Impact of qnrA1, qnrB1 and qnrS1 on the efficacy of ciprofloxacin and levofloxacin in an experimental pneumonia model caused by Escherichia coli with or without the GyrA mutation Ser83Leu.

OBJECTIVES: The aim of this study was to evaluate the impact of qnrA1, qnrB1 and qnrS1 on the in vivo efficacies of ciprofloxacin and levofloxacin in an experimental model of pneumonia caused by Escherichia coli. METHODS: Two isogenic groups of E. coli transformants, based on two ATCC 25922 strains, with or without the GyrA mutation Ser83Leu, and carrying qnrA1, qnrB1 or qnrS1, were used in an experimental pneumonia model. The efficacies of ciprofloxacin (40 mg/kg/day) and levofloxacin (50 and 150 mg/kg/day) were evaluated. RESULTS: For the pneumonia caused by the parental strains lacking qnr genes, both fluoroquinolones significantly (P<0.05) reduced the bacterial lung concentration by >7 log10 cfu/g against E. coli ATCC/pBK and between 5.09 and 6.34 log10 cfu/g against E. coli ATCC-S83L/pBK. The presence of any qnr genes in the strains of both isogenic groups diminished the reduction of bacterial lung concentration with any therapy (P<0.05). Furthermore, all therapeutic schemes reduced the percentage of positive blood cultures in both isogenic groups (P<0.05). Finally, the survival results suggest a higher mortality with the strains expressing qnr genes. CONCLUSIONS: The presence of qnrA1, qnrB1 and qnrS1 in E. coli reduced the efficacy of ciprofloxacin and levofloxacin in a murine pneumonia model.

Contribution of OqxAB efflux pumps to quinolone resistance in extended-spectrum-ß-lactamase-producing Klebsiella pneumoniae.

OBJECTIVES: The aims of this study were to analyse the presence of oqxA and oqxB genes in a collection of extended-spectrum ß-lactamase (ESBL)-producing Klebsiella pneumoniae strains, to determine their chromosomal and/or plasmidic locations and to analyse expression levels in relation to susceptibility or resistance to quinolones. METHODS: A collection of 114 non-repetitive isolates of ESBL-producing K. pneumoniae was used. K. pneumoniae ATCC 27799 and K. pneumoniae ATCC 700603 were also included. Detection of oqxA and oqxB genes was performed by PCR. Testing for chromosomal and/or plasmidic location was carried out using plasmid DNA and subsequent hybridization. oqxA gene expression was analysed using real-time RT-PCR. Transfer of the plasmid-encoded OqxAB was evaluated. RESULTS: The prevalence of both oqxA and oqxB detected in K. pneumoniae was high: 76% and 75%, respectively. Hybridization assays showed that oqxA (16%) and oqxB (13%) were simultaneously present in locations on the chromosome and on large plasmids. The plasmids were transferable by transformation into K. pneumoniae. RT-PCR assays showed higher expression (4-fold) in strains with reduced susceptibility to quinolones than in susceptible strains. Interestingly, K. pneumoniae ATCC 700603 showed an 18-fold higher expression than K. pneumoniae ATCC 27799. These differences were in accordance with quinolone susceptibility. CONCLUSIONS: The prevalence of the OqxAB efflux pump (both chromosomal and plasmid encoded) in ESBL-producing K. pneumoniae is high in Spain and represents a potential reservoir for the spread of these genes. High expression of this pump contributes to reduced susceptibility to quinolones in clinical isolates of ESBL-producing K. pneumoniae.

Exposure to diverse antimicrobials induces the expression of qnrB1, qnrD and smaqnr genes by SOS-dependent regulation.

OBJECTIVES: Direct SOS-dependent regulation of qnrB genes by fluoroquinolones mediated by LexA was reported. The smaqnr gene, on the Serratia marcescens chromosome, and qnrD both contain a putative LexA box. The aim of this study was to evaluate whether smaqnr or qnrD genes are induced via SOS-dependent mechanisms, and to investigate whether other antimicrobial agents induce qnrB, qnrD and smaqnr expression. METHODS: RT-PCR was used to evaluate qnrB1, qnrD and smaqnr expression. Different concentrations of ciprofloxacin, levofloxacin, moxifloxacin and ceftazidime were evaluated as inducers. Additionally, the promoter regions of qnrB1, qnrD and smaqnr were fused transcriptionally to green fluorescent protein and used in reporter gene assays. Disc diffusion assays with different antimicrobial agents were used to detect induction. Measurements of transcriptional induction by ciprofloxacin were carried out using a plate reader. RESULTS: RT-PCR assays showed that qnrB1, qnrD and smaqnr were induced at different concentrations of ciprofloxacin, moxifloxacin, levofloxacin and ceftazidime, increasing transcription 1.5- to 16.3-fold compared with basal expression, and depending on the antimicrobial agent and promoter analysed. The reporter gene assays showed that the qnrB1, qnrD and smaqnr genes were induced by ciprofloxacin, as expected, but also by ceftazidime, ampicillin and trimethoprim in Escherichia coli wild-type strains, but not in the recA-deficient E. coli HB101. Induction was not evident for imipenem or gentamicin. CONCLUSIONS: ß-Lactams and trimethoprim, along with fluoroquinolones, induce transcription of qnrB, qnrD and smaqnr genes using SOS-dependent regulation. These results show the direct SOS-dependent regulation of a low-level fluoroquinolone resistance mechanism in response to other antimicrobials.

In vitro effect of qnrA1, qnrB1, and qnrS1 genes on fluoroquinolone activity against isogenic Escherichia coli isolates with mutations in gyrA and parC.

This article provides an analysis of the in vitro effect of qnrA1, qnrB1, and qnrS1 genes, combined with quinolone-resistant Ser83Leu substitutions in GyrA and/or Ser80Arg in ParC, on fluoroquinolone (FQ) resistance in isogenic Escherichia coli strains. The association of Ser83Leu substitution in GyrA, Ser80Arg substitution in ParC, and qnr gene expression increased the MIC of ciprofloxacin to 2 µg/ml. qnr genes present in E. coli that harbored a Ser83Leu substitution in GyrA increased mutant prevention concentration (MPC) values to 8 to 32 µg/ml. qnr gene expression in E. coli may play an important role in selecting for one-step FQ-resistant mutants.

Smaqnr, a new chromosome-encoded quinolone resistance determinant in Serratia marcescens.

OBJECTIVES: A new pentapeptide repeat (PRP) protein, named SmaQnr, from the clinically relevant species Serratia marcescens, which decreased susceptibility to quinolones when expressed in Escherichia coli, is reported herein. METHODS: In silico analysis revealed the presence of a gene encoding a Qnr-like protein that shares 80% amino acid identity with QnrB1 in the S. marcescens strain Db11. Fragments carrying the coding region and the upstream non-coding sequences of eight clinical isolates were cloned and expressed in E. coli. MIC values of quinolones were determined. RT-PCR was used to study expression of these genes in their natural host. Southern hybridization was used to explore the presence of the gene in the genus Serratia. RESULTS: Recombinant plasmids encoding SmaQnr reduced susceptibility to fluoroquinolones and nalidixic acid in both E. coli ATCC 25922 and DH10B. Sequences upstream of these genes contain a LexA box. Conventional RT-PCR showed transcription of the analysed Smaqnr genes in their natural hosts. Southern blot analysis suggests the presence of similar genes in several species of the genus Serratia. CONCLUSIONS: SmaQnr conferred a reduced susceptibility phenotype against fluoroquinolones in E. coli. These data provide evidence of its possible role in quinolone resistance in S. marcescens. This Gram-negative species may constitute a reservoir for qnr-like quinolone resistance genes.

Mutational analysis of quinolone resistance in the plasmid-encoded pentapeptide repeat proteins QnrA, QnrB and QnrS.

OBJECTIVES: Pentapeptide repeat proteins (PRPs) QnrA, QnrB and QnrS confer reduced susceptibility to quinolones. This study presents an in vitro analysis of the genetic evolution of quinolone resistance mediated by changes in the coding sequences and promoter regions of qnrA1, qnrS1 and qnrB1 genes. METHODS: A random mutagenesis technique was used to predict the evolutionary potential of these PRPs against nalidixic acid and fluoroquinolones. After comparing the amino acid sequences of these and other PRPs protecting bacteria from quinolone activity, several conserved positions were found. The role of these residues in their effect against quinolones was evaluated by site-directed mutagenesis. RESULTS: Three different phenotypes (similar resistance, higher resistance or lower resistance to quinolones) were obtained in the random mutagenesis assays when compared with wild-type phenotypes. Only one mutant increased quinolone resistance: QnrS1 containing D185Y substitution (4-fold for ciprofloxacin). Using site-directed mutagenesis, residues G56, C72, C92, G96, F114, C115, S116, A117 and L159, according to the sequence of QnrA1, were analysed and despite the wide amino acid variability of the PRPs, most conserved residues analysed were critical to QnrA1, QnrB1 and QnrS1. CONCLUSIONS: Amino acid sequences of PRPs QnrA1, QnrB1 and QnrS1 could be already optimized for quinolone resistance. One or several changes appear to be insufficient to obtain variants producing fluoroquinolone clinical resistance (MIC > 1 mg/L). Critical residues for quinolone resistance in PRPs were described. Interestingly, different effects were observed for QnrA1, QnrB1 and QnrS1 with the same substitution in several positions.

Qnr-like pentapeptide repeat proteins in gram-positive bacteria.

OBJECTIVES: To study the role of Qnr-like pentapeptide repeat proteins (PRPs) from several gram-positive species with quinolone resistance in vitro. METHODS: A PCR-based strategy was used to clone and express genes coding for Qnr-like PRPs in Enterococcus faecalis, Enterococcus faecium, Listeria monocytogenes, Clostridium perfringens, C. difficile, Bacillus cereus and B. subtilis in Escherichia coli DH10B. MIC values of nalidixic acid and fluoroquinolones were determined for reference strains and E. coli DH10B harbouring recombinant plasmids containing genes coding for PRPs. RESULTS: Amino acid identity of Qnr-like PRPs in gram-positive strains compared with that of the plasmid-mediated quinolone resistance determinants QnrA1, QnrB1 and QnrS1 was in the range of 16% to 22%. Recombinant plasmids coding for Qnr-like PRPs conferred reduced susceptibility to fluoroquinolones (in the range of 0.016 to 0.064 mg/L for ciprofloxacin) and nalidixic acid (from 6 to 12 mg/L), depending on the antimicrobial agent and PRP. The PRP from B. subtilis showed no protective effect. CONCLUSIONS: The PRPs analysed conferred a reduced susceptibility phenotype in E. coli; the data provide further evidence of the possible roles in quinolone resistance of PRPs from different gram-positive species. These gram-positive species may constitute a reservoir for Qnr-like quinolone resistance proteins.

Prevalence of plasmid-mediated quinolone resistance determinants qnr and aac(6')-Ib-cr in Escherichia coli and Klebsiella pneumoniae producing extended-spectrum ß-lactamases in Spain.

The presence of the plasmid-mediated quinolone resistance determinants qnrA, qnrB, qnrS and aac(6')-Ib-cr was evaluated in a collection of 382 isolates of extended-spectrum ß-lactamase (ESBL)-producing Escherichia coli and Klebsiella pneumoniae collected between February and March 2006 for the nationwide Spanish GEIH-ESBL 2006 project. In total, 14 isolates (3.7%) were positive for qnr genes (3 qnrA1, 5 qnrB-like and 6 qnrS1) and 62 isolates (16.2%) were positive for the mutant variant of aac(6')-Ib-cr. The Aac(6')-Ib-cr enzyme was the most prevalent plasmid-mediated mechanism of quinolone resistance in Spain. Most of the Aac(6')-Ib-cr-producing E. coli isolates (94.2%) carried two mutations in gyrA and two in parC, whilst only 57.2% of K. pneumoniae harbouring this enzyme were gyrA and/or parC mutants. Most qnr plasmids were transferable, but only four were conjugative. Plasmid incompatibility groups were identified for only four plasmids, belonging to FIA, HI2 and I1γ. The most prevalent ESBLs associated with qnr plasmids belonged to the SHV and CTX-M families. The present study highlights the broad geographical spread of qnr-like determinants in Spain and their association with the SHV-12 and CTX-M-9 ESBLs in human clinical isolates.