
==== Front
Microbiology (Reading)
Microbiology (Reading)
micro
micro
Microbiology
1350-0872
1465-2080
Microbiology Society

39230258
001499
10.1099/mic.0.001499
Research Article
Antimicrobials and AMR
Mutations in the efflux regulator gene oqxR provide a simple genetic switch for antimicrobial resistance in Klebsiella pneumoniae
Dawson Catherine J. 12*Catherine.J.Dawson@uon.edu.au

Bartczak Amelia 12amelia.bartczak@uon.edu.au

http://orcid.org/0000-0003-2031-9679
Hassan Karl A. 12*karl.hassan@mq.edu.au;karl.hassan@newcastle.edu.au

1 University of Newcastle, Newcastle, Australia
2 ARC Centre of Excellence in Synthetic Biology, Macquarie University, Sydney, Australia
The authors declare that there are no conflicts of interest.

Catherine J.Dawson, Catherine.J.Dawson@uon.edu.au
Karl A.Hassan, karl.hassan@mq.edu.au;karl.hassan@newcastle.edu.au
Supplement: Four supplementary figures and one supplementary table are available with the online version of this article.

2024
04 9 2024
170 9 00149917 8 2024
22 8 2024
Copyright © 2024 The Authors
2024
https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License. This article was made open access via a Publish and Read agreement between the Microbiology Society and the corresponding author’s institution.

Abstract

Klebsiella pneumoniae is a pathogen of major concern in the global rise of antimicrobial resistance and has been implicated as a reservoir for the transfer of resistance genes between species. The upregulation of efflux pumps is a particularly concerning mechanism of resistance acquisition as, in many instances, a single point mutation can simultaneously provide resistance to a range of antimicrobials and biocides. The current study investigated mutations in oqxR, which encodes a negative regulator of the RND-family efflux pump genes, oqxAB, natively found in the chromosome of K. pneumoniae. Resistant mutants in four K. pneumoniae strains (KP6870155, NTUH-K2044, SGH10, and ATCC43816) were selected from single exposures to 30 µg/mL chloramphenicol and 12 mutants were selected for whole genome sequencing to identify mutations associated with resistance. Resistant mutants generated by single exposures to chloramphenicol, tetracycline, or ciprofloxacin at ≥4 X MIC were replica plated onto all three antibiotics to observe simultaneous cross-resistance to all compounds, indicative of a multidrug resistance phenotype. A variety of novel mutations, including single point mutations, deletions, and insertions, were found to disrupt oqxR leading to significant and simultaneous increases in resistance to chloramphenicol, tetracycline, and ciprofloxacin. The oqxAB-oqxR locus has been mobilized and dispersed on plasmids in many Enterobacteriaceae species and the diversity of these loci was examined to evaluate the evolutionary pressures acting on these genes. Comparison of the promoter regions of oqxR in plasmid-borne copies of the oqxR-oqxAB operon indicated that some constructs may produce truncated versions of the oqxR transcript, which may impact on oqxAB regulation and expression. In some instances, co-carriage of chromosomal and plasmid encoded oqxAB-oqxR was found in K. pneumoniae, implying that there is selective pressure to maintain and expand the efflux pump. Given that OqxR is a repressor of oqxAB, any mutation affecting its expression or function can lead to multidrug resistance. This is in contrast to antibiotic target site mutations that must occur in limited sequence space to be effective and not impact the fitness of the cell. Therefore, oqxR may act as a simple genetic switch to facilitate resistance via OqxAB mediated efflux.

Klebsiella pneumoniae, efflux upregulation, antibiotic resistance, oqxAB, oqxR
http://dx.doi.org/10.13039/501100000923 Australian Research Council FT180100123 Hassan Karl
==== Body
pmcIntroduction

In 2019, antimicrobial resistance (AMR) was associated with approximately 4.95 million deaths worldwide, 1.27 million of which were directly attributed to bacterial AMR, surpassing mortality from diseases such as human immunodeficiency virus infection and malaria [1]. Klebsiella pneumoniae is a pathogen of great concern in the global rise of AMR, with 44% of countries in the WHO European region reporting resistance rates of 50% or above [2]. K. pneumoniae has a remarkable capacity to acquire resistance, and has been implicated as a reservoir for spreading AMR genes with other clinically relevant pathogens [3].

Bacteria employ a number of resistance mechanisms to overcome antibiotic treatment, one of which is export of antimicrobials from the cell via efflux pumps. K. pneumoniae possess a number of intrinsic efflux pumps, major contributors being the RND-family AcrAB and OqxAB pumps that associate with TolC [4]. The OqxAB-TolC pump confers resistance to a range of antibiotics and biocides, including quinolones, tigecycline, ciprofloxacin, nitrofurantoin, tetracycline, benzalkonium chloride, triclosan and SDS [56]. Genes encoding OqxAB are natively located in the chromosome of K. pneumoniae, but were first discovered in a conjugative plasmid in Escherichia coli [7]. Sequencing and phylogenetic analysis has determined that the oqxAB operon probably originated from K. pneumoniae and has been mobilized by IS26-like transposable elements [89]. Transcription of oqxAB is influenced by the global regulator RamA, and by two regulators encoded either side of the oqxAB operon [10]. These are rarA, encoded upstream of oqxAB which activates expression, and oqxR, encoded downstream and involved in repressing oqxAB expression [411]. Mutations in oqxR have been reported to increase expression of the oqxAB pump, leading to increased AMR [4,6, 1113].

The following study reports novel mutations found in oqxR in K. pneumoniae that result in increased resistance to antimicrobial substrates of OqxAB and examines the broader distribution of the oqxAB-oqxR operon in genomes of the Enterobacteriaceae. A mutation at almost any site in the oqxR gene that introduces a frameshift or changes a functionally important amino acid residue, and therefore disrupts its activity as a repressor, allows for increased oqxAB expression and increased resistance. Consequently, oqxR may act as a simple genetic switch to facilitate multidrug resistance, which may be particularly important in the convergence of hypervirulent and multidrug-resistant K. pneumoniae strains.

Methods

Bacterial culturing and strains

Culturing was performed in cation adjusted Mueller-Hinton (MH; Oxoid) or Lennox-LB (LLB; 10 g tryptone, 5 g yeast extract, 5 g NaCl) medium where indicated. Hyper-mucoid strains were grown at 25 °C to reduce capsule production, while all other strains were grown at 37 °C unless otherwise stated. The strains used are listed in Supplementary Material 1 .

Mutant generation

Resistant mutants were generated in K. pneumoniae strains by plating high-density cultures on selective media. Strains of K. pneumoniae were streaked on LLB and incubated overnight. Cells were resuspended from plates in 1× PBS and adjusted to ~1.7×1010 c.f.u. ml–1 (OD600=20). A 25 µl spot corresponding to ~4×108 c.f.u. was transferred to an LLB plate containing 100 µM 2,6-diaminopimelic acid, allowed to dry in a biosafety hood, then incubated at 37 °C for 1 h. Following incubation, spots were scraped from the plate and streaked onto LLB plates containing 30 µg ml−1 chloramphenicol, then incubated overnight. Single colonies were cultured directly from plates under chloramphenicol selection and genomic DNA was extracted for whole genome sequencing.

DNA extraction and whole genome sequencing

Genomic DNA was extracted from liquid cultures using the DNeasy UltraClean Microbial kit (Qiagen). Briefly, 1 ml of overnight culture was used as input and an additional incubation was performed in solution SL (Qiagen) at 70 °C for 10 min prior to vortexing. The rest was performed as per the manufacturer’s instructions. Purity was assessed by nanodrop and concentration quantified by Qubit using the 1× dsDNA Broad Range assay (Thermofisher). Samples were sent for Illumina paired short-read sequencing at SeqCentre (Pittsburgh, USA) using the 200 Mb service. SeqCentre’s methods were as follows: sample libraries were prepared using an Illumina DNA Prep kit and IDT 10 bp UDI indices, and sequenced on an Illumina NextSeq 2000, producing 2×151 bp reads. Demultiplexing, quality control and adapter trimming were performed with bcl-convert v3.9.3.

Data processing and mutation prediction

Quality assessment of raw reads was performed using fastQC v0.12.1 [14] with default settings. Reads were filtered using fastp v0.23.2 [15] with the following parameters: reads were trimmed for adapter sequences (R1: TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG, R2: GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG), 15 bases were trimmed from the front of reads, four from ends, and poly-G tail trimming and poly-X trimming from the 3′ end of reads were also enabled. Mutation prediction was performed using Breseq v0.36.1 [16] using default parameters. Reference sequences were KP6870155 [17], ATCC43816 (Chr: NZ_CP064352), SGH10 (Chr: NZ_CP025080, Plasmid: NZ_CP025081) and NTUH-K2044 (Chr: NC_012731, Plasmid: NC_006625).

Minimal inhibitory concentration assays

MIC assays were conducted according to EUCAST standards [18]. Briefly, twofold serial dilutions of antibiotics (chloramphenicol, tetracycline and ciprofloxacin) were set up in a 96-well plate (Corning) in cation-adjusted MH. Cultures were prepared by resuspending 3–5 colonies from fresh streak plates in cation-adjusted MH and diluted to reach a final inoculum of 5×105 c.f.u. ml−1 in each well. Plates were incubated at 37 °C with shaking (500 r.p.m.) and OD600 read every 10 min in a LogPhase 600 plate reader (Agilent). Three biological replicates, each with three technical replicates, were performed.

Mutation rate determination and generation of mutant pools

Resistance mutation rate from a single exposure to above MIC antimicrobials was determined by plating a high-density culture of K. pneumoniae on selective agar plates. KP6870155 was streaked over the entire surface of an MH plate and incubated overnight. Cells were resuspended from plates in 1× PBS and adjusted to ~1.7×1010 c.f.u. ml–1 (OD600=20). A 30 µl aliquot (~5×108 c.f.u.) was spread onto 15 cm diameter MH agar plates containing antibiotic at 4× MIC of the WT strain (12 mg l−1 chloramphenicol, 4 mg l−1 tetracycline, 0.12 mg l−1 ciprofloxacin). Concurrently, the input culture was diluted 1×10−7 and 100 µl was plated on non-selective agar to determine c.f.u. ml–1. Plates were incubated overnight, protected from light, for 24 h. Plates were then imaged and c.f.u. counted. To determine if resistance was specific to the antibiotic on which mutants were generated, colonies were replica plated in triplicate onto the same panel of antibiotics (chloramphenicol, tetracycline, ciprofloxacin) using sterile filter paper and incubated for a further 24 h protected from light. Plates were imaged, and images were aligned, pseudo-coloured and overlayed using Fiji [19].

Analysis of the oqx operon on plasmid sequences

The oqx operon sequence from KP6870155 (from rarA to oqxR) was queried against the NCBI nucleotide database [20] using blastn and filtered for results that had percentage identity and query coverage of >90% (date accessed 11 March 2023). Given the large number of Klebsiella chromosomal sequences in the database, searches were also performed with exclusion of hits to Klebsiella (taxid: 570) sequences to allow a greater diversity of variants from other taxa to be identified. This resulted in 321 sequence hits which were exported to a hit-table. These were filtered for hits with an alignment length >4900 nt. Using a custom script, ‘entrez_ncbi_gene_slice.sh’, hit sequences including 10 000 bp upstream and downstream (or when this was not possible, the longest available sequence) were downloaded from the NCBI database utilizing the entrez-direct v16.2 tools [21]. Downloaded sequences were converted to fasta format using the any2fasta.pl script (https://github.com/tseemann/any2fasta) and concatenated into a single multifasta file. CD-hit-est v4.8.1 was run on the concatenated sequences with default settings categorizing sequences into 188 clusters [22]. A representative sequence from each cluster was extracted and re-annotated with Prokka v1.14.6 to ensure consistency in annotations [23]. Splitting the sequences into four subsets, Clinker v0.0.27 was used with default settings to create sequence similarity graphs [24]. Graphs were manually inspected and representative sequences selected based on any divergence from the canonical insertion sequence (IS)-flanked oqxAB-oqxR operon previously reported for pOLA52 [7]. Sequences from K. pneumoniae were manually selected. From these representative sequences a focused similarity graph was constructed using Clinker.

Alignment and prediction of oqxA and oqxR promoter sequences

A region inclusive of the gene and 200 bp upstream of the annotated coding sequence was selected. Sequences were aligned using MAFFT v7.515 [25] with the default L-INS-i parameters. Alignments were visualized using Unipro UGENE [26]. Promoter sequences were predicted using BPROM on the Softberry online webserver (http://www.softberry.com/ accessed 14 August 2023).

Identification of Tn6010 in sequences

The sequence for Tn6010 was obtained from pOLA52 (GenBank ID: EU370913.1) as reported by Norman et al. [8]. A custom database was built from the selected sequences using Blast v2.12.0, and the Tn6010 sequence was then queried against the database using default settings.

Data availability

Files and custom scripts used in data analysis are available from the GitHub repository: https://github.com/Cat-Jane/Kp_oqxR_mutants. Genome sequence read data were deposited in the GenBank sequence read archive under accession numbers: SRS22372387–SRS22372402.

Results

Mutations in oqxR promote resistance to chloramphenicol

When constructing K. pneumoniae mutants for other projects, we noted a high frequency of false positive antibiotic-resistant colonies. To identify the cause(s) of spontaneous resistance, mutants were generated by plating K. pneumoniae KP6870155 at high culture densities on chloramphenicol selective media. Six colonies were selected and subjected to whole genome sequencing. Short reads of the chloramphenicol-resistant mutants were assessed for mutations relative to their ancestral parent strain using Breseq [16]. In all resistant strains, mutations were present in the oqxR gene. A variety of mutations were observed including SNPs, 24 bp duplications and a 179 bp deletion (Table 1). To determine if mutations in oqxR were specifically selected in strain KP6870155 or whether similar mutations would be selected in other strains, we isolated chloramphenicol-resistant mutants of three hypervirulent strains of K. pneumoniae, ATCC 43816, SGH10 and NTUH-K2044, using the same approach. The genomes of two resistant colonies from each strain were sequenced, and it was found that they all carried a mutation in oqxR (Table 1). As OqxR is a known negative regulator of the genes encoding the OqxAB efflux pump, it was hypothesized that these mutations were leading to upregulation of the pump to confer resistance. To investigate this further, three KP6870155 strains with distinct oqxR mutations were chosen for further analyses: a single amino acid change (A32D), a 24 bp duplication (24dup) and a point mutation leading to a premature stop codon (C100*).

Table 1. Mutations in oqxR in chloramphenicol-resistant K. pneumoniae

Strain	Position	oqxR mutation	OqxR change	
KP6870155	1 169 748	C→A	A32D (GCC→GAC)	
KP6870155	1 169 748	C→A	A32D (GCC→GAC)	
KP6870155	1 169 859	(CAACGGCTCAATTCATCTTGGCCG)1→2	24 bp duplication → insertion of LGRNGSIH	
KP6870155	1 169 953	C→A	C100* (TGC→TGA)	
KP6870155	1 169 859	(CAACGGCTCAATTCATCTTGGCCG)1→2	24 bp duplication → insertion of LGRNGSIH	
KP6870155	1 169 843	T→C	S64P (TCA→CCA)	
ATCC 43816	5 327 417	G→C	A32P (GCC→CCC)	
ATCC 43816	5 327 529	(CAACGGCTCAATTCATCTTGGCCG)1→2	24 bp duplication → insertion of LGRNGSIH	
SGH10	1 174 013	Δ179 bp	51RDGIIV →51QIGCR*	
SGH10	1 173 875	G→T	R5L (CGC→CTC)	
NTUH-K2044	4 041 038	(ATGAATTGAGCCGTTGCGGCCAAG)1→2	24 bp duplication → insertion of LGRNGSIH	
NTUH-K2044	4 041 038	(ATGAATTGAGCCGTTGCGGCCAAG)1→2	24 bp duplication → insertion of LGRNGSIH	
* indicates stop codon. Underlines highlight DNA bases that were mutated within the codon leading to an amino acid change.

oqxR mutants have increased tolerance to a range of antimicrobials

The MICs of several antibiotics for the selected oqxR mutants were tested to determine if the mutations impacted resistance. Chloramphenicol, tetracycline and ciprofloxacin are known substrates of the OqxAB efflux pump and each fall into a different antibiotic class. The mutants were selected on chloramphenicol and, as expected, each showed high resistance to this antibiotic with the A32D mutant at 66×, and the 24dup and C100* mutants at 133× the MIC of the parental strain (Fig. 1a) . The difference in MIC was less pronounced for tetracycline with all mutant strains showing resistance at 4× the MIC of the parental strain. Ciprofloxacin MICs were 33× higher than the parental strain for all mutants. These values are classed as resistant in reference to EUCAST susceptibility breakpoints, noting that tetracycline has been removed as a recommended agent for treatment in Enterobacterales in the 2023 revision [27]. There were no obvious differences between the strains in the absence of selection in terms of exponential growth rates in MH medium, but the oqxR mutants did appear to have a lower stationary phase density (Fig. 1b) .

Fig. 1. (a) MIC of KP6870155 WT and oqxR mutant strains. (b) Growth curves of KP6870155 WT and various oqxR mutants, conducted in Muller-Hinton media. Error bars represent sd, average of three biological replicates.

Mutation rate determination

To determine the rate of antibiotic-resistant mutants selected by one exposure to antibiotic, high-density cultures were plated on media containing 4× MIC of chloramphenicol, tetracycline and ciprofloxacin, and the number of resistant c.f.u. relative to the input were counted. It was also of interest whether different antibiotics would select for greater numbers of mutants than others. From an input culture of 1.74×1010 c.f.u. ml−1 (sd=1.94×109), a similar number of mutants arose under selection for each antibiotic at ~1×104 c.f.u. ml–1 (Fig. 2a), suggesting there was no significant difference between the antibiotics tested in terms of promoting mutant selection at the concentrations used. To help distinguish between mutations in oqxR-oqxAB and other mutations, such as tetracycline and ciprofloxacin target site mutations [2829], colonies that arose on the first selective plate were replica plated onto other antibiotics. Resistance to all antibiotics is indicative of a multidrug-resistant phenotype like that conferred by OqxAB, whereas antibiotic-specific resistance could have arisen through target site mutation. Almost all colonies (>99%) were also able to grow on the other antibiotics tested, suggesting that resistance may be due to efflux pump upregulation rather than a specific target modification or enzymatic antibiotic inactivation (Fig. 2b).

Fig. 2. (a) Number of resistant mutants that arose from a single exposure to antibiotic selection at 4× the MIC, relative to the WT input culture of 1.74×1010 c.f.u. ml−1. (b) Replica plates of resistant mutants across different antibiotics taken from the initial plate. Images have been pseudo-coloured and overlayed in the final panel.

Sequence conservation of the oqxAB-oqxR operon between chromosome and plasmids

Aside from K. pneumoniae chromosomes, the oqxAB-oqxR genes are found on plasmids carried by Enterobacteriaceae hosts including Escherichia coli, Shigella flexneri and Salmonella enterica [69]. Given how readily chromosomal oqxR mutants can be selected by antibiotics in K. pneumoniae, it was of interest to investigate the conservation of oqxAB-oqxR in these plasmids, which may demonstrate high levels of diversity. There were a large number of non-Klebsiella genome sequences that contained the oqxAB-oqxR operon (n=321), so sequences were clustered based on similarity and sequences representative of the major clusters that differed from the oqxAB-oqxR structure previously reported for pOLA52 [7] were selected for inspection by amino acid sequence similarity graphs using Clinker (Fig. 3) . The oqxR regulator gene sequence was consistently found in plasmids carrying the pump, but the positive regulator gene rarA does not appear to have been mobilized onto plasmids. This may be due to RarA being involved in global regulation [1011] and therefore carriage on plasmids may disrupt regular cellular function and reduce fitness.

Fig. 3. Comparison of oqxAB-oqxR sequences from various Enterobacteriaceae. (a) Alignment of the oqxR promoter region in chromosomal and plasmid sequences. A representative sequence was chosen from each promoter grouping; the coloured boxes correspond to plasmids in (b) that had the same promoter region. The K. pneumoniae MGH78578 and St pHK06-53 sequences published by Chan et al. [30] are included for comparison. Highlighted are the predicted −35 and −10 promoter sequences (purple), transcription start site (yellow), transposon stop codon (red) and oqxR start codon (orange). Kp=K. pneumoniae, St=Salmonella typhimurium, Ec=E. coli. (b) Comparison of oqxAB-oqxR sequences and their genomic context. The IS sequences are shown in dark red. Coloured boxes indicate sequences with identical oqxR promoter regions and how they relate to (a). Sequences that contain Tn6010 are indicated by a grey star, and sequences from the same chromosome are indicated by orange circles.

Several K. pneumoniae strains were found to be carrying oqxAB-oqxR both on a plasmid and on the chromosome. In one instance, K. pneumoniae strain 555, the plasmid copy of oqxAB was frameshifted and may no longer be functional, but this may be a sequencing artefact (Fig. S1, available in the online version of this article). In other instances of co-carriage (strains KP32558 and Nord9 R85), the chromosomal and plasmid oqxAB sequences were highly similar (identity ≥99%) with no frameshift mutations, suggesting that both copies are functional. However, the oqxR gene on Nord9 pR85_1 is truncated, suggesting that the regulator is non-functional (Fig. S2). The oqxR gene in pR85_1 is adjacent to an IS4 and a second copy of this insertion sequence is encoded upstream of the oqxAB genes. Therefore, the truncation of oqxR may have resulted from the actions of IS4, which may have mobilized the cluster, but only partially captured oqxR.

Comparisons at the amino acid level between chromosomally and plasmid encoded sequences of OqxR revealed that most (15/19) were identical to that of WT OqxR (Fig. S3). The chromosomally encoded OqxR in the Nord 9 strain (CP091582.1) has a single amino acid change, V130A, which may result in a similar resistance phenotype to the spontaneous mutants discovered in this study. As noted above, this strain also carries a plasmid with a truncated sequence of oqxR which, in combination with the point mutation in the chromosomal copy of oqxR, suggests that OqxAB expression may be high in this strain. The OqxR sequences for K. pneumoniae strain 555, K. pneumoniae KP32558 and pKP14052-MCR-1 (CP043932.1, CP076030.1 and MH715960.1) have an amino acid change at S15R which similarly may lead to oqxAB de-repression. Strain 555 carries a frameshifted plasmid copy of oqxAB.

Given that the majority of sequences for oqxR were identical to the WT sequence, it was of interest to determine if there were any differences in the promoter/operator regions between plasmid and chromosomal copies of the oqxAB-oqxR operon. We therefore examined the regions upstream of representative oqxA and oqxR loci to identify differences in their promoter/operator regions compared to the canonical chromosomal sequence. The sequences upstream of oqxA were identical between plasmid and chromosomal copies, suggesting that the binding sites for regulators and the transcription start site (TSS) are unaltered (Fig. S4).

In contrast, there were significant variations in the promoter regions of the oqxR loci. Each of the chromosomal sequences had the same promoter region, identical to that previously reported by Chan et al. [30] with a predicted TSS 62 bp upstream of the oqxR gene (Fig. 3a). The plasmid oqxR promoter/operator regions clustered into three groups based on sequence similarity, with two further sequences that did not fall into any of these groups. There was no correlation of groups with species origin, since plasmids from E. coli, Salmonella enterica and K. pneumoniae fell into each cluster, suggesting inter-species mobility. Many predicted oqxR promoter sequences on the plasmids matched those reported previously [30], but there were some distinct differences. Plasmids falling into the group with E. coli pOLA52 differed from the Salmonella enterica Typhimurium pHK06-53 grouping by a single nucleotide change C→T located 38 bases upstream of the oqxR start codon. This resulted in the promoter sequence predicted by BPROM to be upstream of the start codon, which would result in a full oqxR transcript being produced (Fig. 3a), unlike pHK06-53 [30]. Similarly, the BPROM predicted promoter sequence for Salmonella enterica Typhimurium pNDN5_LS002 was located well upstream of the oqxR start codon, which would also imply that the oqxR transcript produced would include the full-length oqxR gene. In most cases the predicted promoter sequences are yet to be verified experimentally, but these findings suggest that in at least some instances of plasmid carriage of the oqxAB-oqxR operon, the OqxR regulator could be fully functional.

Previous studies have suggested that oqxAB-oqxR can be mobilized by IS26-like transposases [9]. In the vast majority of plasmids oqxAB-oqxR are flanked by IS26 in Tn6010. Tn6010 was first identified in pOLA52 [8] and is found in a number of other sequences (Fig. 3b, grey stars). In one instance, Tn6010 was present on the chromosome, in K. pneumoniae strain STEFF_2, which also had the native oqx operon intact (Fig. 3b, orange circles). K. pneumoniae STEFF_2 carries two plasmids, neither of which carries the oqx operon or IS26. Tn6010 is not found in all plasmids carrying oqxAB-oqxR, suggesting these genes may have been acquired via an alternative element, or that multiple re-arrangement events have taken place modifying the initial Tn6010 sequence. To highlight this diversity, a select number of representative sequences were included in Fig. 3b. In some plasmids ISs do flank oqxAB-oqxR but are not directly adjacent to the operon on one or both sides. Overall, the array of plasmids containing a copy of the operon is quite diverse, suggesting high levels of selection for oqxAB-oqxR carriage on plasmids.

Discussion

While efflux pumps play an important role in intrinsic resistance and other cellular processes, it is typically their regulators that determine efflux pump expression levels, and thus determine their contribution to resistance to a particular antimicrobial [31]. Hypervirulent strains of K. pneumoniae NTUH-K2044, ATCC 43816 and SGH10 were included in our analysis as it was of interest to determine whether mutations relating to efflux upregulation would be selected in these strains, potentially increasing their multidrug resistance profile. The convergence of hypervirulent and multidrug-resistant K. pneumoniae strains, particularly in Asian regions, is of great concern [3234]. Furthermore, mutations in oqxR have been highlighted as a potential contributor to the success of multidrug-resistant isolates of K. pneumoniae (e.g. Clonal Group 258) [12].

As demonstrated in this study, a wide range of mutations in oqxR can facilitate the switch from susceptible to resistant for a number of antibiotics. This is consistent with previous observations of mutations in oqxR leading to upregulation of oqxAB and subsequently increased resistance [411 35]. Whether mutations in oqxR arise spontaneously under selective pressure or are present in the population at low frequencies and become enriched under selection is difficult to determine. There was no overlap between the mutations in oqxR found experimentally here and those previously reported, suggesting that effectively any mutation causing loss of function can be selected, rather than those in specific nucleotide locations. As such, oqxR mutation could be a simple genetic switch providing resistance in K. pneumoniae. It is important to report the breadth of mutations in regulators that can lead to increased levels of resistance, as this is fundamental to increasing the reliability of predicting resistance profiles from whole genome sequencing [35].

Whereas intrinsic oqxAB expression can provide low to intermediate levels of resistance, overexpression of oqxAB is required for strains to be considered non-susceptible, especially to ciprofloxacin [3036]. Increased efflux pump expression can be associated with a fitness cost, although modifications in metabolic pathways can compensate for the negative effects of efflux pump overexpression [37]. We did not observe any significant fitness cost for oqxR mutations as measured by growth curves compared to the WT, but this was in rich lab media and may not reflect the environmental conditions of an active infection. A clinical isolate of K. pneumoniae with a mutation in oqxR resulting in upregulation of OqxAB was more virulent in a Caenorhabditis elegans model compared to the same strain without a mutation in oqxR, suggesting that fitness of the OqxAB overproducing strain was not significantly impacted [4]. Given the widespread dispersion of oqxAB-oqxR onto various plasmids, this would suggest that there may be selective pressure to maintain the efflux pump, and therefore the benefit must outweigh any potential cost.

It is possible that not all of the resistant mutants that were observed in the replica plating experiments harboured mutations in oqxR. For instance, mutations in the quinolone resistance-determining regions of gyrA, gyrB, parC and parE are commonly associated with resistance to ciprofloxacin [38]. However, these mutations would not confer resistance to either tetracycline or chloramphenicol and the majority of colonies when replica plated were able to survive on all three antibiotics, suggesting a non-specific mechanism of resistance. Furthermore, mutations in topoisomerase enzyme genes must occur within a relatively restricted region and at specific sites to maintain function of the enzyme but prevent quinolone binding. In contrast, a mutation at almost any site in oqxR that introduces a frameshift or changes a functionally important amino acid residue in the encoded protein will allow for increased oqxAB expression and therefore a change in resistance.

While the oqxAB-oqxR operon was frequently identified on plasmids and found to be highly conserved (nucleotide sequence identity >99%), significant diversity was identified in the promoter region upstream of oqxR. Chan et al. [30] identified that in Salmonella enterica subsp. Typhimurium strain PY1, expression of the oqxR-oqxAB operon cloned from the plasmid pHK06-53 led to increased resistance to ciprofloxacin. In contrast, expression of oqxR-oqxAB cloned from the chromosome of K. pneumoniae strain MGH78578 did not result in increased ciprofloxacin resistance. They identified that the promoter region of oqxR was truncated in the pHK06-53 plasmid-borne copy of the oqxR-oqxAB operon and showed that this resulted in transcription of oqxR from an alternative start site that reduced its level of expression and shortened the OqxR protein sequence. This reduced its binding capacity as a negative regulator, which in turn reduced repression of the oqxAB operon [30]. The authors hypothesized that this truncation was the result of the promoter being partially captured in a transposon, Tn6010 [30]. The protein sequence of the truncated OqxR is yet to be experimentally verified, but if the predicted TSS is correct, this would lead to the omission of at least 12 aa from the N-terminus of the protein. It is interesting to note that the alternative promoter and TSS proposed by Chan et al. [30] is present in both chromosomal and plasmid-based oqxR sequences (Fig. 3a), so could also direct transcription in the native chromosomal context. Presumably, it would not be beneficial to produce full-length and truncated versions of OqxR from the chromosomally encoded operon, so there may be additional unknown mechanisms of regulation to prevent this.

In the current work, we found evidence to suggest that at least some plasmid sequences of oqxR, including pNDM5_LS002 found in Salmonella enterica subsp. Typhimurium, could produce a full-length transcript and subsequently fully functional OqxR. Furthermore, some sequences that contained Tn6010, such as E. coli pOLA52, E. coli pWP2-W18-CRE-03_1 and K. pneumoniae STEFF_2, had predicted promoters that would also produce a full-length oqxR transcript due to a single point mutation in the region between IS26 and the oqxR start codon. This raises questions around the regulation and expression level of the OqxAB pump in these circumstances and what the contribution to resistance would be.

There were a number of K. pneumoniae strains that carried both a chromosomal and plasmid encoded copy of the oqxAB-oqxR operon. One study that investigated the presence of oqxAB in ESBL (extended-spectrum beta-lactamase) K. pneumoniae strains found that co-carriage of oqxAB on both the chromosome and plasmids occurred in 13% of isolates tested (n=114) [36]. The reason for this is unclear, but it has been shown that expression of oqxAB from plasmids can increase by >80-fold [9], so this may be an alternative way of gaining resistance from OqxAB expression without selecting for mutations in the chromosomally located oqxR gene. It is also possible that regulators from the chromosomally encoded operon could interact with the plasmid copy of oqxAB, or vice versa. Additionally, co-carriage could provide a source for recombination to interchange chromosomal and plasmid copies of the operon. This could enable the fine-tuning of OqxAB expression in response to environmental challenges.

The inherent danger of efflux pumps to infection control is that a single mutation can lead to broad resistance to a range of antimicrobials. The oqxR regulator clearly plays an important role in determining the resistance profile associated with oqxAB pump expression. In essence, the OqxAB pump has a fairly similar range of substrate recognition to that of AcrAB, an RND pump broadly present in Enterobacteriaceae, the primary difference being that acrAB is carried natively in the chromosome of most lineages, whereas oqxAB is generally restricted to K. pneumoniae and Enterobacter sp. [9]. To date, there is no documented carriage of acrAB on plasmids, but there is clear evidence of mobilization of oqxAB broadly on plasmids. It is interesting that a pump that presumably only has utility when combined with a TolC outer membrane component would be mobilized and transferred between strains on a plasmid, and could potentially explain the limited host range of the oqxAB operon as it may only have utility in strains that possess TolC. It would be interesting to know if OqxAB competes with AcrAB for TolC attachment. Indeed, membrane facilitator proteins from various multi-component efflux transporters have been shown to have differing affinities for TolC, probably contributing to the recruitment and stability of pump-TolC complexes [39]. Both OqxAB and AcrAB have been shown to contribute to tigecycline, nitrofurantoin and ciprofloxacin resistance in K. pneumoniae [4042]. The interplay between these efflux pumps warrants further investigation, and the spread of oqxAB should be monitored due to its potential contribution to multidrug resistance.

supplementary material

10.1099/mic.0.001499 Supplementary Material 1.

Acknowledgements

We would like to thank Francesca Short for kindly providing the K. pneumoniae strains NTUH-K2044, SGH10 and ATCC43816.

Abbreviations

AMR antimicrobial resistance

CFU colony forming unit

LLB Lennox-LB

MH Mueller-Hinton

MIC minimal inhibitory concentration

PBS phosphate buffered saline

RND resistance nodulation division

TSS transcription start site

WHO World Health Organisation

WT wild type

Funding: K.A.H. has been supported by an Australian Research Council Future Fellowship (FT180100123). C.J.D. and A.B. are each supported by an Australian Government Research Training Programme (RTP) Scholarship.

Accession No: Genome sequence read data generated in this project have been deposited in the GenBank sequence read archive in BioProject PRJNA1148794 (sequential SRA accessions: SRS22372387–SRS22372402).
==== Refs
References

1. Murray CJL Ikuta KS Sharara F Swetschinski L Robles Aguilar G et al Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis Lancet 2022 399 629 655 10.1016/S0140-6736(21)02724-0 35065702
2. ECDC, WHO Antimicrobial resistance surveillance in Europe 2022 - 2020 data. European Centre for Disease Prevention and Control 2022 n.d https://www.ecdc.europa.eu/en/publications-data/antimicrobial-resistance-surveillance-europe-2022-2020-data accessed 24-November-2022
3. Wyres KL Holt KE Klebsiella pneumoniae as a key trafficker of drug resistance genes from environmental to clinically important bacteria Curr Opin Microbiol 2018 45 131 139 10.1016/j.mib.2018.04.004 29723841
4. Bialek-Davenet S Lavigne J-P Guyot K Mayer N Tournebize R et al Differential contribution of AcrAB and OqxAB efflux pumps to multidrug resistance and virulence in Klebsiella pneumoniae J Antimicrob Chemother 2015 70 81 88 10.1093/jac/dku340 25193085
5. Hansen LH Jensen LB Sørensen HI Sørensen SJ Substrate specificity of the OqxAB multidrug resistance pump in Escherichia coli and selected enteric bacteria J Antimicrob Chemother 2007 60 145 147 10.1093/jac/dkm167 17526501
6. Li J Zhang H Ning J Sajid A Cheng G et al The nature and epidemiology of OqxAB, a multidrug efflux pump Antimicrob Resist Infect Control 2019 8 44 10.1186/s13756-019-0489-3 30834112
7. Sørensen AH Hansen LH Johannesen E Sørensen SJ Conjugative plasmid conferring resistance to olaquindox Antimicrob Agents Chemother 2003 47 798 799 10.1128/AAC.47.2.798-799.2003 12543696
8. Norman A Hansen LH She Q Sørensen SJ Nucleotide sequence of pOLA52: a conjugative IncX1 plasmid from Escherichia coli which enables biofilm formation and multidrug efflux Plasmid 2008 60 59 74 10.1016/j.plasmid.2008.03.003 18440636
9. Wong MHY Chan EWC Chen S Evolution and dissemination of OqxAB-like efflux pumps, an emerging quinolone resistance determinant among members of Enterobacteriaceae Antimicrob Agents Chemother 2015 59 3290 3297 10.1128/AAC.00310-15 25801572
10. Jiménez-Castellanos J-C Wan Ahmad Kamil WNI Cheung CHP Tobin MS Brown J et al Comparative effects of overproducing the AraC-type transcriptional regulators MarA, SoxS, RarA and RamA on antimicrobial drug susceptibility in Klebsiella pneumoniae J Antimicrob Chemother 2016 71 1820 1825 10.1093/jac/dkw088 27029850
11. Veleba M Higgins PG Gonzalez G Seifert H Schneiders T Characterization of RarA, a novel AraC family multidrug resistance regulator in Klebsiella pneumoniae Antimicrob Agents Chemother 2012 56 4450 4458 10.1128/AAC.00456-12 22644028
12. Bowers JR Kitchel B Driebe EM MacCannell DR Roe C et al Genomic analysis of the emergence and rapid global dissemination of the clonal group 258 Klebsiella pneumoniae pandemic PLoS One 2015 10 e0133727 10.1371/journal.pone.0133727 26196384
13. Wan Nur Ismah WAK Takebayashi Y Findlay J Heesom KJ Avison MB Impact of OqxR loss of function on the envelope proteome of Klebsiella pneumoniae and susceptibility to antimicrobials J Antimicrob Chemother 2018 73 2990 2996 10.1093/jac/dky293 30053019
14. Andrews S FastQC: a quality control tool for high throughput sequence data 2010 http://www.bioinformatics.babraham.ac.uk/projects/fastqc
15. Chen S Zhou Y Chen Y Gu J fastp: an ultra-fast all-in-one FASTQ preprocessor Bioinformatics 2018 34 i884 i890 10.1093/bioinformatics/bty560 30423086
16. Deatherage DE Barrick JE Identification of mutations in laboratory-evolved microbes from next-generation sequencing data using breseq Methods Mol Biol 2014 1151 165 188 10.1007/978-1-4939-0554-6_12 24838886
17. Semenec L Cain AK Dawson CJ Liu Q Dinh H et al Cross-protection and cross-feeding between Klebsiella pneumoniae and Acinetobacter baumannii promotes their co-existence Nat Commun 2023 14 702 10.1038/s41467-023-36252-2 36759602
18. EUCAST Determination of minimum inhibitory concentrations (MICs) of antibacterial agents by broth dilution Clin Microbiol Infect 2003 9 ix xv 10.1046/j.1469-0691.2003.00790.x
19. Schindelin J Arganda-Carreras I Frise E Kaynig V Longair M et al Fiji: an open-source platform for biological-image analysis Nat Methods 2012 9 676 682 10.1038/nmeth.2019 22743772
20. Johnson M Zaretskaya I Raytselis Y Merezhuk Y McGinnis S et al NCBI BLAST: a better web interface Nucleic Acids Res 2008 36 W5 9 10.1093/nar/gkn201 18440982
21. Kans J Entrez direct: E-utilities on the Unix Command Line National Center for Biotechnology Information (US) 2023 https://www.ncbi.nlm.nih.gov/books/NBK179288 accessed 13-April-2023
22. Li W Godzik A Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences Bioinformatics 2006 22 1658 1659 10.1093/bioinformatics/btl158 16731699
23. Seemann T Prokka: rapid prokaryotic genome annotation Bioinformatics 2014 30 2068 2069 10.1093/bioinformatics/btu153 24642063
24. Gilchrist CLM Chooi Y-H clinker & clustermap.js: automatic generation of gene cluster comparison figures Bioinformatics 2021 37 2473 2475 10.1093/bioinformatics/btab007 33459763
25. Katoh K Misawa K Kuma K Miyata T MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform Nucleic Acids Res 2002 30 3059 3066 10.1093/nar/gkf436 12136088
26. Okonechnikov K Golosova O Fursov M UGENE team Unipro UGENE: a unified bioinformatics toolkit Bioinformatics 2012 28 1166 1167 10.1093/bioinformatics/bts091 22368248
27. EUCAST Breakpoint tables for interpretation of MICs and zone diameters. Version 13.0 2023 https://www.eucast.org/clinical_breakpoints accessed 23-January-2023
28. Grossman TH Tetracycline antibiotics and resistance Cold Spring Harb Perspect Med 2016 6 a025387 10.1101/cshperspect.a025387 26989065
29. Zhao X Drlica K Restricting the selection of antibiotic-resistant mutants: a general strategy derived from fluoroquinolone studies Clin Infect Dis 2001 33 S147 S156 10.1086/321841 11524712
30. Chan BK-W Wong MH-Y Chan EW-C Chen S Transcriptional regulation and functional characterization of the plasmid-borne oqxAB genes in Salmonella typhimurium Microbiol Spectr 2022 10 e0217021 10.1128/spectrum.02170-21 35315694
31. Webber MA Piddock LJV The importance of efflux pumps in bacterial antibiotic resistance J Antimicrob Chemother 2003 51 9 11 10.1093/jac/dkg050 12493781
32. Rafiq Z Sam N Vaidyanathan R Whole genome sequence of Klebsiella pneumoniae U25, a hypermucoviscous, multidrug resistant, biofilm producing isolate from India Mem Inst Oswaldo Cruz 2016 111 144 146 10.1590/0074-02760150423 26872343
33. Wyres KL Nguyen TNT Lam MMC Judd LM van Vinh Chau N et al Genomic surveillance for hypervirulence and multi-drug resistance in invasive Klebsiella pneumoniae from South and Southeast Asia Genome Med 2020 12 11 10.1186/s13073-019-0706-y 31948471
34. Lan P Jiang Y Zhou J Yu Y A global perspective on the convergence of hypervirulence and carbapenem resistance in Klebsiella pneumoniae J Glob Antimicrob Resist 2021 25 26 34 10.1016/j.jgar.2021.02.020 33667703
35. Wan Nur Ismah WAK Takebayashi Y Findlay J Heesom KJ Jiménez-Castellanos J-C et al Prediction of fluoroquinolone susceptibility directly from whole-genome sequence data by using liquid chromatography-tandem mass spectrometry to identify mutant genotypes Antimicrob Agents Chemother 2018 62 e01814-17 10.1128/AAC.01814-17 29263066
36. Rodríguez-Martínez JM Díaz de Alba P Briales A Machuca J Lossa M et al Contribution of OqxAB efflux pumps to quinolone resistance in extended-spectrum-β-lactamase-producing Klebsiella pneumoniae J Antimicrob Chemother 2013 68 68 73 10.1093/jac/dks377 23011289
37. Olivares Pacheco J Alvarez-Ortega C Alcalde Rico M Martínez JL Metabolic compensation of fitness costs is a general outcome for antibiotic-resistant Pseudomonas aeruginosa mutants overexpressing efflux pumps mBio 2017 8 e00500-17 10.1128/mBio.00500-17 28743808
38. Correia S Poeta P Hébraud M Capelo JL Igrejas G Mechanisms of quinolone action and resistance: where do we stand? J Med Microbiol 2017 66 551 559 10.1099/jmm.0.000475 28504927
39. Tikhonova EB Dastidar V Rybenkov VV Zgurskaya HI Kinetic control of TolC recruitment by multidrug efflux complexes Proc Natl Acad Sci USA 2009 106 16416 16421 10.1073/pnas.0906601106 19805313
40. Albarri O AlMatar M Öcal MM Köksal F Overexpression of efflux pumps AcrAB and OqxAB contributes to ciprofloxacin resistance in clinical isolates of K. pneumoniae CPPS 2022 23 356 368 10.2174/1389203723666220630162920
41. Xu Q Jiang J Zhu Z Xu T Sheng Z-K et al Efflux pumps AcrAB and OqxAB contribute to nitrofurantoin resistance in an uropathogenic Klebsiella pneumoniae isolate Int J Antimicrob Agents 2019 54 223 227 10.1016/j.ijantimicag.2019.06.004 31200021
42. Zhong X Xu H Chen D Zhou H Hu X et al First emergence of acrAB and oqxAB mediated tigecycline resistance in clinical isolates of Klebsiella pneumoniae pre-dating the use of tigecycline in a Chinese hospital PLoS One 2014 9 e115185 10.1371/journal.pone.0115185 25503276
