Modulation of Membrane Influx and Efflux in Escherichia coli Sequence Type 131 Has an Impact on Bacterial Motility, Biofilm Formation, and Virulence in a Caenorhabditis elegans Model

Abstract

Energy-dependent efflux overexpression and altered outer membrane permeability (influx) can promote multidrug resistance (MDR). The present study clarifies the regulatory pathways that control membrane permeability in the pandemic clone Escherichia coli sequence type 131 (ST131) and evaluates the impact of efflux and influx modulations on biofilm formation, motility, and virulence in the Caenorhabditis elegans model. Mutants of two uropathogenic E. coli (UPEC) strains, MECB5 (ST131; H30-Rx) and CFT073 (ST73), as well as a fecal strain, S250 (ST131; H22), were in vitro selected using continuous subculture in subinhibitory concentrations of ertapenem (ETP), chloramphenicol (CMP), and cefoxitin (FOX). Mutations in genes known to control permeability were shown for the two UPEC strains: MECB5-FOX (deletion of 127 bp in marR; deletion of 1 bp and insertion of an IS1 element in acrR) and CFT073-CMP (a 1-bp deletion causing a premature stop in marR). We also demonstrated that efflux phenotypes in the mutants selected with CMP and FOX were related to the AcrAB-TolC pump, but also to other efflux systems. Alteration of membrane permeability, caused by underexpression of the two major porins, OmpF and OmpC, was shown in MECB5-ETP and mutants selected with FOX. Lastly, our findings suggest that efflux pump-overproducing isolates (CMP mutants) pose a serious threat in terms of virulence (significant reduction in worm median survival) and host colonization. Lack of porins (ETP and FOX mutants) led to a high level of antibiotic resistance in an H30-Rx subclone. Nevertheless, this adaptation created a physiological disadvantage (decreased motility and ability to form biofilm) associated with a low potential for virulence.

FIG 1

Schematic diagram of the genomic region comprising the acrR, marR, and soxR genes in E. coli K-12 substrain MG1655 (accession number NC_000913.3). The nucleotide mutations and the deduced protein changes for the three parental strains and their isogenic mutants are shown.

FIG 2

Relative mRNA expression levels of genes implicated in MDR regulation (A), membrane efflux (B), and permeability (C) for the ST131 H30-Rx E. coli strain MECB5 and its mutants in vitro selected with ETP, CMP, and FOX. The log-transformed averages of relative fold changes of mutants compared to the parental strain for each antibiotic are presented. The error bars represent the standard deviations from three different RNA preparations. Significant differences from the parental MECB5 using Dunnett's test are indicated; *, P < 0.05; **, P < 0.01; ***, P < 0.001.

FIG 3

Comparative results of swimming and swarming assays for MECB5, S250, and CFT073 parental strains and their isogenic mutants selected with ETP, CMP, and FOX. The error bars represent the standard deviations from at least three independent assays. Significant differences from the parental strains using Dunnett's test are indicated; *, P < 0.05; ***, P < 0.001.

FIG 4

Effects of ETP, CMP, and FOX resistance on E. coli biofilm formation. The kinetics of the early phase of biofilm formation for strains MECB5 (A), S250 (B), and CFT073 (C) and their respective mutants were determined by the BioFilm ring test (BioFilm Control). Means ± standard errors of the mean of BFIs for at least three independent replicates are presented. Statistical differences between the strains and their mutants at each time were obtained by using two-way ANOVA, followed by Dunnett's multiple-comparison test.