ATH9 VJXN00000000 5,446,905 334 57.30 1 5,155 ST147
ATH10 VJXM00000000 5,364,335 258 57.30 1 5,079 ST147
ATH15 VJXL00000000 5,681,214 334 57.08 3 5,439 ST258
ATH16 VJXK00000000 5,682,327 335 57.08 2 5,443 ST258
ATH17 VJXJ00000000 5,716,616 376 56.94 5 5,492 ST258
ATH18 VJXI00000000 5,713,722 399 56.95 11 5,486 ST258
ATH21 VJXH00000000 5,521,083 327 57.20 2 5,272 ST258
ATH22 VJXG00000000 5,633,725 285 57.12 4 5,368 ST258
ATH23 VJXF00000000 5,605,330 238 57.11 4 5,351 ST258
ATH24 VJXE00000000 5,604,784 256 57.11 3 5,359 ST258
ATH25 VJXD00000000 5,748,458 329 57.02 6 5,524 ST258
ATH26 VJXC00000000 5,631,750 305 57.10 3 5,411 ST258
ATH29 VJXA00000000 5,600,153 307 57.11 3 5,349 ST258
ATH30 VJXB00000000 5,600,123 264 57.11 3 5,348 ST258
Open in a new tab

Phylogenetic analysis

MLST analysis with the 7 housekeeping genes gapA, infB, mdh, pgi, phoE, rpoB and tonB classified ATH9/ATH10 to ST147 and the other 14 isolates to ST258. A phylogenetic tree (Fig 1) was constructed based on the core genome alignment using the 16 draft assemblies and 143 complete genomes from GenBank (Table S2). Strains TGH13 (Genbank Accession No. CP012745, ST147) and AUSMDU00008079 (GenBank Accession No. CP022691–CP022694, ST258) were phylogenetically closest to ATH9/ATH10 and the other 14 isolates, respectively (Fig 1). Therefore, TGH13 and AUSMDU0008079 genomes were employed as references to identify genetic variations.

Figure 1.

Figure 1.

Open in a new tab

ML-based phylogeny of K. pneumoniae, containing the 16 isolates and 143 isolates with complete genomes (A). Bootstrap values (percentage) are indicated by a colour scale from red (1%) to green (100%). Specifically, the closely related K. pneumoniae genomes are shown in (B). The ST147 isolates ATH9 and ATH10 are highlighted in orange; and the 14 ST258 isolates are in blue. Reference strains for genetic variation analysis are in bold. The sequence type is shown after the isolate name for each genome. Scale bars in (A) and (B) indicate the number of substitutions per nucleotide site.

Mobile elements

Across the 16 isolates, 61 putative plasmids were assembled, with 1 (ATH10) to 11 (ATH18) plasmids per isolate and the plasmid size ranging from 1.1 to 200 Kb (Table S5). Seventeen plasmids were assigned to known incompatibility (Inc) groups including ColRNAI, IncFIB, IncFIC, IncFII, IncHI1B and IncX3; the most common Inc group was ColRNAI (detected using ColRNAI_DQ298019 probe) which was present in 6 colistin-susceptible and 5 colistin-resistant isolates.48 Additionally, 392 insertion sequences (ISs) were predicted with the length ranging from 294 to 3,030 bp, representing 55–58 unique ISs from 20 families (Fig 2 and Table S6). ISNCY accounted for 33.5% of the total ISs and was the most abundant (Fig 2). Strains ATH10, ATH16 and ATH17 contained much less ISs (<10) than the others (Fig 2). Strains ATH7/ATH8, ATH23/ATH24 and ATH30/ATH29 contained similar IS elements, indicating their close phylogenetic relationships.

Figure 2.

Figure 2.

Open in a new tab

Predicted insertion sequences in the 16 K. pneumoniae isolates.

Genomic content and resistance genes

Among the 16 isolates, 449 genes were predicted to confer resistance to different antimicrobial classes including fluoroquinolones (gyrAB, parC and qnrA1), fosfomycin (uhpT and fosA6), aminoglycosides [aac(6’)-Ib10, aac(6’)-Ib7, aadA, aadA2, ant(2”)-Ia, aph(3’)-Ia, aph(3”)-Ib and aph(6)-Id], chloramphenicol (catI), diaminopyrimidines (dfrA12 and dfrA14), macrolides (mphA and mrx) and sulfonamides (sul1 and sul2) (Fig 3 and Table S7). Totally, 146 efflux pump genes (e.g. emrB, marA, msbA, msrB, ompK37, oqxAB, patA and vgaC) were identified in the 16 isolates, and likely rendered the isolates resistant to a broad range of antibiotics; 80 regulators (e.g. emrR, hns, baeR) were discovered and assumed to regulate the expression of efflux pump genes. Notably, all the 16 genomes harboured β-lactamase genes, including blaKPC-2, blaSHV-11, blaSHV-5, blaTEM-1 and blaVIM-1. In addition, 36–60 genomic island fragments per isolate were predicted, with the length ranging from 4.0 to 386.9 Kb (Table S8). Among them, 104 island fragments (12.4% of 840 island fragments in the 16 isolates) harboured a total of 203 resistance genes (3–17 resistance genes per isolate), conferring resistance to β-lactams, aminoglycosides, chloramphenicol, and diaminopyrimidines.

Figure 3.

Figure 3.

Open in a new tab

The absence (blank) and presence (colour) of the predicted antibiotic resistance genes in the 16 K. pneumoniae isolates.

Genetic variations in colistin-susceptible and -resistant isolates from the same patients

By mapping the reads to their corresponding reference genomes, 2,762 SNPs were identified among the 16 isolates with 120–195 SNPs per isolate. Totally, 303 (10.9%) SNPs were exclusively discovered in either colistin-susceptible or -resistant isolates; among them, 266 SNPs were present in 100% of the sequencing reads (Tables 3 and S9; Fig 4). Interestingly, 186 unique SNPs were identified between ATH25 and ATH26, indicating their significant genetic discrepancies (Table S9). Strains ATH7 and ATH21 contained a stop-gain (E132Stop) and a frameshift variation (I115fs) in ramR, respectively; both might abolish its transcriptional repression on the downstream regulator ramA and potentially induced the expression of AcrAB efflux pump, thereby rendering strains resistant to tigecycline.49

Table 3.

SNPs identified by comparative genomics analysis.

Isolate Total No. of SNPs compared to the reference genome No. of unique SNPs No. of SNPs between colistin-susceptible and -resistant strains
ATH7 121 8 15
ATH8 120 7
ATH9 157 5 14
ATH10 161 9
ATH15 193 6 13
ATH16 194 7
ATH17 188 4 15
ATH18 195 11
ATH21 159 19 32
ATH22 153 13
ATH23 188 5 10
ATH24 188 5
ATH25 179 89 186
ATH26 187 97
ATH30 188 6 18
ATH29 191 12
Open in a new tab

Figure 4.

Figure 4.

Open in a new tab

Unique nonsynonymous SNPs of the 16 K. pneumoniae isolates. The circular tracks, from outermost to innermost, represent the genes on positive and negative strands of the reference genomes, SNPs in colistin-susceptible (green) and -resistant isolates (red), amino acid substitutions, and the affected genes. The genes are colour-coded according to the predicted Clusters of Orthologous Groups (COG) functional categories of reference genomes using eggNOG tool.64 For SNP tracks, from outermost to innermost they are ATH9 and ATH10 in (A), and ATH7, ATH8, ATH15, ATH16, ATH17, ATH18, ATH21, ATH22, ATH23, ATH24, ATH25, ATH26, ATH 30 and ATH29 in (B); in each track, the SNPs are denoted by solid circles according to their quality score, with a range of 20–100 (from the lightest to darkest shading). The phoQ SNPs are highlighted by red arrows. Within 1.35–1.40 Mb (B), 73 SNPs were identified mostly in CI103_06880 (integrase) and CI103_07080–7095 (hypothetical proteins). No SNP track is depicted in (B) if there is no SNPs on reference plasmids.

Overall, 102 DNA breakpoints were predicted in 16 isolates, with 4–10 per isolate (Table S10). Comparative genomic analysis determined a core genome consisting 4,224 genes (≥99% identity in amino acid sequence), and a pan genome of totally 6,708 genes (Table S11). ATH9 and ATH10 showed significant differences from the other isolates in genome content, consistent with their distinct sequence type classification (ST147, Fig S1).

Genetic variations associated with colistin susceptibility and resistance

Single nucleotide variations were identified in the genes that are associated with colistin resistance in K. pneumoniae, including TCSs and lipid A modification genes (Table 4). The ST147 isolates ATH9 and ATH10 contained a unique missense variation (T140P) in pmrB compared to their reference TGH13. All 14 ST258 isolates contained a V53G variation in phoP compared to the reference AUSMDU00007089; whereas the two ST147 isolates ATH9 and ATH10 possessed an intact phoP compared to the reference TGH13. Missense variations of phoQ were identified in 10 of 16 isolates (Table 4; Fig 5). In 4 colistin-susceptible isolates ATH9, ATH23, ATH25 and ATH30, single missense variations (G39S, V446G or L239P) were identified in phoQ; while the corresponding colistin-resistant isolates (ATH10, ATH24, ATH26 and ATH29) obtained from the same patients after colistin treatment possessed an intact phoQ (Table 4). The other three colistin-susceptible ST258 isolates (ATH15, ATH17 and ATH21) shared a common missense variation phoQN253T that was caused by a nucleotide substitution (758A>C). Their corresponding colistin-resistant isolates had either an additional missense variation (D438H in ATH16, and T439P in ATH18), or a variation (N253P in ATH22) caused by dinucleotide substitution (757_758delAAinsCC) in phoQ. This dinucleotide substitution occurred very likely via acquiring a second nucleotide substitution (757A>C) in the same codon (AAC for Asn253) of phoQ. In addition, a premature mutation (E307Stop) was discovered in pmrC of ATH9 and ATH10, indicating that both were unable to modify lipid A with pEtN. Strain ATH21 contained a specific missense variation (R306S) in arnB (UDP-4-amino-4-deoxy-L-arabinose:oxoglutarate aminotransferase) and an in-frame deletion (E176del) in ugd (UDP-glucose 6-dehydrogenase). Both genes belonged to the arnBCADTEF-ugd loci that are responsible for lipid A modification with L-Ara4N.20 Both ST147 isolates ATH9 and ATH10 shared a common variation in the arn promoter (−58A>G). Isolate ATH7 contained an S204Y variation in the LPS heptosyltransferase gene rfaQ (Table S9).

Table 4.

Genetic variations contributing to colistin resistance.

Isolate Colistin susceptibility Category of resistance mechanism mgrB pmrB phoP phoQ pmrC arnB ugd
CI103_10160/AOG30_10755a CI103_03610/AOG30_18810 CI103_16945/AOG30_17125 CI103_16950/AOG30_17130 CI103_18860/AOG30_18800 CI103_01375/AOG30_01390 CI103_08920 /AOG30_09695
ATH7 S V53G (158T>Gb)
ATH8 R I mgrB::N25ISKpn26-like V53G (158T>G)
ATH15 S mgrB::N25ISKpn26-like V53G (158T>G) N253T (758A>C)
ATH16 R II mgrB::N25ISKpn26-like V53G (158T>G) N253T (758A>C), D438H (1312G>C)
ATH17 S mgrB::N25ISKpn26-like V53G (158T>G) N253T (758A>C)
ATH18 R II mgrB::N25ISKpn26-like V53G (158T>G) N253T (758A>C), T439P (1315A>C)
ATH21 S mgrB::N25ISKpn26-like V53G (158T>G) N253T (758A>C) R306S (916C>A) E176Del (527_529delAAG)
ATH22 R II mgrB::N25ISKpn26-like V53G (158T>G) N253P (757_758delAAinsCC)
ATH23 S mgrB::N25ISKpn26-like V53G (158T>G) V446G (1337T>G)
ATH24 R I mgrB::N25ISKpn26-like V53G (158T>G)
ATH25 S mgrB::N25ISKpn26-like V53G (158T>G) L239P (716T>C)
ATH26 R I mgrB::N25ISKpn26-like V53G (158T>G)
ATH30 S mgrB::N25ISKpn26-like V53G (158T>G) V446G (1337T>G)
ATH29 R I mgrB::N25ISKpn26-like V53G (158T>G)
ATH9 S mgrB::N25ISKpn26-like T140P (418A>C) −38G>T G39S (115G>A) E307Stop (919G>T) −58A>G
ATH10 R I mgrB::N25ISKpn26-like T140P
(418A>C)		 −38G>T E307Stop (919G>T) −58A>G
Open in a new tab
a

Loci in AUSMDU0008079/TGH13 genome.

b

Nucleotide variations.

Figure 5.

Figure 5.

Open in a new tab

Mutations in PhoQ and MgrB. The mgrB inactivation (via ISKpn26-like integration) is denoted by red ‘×’. The nonsynonymous SNPs in phoQ are indicated by blue (identified exclusively in colistin-resistant strains) and red (the rest) dots. The function domains of PhoQ were assigned based on the literature.65

Very surprisingly, 15 K. pneumoniae isolates exhibited integration of an IS element in mgrB with the only exception of ATH7 (Table 4; Fig 5). Further PCR amplification and Sanger sequencing confirmed that the insertion occurred at the 75 nt position of mgrB in the forward direction among 13 isolates, the same position but in the reverse direction in ATH9 and ATH10. The 1,196-bp IS element discovered in this study belonged to IS5 family and showed 99% nucleotide identity with ISKpn26, therefor designated ISKpn26-like element.50

Discussion

In this study, we undertook comparative genomic analysis with 16 colistin-susceptible and -resistant clinical isolates and identified the genetic variations potentially responsible for high-level colistin resistance in K. pneumoniae. The K. pneumoniae ST258 group is an important cause of MDR hospital outbreaks,51 while ST147 is a newly emerged human-related K. pneumoniae sequence type and also associated with MDR outbreaks.52, 53 Among the 16 isolates, 14 were classified as ST258, whereas ATH9 and ATH10 were grouped to ST147 (Table 2). Consistent with their XDR profiles, the assembled genomes harboured a variety of antibiotic resistance genes (Table S7; Fig 3). Plasmids, ISs and genomic islands were predicted and highly likely they contributed to the acquisition of MDR (Tables S5, S6 and S8; Fig 2). With TGH13 and AUSMDU00008079 genomes as references, SNPs, gene presence/absence variations, and structural variations were identified in 16 isolates (Tables 3 and S9–11; Figs 4 and S1). Key genetic variations associated with polymyxin resistance were identified, including an ISKpn26-like element insertion in mgrB at 75-nt and several nonsynonymous variations in phoQ, pmrC, arnB and ugd (Table 4; Fig 5).

K. pneumoniae commonly evolves colistin resistance via pEtN and L-Ara4N modifications on the lipid A component of LPS.15–17 The expression of relevant genes (pmrC and arn operon) are controlled by TCSs PmrAB and PhoPQ.18, 20 Environmental stimuli (e.g. high concentrations of ferric iron for PmrB, low concentrations of divalent cationic ions, and presence of cationic antimicrobial peptides for PhoQ) can trigger the auto-phosphorylation of membrane-bound kinases PmrB and PhoQ, which in turn activate the cognate regulators PmrA and PhoP by phosphorylation, respectively.54 In K. pneumoniae, the activated PhoP promotes the transcription of a large number of genes, including host adaptation genes, arn operon and connecter pmrD.18, 55, 56 PmrD in turn binds to phosphorylated PmrA, and thereafter leading to persistent expression of PmrA-activated genes (e.g. pmrCAB, ugd).18, 55, 56 Mutations in pmrAB and phoPQ can render K. pneumoniae resistant to polymyxins.18, 22 In our study, all the colistin-susceptible and their corresponding colistin-resistant isolates had identical pmrAB and phoP, indicating that they might not contribute to polymyxin resistance. Whereas the genetic variations in phoQ were identified in 7 colistin-susceptible and 3 colistin-resistant isolates (Table 4), suggesting that the phoQ mutations may not always play a role in acquisition of polymyxin resistance. Specifically, as shown in Fig 5, the 7 colistin-susceptible isolates (ATH9, ATH15, ATH17, ATH21, ATH23, ATH25 and ATH30) had a mutated phoQ compared to their reference genomes, while the corresponding colistin-resistant isolates obtained from the same patients after colistin treatment contained either an intact (ATH10, ATH24, ATH26 and ATH29) or a secondary mutation in phoQ (ATH16, ATH18 and ATH22). Our results indicates that certain genetic variations in phoQ may not contribute to colistin resistance, while other genetic variations may lead to high-level colistin resistance.

Interestingly, apart from phoQ variations, in 15 isolates we identified the insertion of the ISKpn26-like element at 75-nt of mgrB either in the forward (13 isolates) or reverse direction (2 isolates) (Table 4). MgrB is a small membrane protein with only 47 amino acids and is a negative regulator of PhoPQ (Fig 5).22 Previous studies showed that inactivation of mgrB resulted in the upregulation of PhoPQ, which in turn induced lipid A modifications and high-level polymyxin resistance in K. pneumoniae.21, 22 Epidemiologically, mgrB inactivation (e.g. non-synonymous mutations, insertion, deletion and integration with IS element) is widely reported in carbapenemase-producing K. pneumoniae.7, 15, 21, 23, 26, 57 These 8 ‘pairs’ of isolates are not isogenic strains that are commonly used in molecular microbiological studies, but were collected before and after colistin treatment in individual patients. Together, our results showed that the mechanisms of colistin resistance in the 8 isolates (ATH8, ATH10, ATH16, ATH18, ATH22, ATH24, ATH26 and ATH29) were mainly mediated by (i) inactivation of mgrB only (ATH8, ATH10, ATH24, ATH26 and ATH29), and (ii) inactivation of mgrB plus second-site mutations in phoQ (ATH16, ATH18 and ATH22) as shown in Table 4. Previous studies showed mgrB disruption (mgrB::N25ISKpn26-like) caused polymyxin resistance in ST258 Greek isolates.23 Missense variations of phoQ (N253T, V446G, L239P and G39S) were identified in mgrB-disrupted isolates (ATH9, ATH15, ATH17, ATH21, ATH23, ATH25 and ATH30) (Table 4), which potentially rendered these strains susceptible to polymyxins. Specifically, as shown in Fig 5, these phoQ SNPs occurred in the transmembrane domain (G39S), HAMP domain (present in histidine kinases, adenylate cyclases, methyl accepting proteins and phosphatases, N253T and L239P) and histidine kinase domain (V446G). Among them, N253T and V446G were previously discovered along with mgrB disruption in colistin-resistant K. pneumoniae clinical isolates.23 Individually complementation of mutant phoQN253T or phoQV446G in mgrB-disrupted isolate reduced the polymyxin MICs, suggesting that N253T and V446G might exert partial suppressor effect on mgrB inactivation.23 Previous structural analysis also showed that substitution of N253 by a residue with a shorter side chain (e.g. Thr in the present study) would decrease PhoQ activity.58 Therefore, the phoQ mutations (N253T, V446G, L239P and G39S) identified in our study might cause dysfunction of PhoQ and stopped the downstream events on lipid A modifications, and therefore the mgrB-disruption did not cause colistin resistance in these susceptible isolates (i.e. ATH9, ATH15, ATH17, ATH21, ATH23, ATH25 and ATH30). This is the first report of intergenic suppressions in mgrB-disrupted, colistin-susceptible, carbapenemase-producing MDR K. pneumoniae clinical isolates.

Previous studies showed that mgrB disruption is not necessarily associated with significant fitness cost.27, 59, 60 The induction of phoPQ expression via mgrB disruption not only confers K. pneumoniae resistance to polymyxins and cationic antimicrobial peptides, but also enhances virulence by attenuating the host defence response activation.59, 60 Interestingly, a few other studies reported reduced virulence,61 loss of hypermucoviscosity,62 and declined growth in mgrB-disrupted K. pneumoniae strains,63 indicating that the fitness cost and benefits of mgrB inactivation are strain-dependent. In our colistin-susceptible strains, mgrB disruption was accompanied with phoQ mutations, suggesting that mgrB disruption alone might result in certain fitness cost during host adaptation and K. pneumoniae may retain its fitness by acquiring function-impairing phoQ mutations. Whereas under colistin selection pressure, the PhoQ function in 3 isolates (ATH16, ATH18 and ATH22) could be ‘restored’ by the second-site mutations (D438H, T439P and N253P [caused by second nucleotide substitution 757A>C following 758A>C]), thereby leading to lipid A modifications and colistin resistance. Other genetic variations may also contribute to the suppression effect on mgrB-disruption in colistin-susceptible isolates. In colistin-susceptible ATH21, the arnBR306S and ugdE176del might abolish lipid A modification with L-Ara4N (Table 4). In colistin-susceptible isolate ATH7, the rfaQS204Y variation might affect the heptosyltransferase activity and then impair lipid A - core oligosaccharide biosynthesis (Table S9). In contrast, these variations were absent in the colistin-resistant isolates ATH22 and ATH8. Genetic variations of pmrBT140P, pmrCE307Stop, phoPV53G and −58A>G in arnB promoter were also identified in both colistin-susceptible and -resistant isolates (Table 4), indicating that they are irrelevant to colistin resistance.

In summary, key genetic variations were identified in 16 colistin-susceptible and -resistant MDR K. pneumoniae clinical isolates by genome sequencing and comparative genomic analyses. These variations involved an ISKpn26-like element insertion in mgrB (at nucleotide position 75) and 7 missense mutations in phoQ (G39S, L239P, N253T, N253P, D438H, T439P and V446G). Importantly, this is the first report of phoQ mutations accompanied with disrupted-mgrB in colistin-susceptible K. pneumoniae clinical isolates, and we also discovered secondary phoQ mutations in mgrB-disrupted, colistin-resistant strains. Our findings highlight the multifaceted molecular mechanisms of colistin resistance in K. pneumoniae.

Supplementary Material

Supplementary Figure

Figure S1. Gene presence/absence variations across the 16 isolates.

Supplementary Tables

Acknowledgements

J.L. and T.V. are supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health (Grant number R01 AI132154). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Allergy and Infectious Diseases or the National Institutes of Health. J.L. and T.V. are also supported by an Australian National Health and Medical Research Council (NHMRC) project grant (APP1104581). T.V. is an Australian NHMRC Career Development Research Fellow. J.L. is an Australian NHMRC Principal Research Fellow. The authors thank Monash MicroMon (Clayton, Australia) for genome sequencing services.

Footnotes

Part of the work has been presented at 28th European Congress of Clinical Microbiology and Infectious Diseases at April 21–24, 2018, Madrid, Spain.

Transparency declarations

The authors declare no conflicts of interest.

References

  • 1.World Health Organization. Global priority list of antibiotic-resistant bacteria to guide research, discovery, and development of new antibiotics 2017. Available from: https://www.who.int/medicines/publications/WHO-PPL-Short_Summary_25Feb-ET_NM_WHO.pdf?ua=1.
  • 2.Munoz-Price LS, Poirel L, Bonomo RA, Schwaber MJ, Daikos GL, Cormican M, et al. Clinical epidemiology of the global expansion of Klebsiella pneumoniae carbapenemases. Lancet Infect Dis. 2013;13(9):785–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Pragasam AK, Shankar C, Veeraraghavan B, Biswas I, Nabarro LE, Inbanathan FY, et al. Molecular mechanisms of colistin resistance in Klebsiella pneumoniae causing bacteremia from India—a first report. Front Microbiol. 2016;7:2135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Brooks B, Olm MR, Firek BA, Baker R, Thomas BC, Morowitz MJ, et al. Strain-resolved analysis of hospital rooms and infants reveals overlap between the human and room microbiome. Nat Commun. 2017;8(1):1814. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Navon-Venezia S, Kondratyeva K, Carattoli A. Klebsiella pneumoniae: a major worldwide source and shuttle for antibiotic resistance. FEMS Microbiol Rev. 2017;41(3):252–75. [DOI] [PubMed] [Google Scholar]
  • 6.Rojas LJ, Salim M, Cober E, Richter SS, Perez F, Salata RA, et al. Colistin resistance in carbapenem-resistant Klebsiella pneumoniae: laboratory detection and impact on mortality. Clin Infect Dis. 2017;64(6):711–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Cheng YH, Lin TL, Pan YJ, Wang YP, Lin YT, Wang JT. Colistin resistance mechanisms in Klebsiella pneumoniae strains from Taiwan. Antimicrob Agents Chemother. 2015;59(5):2909–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Velkov T, Roberts KD, Nation RL, Thompson PE, Li J. Pharmacology of polymyxins: new insights into an ‘old’ class of antibiotics. Future Microbiol. 2013;8(6):711–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Landman D, Georgescu C, Martin DA, Quale J. Polymyxins revisited. Clin Microbiol Rev. 2008;21(3):449–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Li J, Nation RL, Turnidge JD, Milne RW, Coulthard K, Rayner CR, et al. Colistin: the re-emerging antibiotic for multidrug-resistant Gram-negative bacterial infections. Lancet Infect Dis. 2006;6(9):589–601. [DOI] [PubMed] [Google Scholar]
  • 11.Velkov T, Roberts KD, Thompson PE, Li J. Polymyxins: a new hope in combating Gram-negative superbugs? Future Med Chem. 2016;8(10):1017–25. [DOI] [PubMed] [Google Scholar]
  • 12.Urakawa H, Yamada K, Komagoe K, Ando S, Oku H, Katsu T, et al. Structure-activity relationships of bacterial outer-membrane permeabilizers based on polymyxin B heptapeptides. Bioorg Med Chem Lett. 2010;20(5):1771–5. [DOI] [PubMed] [Google Scholar]
  • 13.Deris ZZ, Akter J, Sivanesan S, Roberts KD, Thompson PE, Nation RL, et al. A secondary mode of action of polymyxins against Gram-negative bacteria involves the inhibition of NADH-quinone oxidoreductase activity. J Antibiot (Tokyo) 2014;67(2):147–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Sampson TR, Liu X, Schroeder MR, Kraft CS, Burd EM, Weiss DS. Rapid killing of Acinetobacter baumannii by polymyxins is mediated by a hydroxyl radical death pathway. Antimicrob Agents Chemother. 2012;56(11):5642–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Wright MS, Suzuki Y, Jones MB, Marshall SH, Rudin SD, van Duin D, et al. Genomic and transcriptomic analyses of colistin-resistant clinical isolates of Klebsiella pneumoniae reveal multiple pathways of resistance. Antimicrob Agents Chemother. 2015;59(1):536–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Liu YY, Wang Y, Walsh TR, Yi LX, Zhang R, Spencer J, et al. Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. Lancet Infect Dis. 2016;16(2):161–8. [DOI] [PubMed] [Google Scholar]
  • 17.Leung LM, Cooper VS, Rasko DA, Guo Q, Pacey MP, McElheny CL, et al. Structural modification of LPS in colistin-resistant, KPC-producing Klebsiella pneumoniae. J Antimicrob Chemother. 2017;72(11):3035–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Cheng HY, Chen YF, Peng HL. Molecular characterization of the PhoPQ-PmrD-PmrAB mediated pathway regulating polymyxin B resistance in Klebsiella pneumoniae CG43. J Biomed Sci. 2010;17:60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kim SY, Choi HJ, Ko KS. Differential expression of two-component systems, pmrAB and phoPQ, with different growth phases of Klebsiella pneumoniae in the presence or absence of colistin. Curr Microbiol. 2014;69(1):37–41. [DOI] [PubMed] [Google Scholar]
  • 20.Olaitan AO, Morand S, Rolain JM. Mechanisms of polymyxin resistance: acquired and intrinsic resistance in bacteria. Front Microbiol. 2014;5:643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Cannatelli A, Giani T, D’Andrea MM, Di Pilato V, Arena F, Conte V, et al. MgrB inactivation is a common mechanism of colistin resistance in KPC-producing Klebsiella pneumoniae of clinical origin. Antimicrob Agents Chemother. 2014;58(10):5696–703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Cannatelli A, D’Andrea MM, Giani T, Di Pilato V, Arena F, Ambretti S, et al. In vivo emergence of colistin resistance in Klebsiella pneumoniae producing KPC-type carbapenemases mediated by insertional inactivation of the PhoQ/PhoP mgrB regulator. Antimicrob Agents Chemother. 2013;57(11):5521–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Pitt ME, Elliott AG, Cao MD, Ganesamoorthy D, Karaiskos I, Giamarellou H, et al. Multifactorial chromosomal variants regulate polymyxin resistance in extensively drug-resistant Klebsiella pneumoniae. Microb Genom. 2018;4:doi: 10.1099/mgen.0.000158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Otter JA, Doumith M, Davies F, Mookerjee S, Dyakova E, Gilchrist M, et al. Emergence and clonal spread of colistin resistance due to multiple mutational mechanisms in carbapenemase-producing Klebsiella pneumoniae in London. Sci Rep. 2017;7(1):12711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Choi MJ, Ko KS. Mutant prevention concentrations of colistin for Acinetobacter baumannii, Pseudomonas aeruginosa and Klebsiella pneumoniae clinical isolates. J Antimicrob Chemother. 2014;69(1):275–7. [DOI] [PubMed] [Google Scholar]
  • 26.Olaitan AO, Diene SM, Kempf M, Berrazeg M, Bakour S, Gupta SK, et al. Worldwide emergence of colistin resistance in Klebsiella pneumoniae from healthy humans and patients in Lao PDR, Thailand, Israel, Nigeria and France owing to inactivation of the PhoP/PhoQ regulator mgrB: an epidemiological and molecular study. Int J Antimicrob Agents. 2014;44(6):500–7. [DOI] [PubMed] [Google Scholar]
  • 27.Cannatelli A, Santos-Lopez A, Giani T, Gonzalez-Zorn B, Rossolini GM. Polymyxin resistance caused by mgrB inactivation is not associated with significant biological cost in Klebsiella pneumoniae. Antimicrob Agents Chemother. 2015;59(5):2898–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Clinical and Laboratory Standards Institute. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, M07-Ed11. CLSI, Wayne, PA, USA: 2018. [Google Scholar]
  • 29.Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M, Kulikov AS, et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol. 2012;19(5):455–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Antipov D, Hartwick N, Shen M, Raiko M, Lapidus A, Pevzner PA. plasmidSPAdes: assembling plasmids from whole genome sequencing data. Bioinformatics. 2016;32(22):3380–7. [DOI] [PubMed] [Google Scholar]
  • 31.Gurevich A, Saveliev V, Vyahhi N, Tesler G. QUAST: quality assessment tool for genome assemblies. Bioinformatics. 2013;29(8):1072–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Rissman AI, Mau B, Biehl BS, Darling AE, Glasner JD, Perna NT. Reordering contigs of draft genomes using the Mauve aligner. Bioinformatics. 2009;25(16):2071–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Seemann T Prokka: rapid prokaryotic genome annotation. Bioinformatics. 2014;30(14):2068–9. [DOI] [PubMed] [Google Scholar]
  • 34.Xie Z, Tang H. ISEScan: automated identification of insertion sequence elements in prokaryotic genomes. Bioinformatics. 2017;33(21):3340–7. [DOI] [PubMed] [Google Scholar]
  • 35.Jia B, Raphenya AR, Alcock B, Waglechner N, Guo P, Tsang KK, et al. CARD 2017: expansion and model-centric curation of the comprehensive antibiotic resistance database. Nucleic Acids Res. 2017;45(D1):D566–D73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Bertelli C, Laird MR, Williams KP, Simon Fraser University Research Computing G, Lau BY, Hoad G, et al. IslandViewer 4: expanded prediction of genomic islands for larger-scale datasets. Nucleic Acids Res. 2017;45(W1):W30–W5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.MLST database for Klebsiella pneumoniae [Internet]. 2018. Available from: http://bigsdb.web.pasteur.fr/klebsiella/.
  • 38.Treangen TJ, Ondov BD, Koren S, Phillippy AM. The Harvest suite for rapid core-genome alignment and visualization of thousands of intraspecific microbial genomes. Genome Biol. 2014;15(11):524. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Stamatakis A RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30(9):1312–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Letunic I, Bork P. Interactive Tree Of Life (iTOL) v4: recent updates and new developments. Nucleic Acids Res. 2019;47(W1):W256–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Liao Y, Smyth GK, Shi W. The Subread aligner: fast, accurate and scalable read mapping by seed-and-vote. Nucleic Acids Res. 2013;41(10):e108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Harrison P, Seemann T. From high-throughput sequencing read alignments to confident, biologically relevant conclusions with Nesoni 2009. Available from: http://www.vicbioinformatics.com/nesoni_ba2009_poster.pdf.
  • 43.Cameron DL, Schroder J, Penington JS, Do H, Molania R, Dobrovic A, et al. GRIDSS: sensitive and specific genomic rearrangement detection using positional de Bruijn graph assembly. Genome Res. 2017;27(12):2050–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Carver T, Harris SR, Berriman M, Parkhill J, McQuillan JA. Artemis: an integrated platform for visualization and analysis of high-throughput sequence-based experimental data. Bioinformatics. 2012;28(4):464–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Page AJ, Cummins CA, Hunt M, Wong VK, Reuter S, Holden MT, et al. Roary: rapid large-scale prokaryote pan genome analysis. Bioinformatics. 2015;31(22):3691–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Choi Y, Chan AP. PROVEAN web server: a tool to predict the functional effect of amino acid substitutions and indels. Bioinformatics. 2015;31(16):2745–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.The European Committee on Antimicrobial Susceptibility Testing. Breakpoint tables for interpretation of MICs and zone diameters, version 8.0 2018. Available from: http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/Breakpoint_tables/v_8.0_Breakpoint_Tables.pdf. [Google Scholar]
  • 48.Carattoli A, Zankari E, Garcia-Fernandez A, Voldby Larsen M, Lund O, Villa L, et al. In silico detection and typing of plasmids using PlasmidFinder and plasmid multilocus sequence typing. Antimicrob Agents Chemother. 2014;58(7):3895–903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Chiu SK, Huang LY, Chen H, Tsai YK, Liou CH, Lin JC, et al. Roles of ramR and tet(A) mutations in conferring tigecycline resistance in carbapenem-resistant Klebsiella pneumoniae clinical isolates. Antimicrob Agents Chemother. 2017;61(8):e00391–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Novovic K, Trudic A, Brkic S, Vasiljevic Z, Kojic M, Medic D, et al. Molecular epidemiology of colistin-resistant, carbapenemase-producing Klebsiella pneumoniae in Serbia from 2013 to 2016. Antimicrob Agents Chemother. 2017;61(5):e02550–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Deleo FR, Chen L, Porcella SF, Martens CA, Kobayashi SD, Porter AR, et al. Molecular dissection of the evolution of carbapenem-resistant multilocus sequence type 258 Klebsiella pneumoniae. Proc Natl Acad Sci U S A. 2014;111(13):4988–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Lascols C, Peirano G, Hackel M, Laupland KB, Pitout JD. Surveillance and molecular epidemiology of Klebsiella pneumoniae isolates that produce carbapenemases: first report of OXA-48-like enzymes in North America. Antimicrob Agents Chemother. 2013;57(1):130–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Guo L, An J, Ma Y, Ye L, Luo Y, Tao C, et al. Nosocomial outbreak of OXA-48-producing Klebsiella pneumoniae in a Chinese hospital: clonal transmission of ST147 and ST383. PLoS One. 2016;11(8):e0160754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Bader MW, Sanowar S, Daley ME, Schneider AR, Cho U, Xu W, et al. Recognition of antimicrobial peptides by a bacterial sensor kinase. Cell. 2005;122(3):461–72. [DOI] [PubMed] [Google Scholar]
  • 55.Mitrophanov AY, Jewett MW, Hadley TJ, Groisman EA. Evolution and dynamics of regulatory architectures controlling polymyxin B resistance in enteric bacteria. PLoS Genet. 2008;4(10):e1000233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Luo SC, Lou YC, Rajasekaran M, Chang YW, Hsiao CD, Chen C. Structural basis of a physical blockage mechanism for the interaction of response regulator PmrA with connector protein PmrD from Klebsiella pneumoniae. J Biol Chem. 2013;288(35):25551–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Giordano C, Barnini S, Tsioutis C, Chlebowicz MA, Scoulica EV, Gikas A, et al. Expansion of KPC-producing Klebsiella pneumoniae with various mgrB mutations giving rise to colistin resistance: the role of ISL3 on plasmids. Int J Antimicrob Agents. 2018;51(2):260–5. [DOI] [PubMed] [Google Scholar]
  • 58.Matamouros S, Hager KR, Miller SI. HAMP domain rotation and tilting movements associated with signal transduction in the PhoQ sensor kinase. mBio. 2015;6(3):e00616–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Kidd TJ, Mills G, Sa-Pessoa J, Dumigan A, Frank CG, Insua JL, et al. A Klebsiella pneumoniae antibiotic resistance mechanism that subdues host defences and promotes virulence. EMBO Mol Med. 2017;9(4):430–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Crawford MA, Fisher DJ, Leung LM, Lomonaco S, Lascols C, Cannatelli A, et al. CXC chemokines exhibit bactericidal activity against multidrug-resistant Gram-negative pathogens. mBio. 2017;8(6):e01549–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Wand ME, Bock LJ, Sutton JM. Retention of virulence following colistin adaptation in Klebsiella pneumoniae is strain-dependent rather than associated with specific mutations. J Med Microbiol. 2017;66(7):959–64. [DOI] [PubMed] [Google Scholar]
  • 62.Bardet L, Baron S, Leangapichart T, Okdah L, Diene SM, Rolain JM. Deciphering Heteroresistance to Colistin in a Klebsiella pneumoniae Isolate from Marseille, France. Antimicrob Agents Chemother. 2017;61(6):e00356–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Zhang H, Zhao D, Shi Q, Quan J, Li X, Yu Y. The mcr-1 gene has no effect on colistin resistance when it coexists with inactivated mgrB gene in Klebsiella pneumoniae. Microb Drug Resist. 2018;24(8):1117–20. [DOI] [PubMed] [Google Scholar]
  • 64.Huerta-Cepas J, Szklarczyk D, Forslund K, Cook H, Heller D, Walter MC, et al. eggNOG 4.5: a hierarchical orthology framework with improved functional annotations for eukaryotic, prokaryotic and viral sequences. Nucleic Acids Res. 2016;44(D1):D286–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Lemmin T, Soto CS, Clinthorne G, DeGrado WF, Dal Peraro M. Assembly of the transmembrane domain of E. coli PhoQ histidine kinase: implications for signal transduction from molecular simulations. PLoS Comput Biol. 2013;9(1):e1002878. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figure

Figure S1. Gene presence/absence variations across the 16 isolates.

NIHMS1590989-supplement-Supplementary_Figure.docx (3.6MB, docx)
Supplementary Tables
NIHMS1590989-supplement-Supplementary_Tables.xlsx (1.3MB, xlsx)

RESOURCES