RavA-ViaA antibiotic response is linked to Cpx and Zra2 envelope stress systems in Vibrio cholerae

Abstract

The RavA-ViaA complex was previously found to sensitize Escherichia coli to aminoglycosides (AGs) in anaerobic conditions, but the mechanism is unknown. AGs are antibiotics known for their high efficiency against Gram-negative bacteria. In order to elucidate how the expression of the ravA-viaA genes increases bacterial susceptibility to aminoglycosides, we aimed at identifying partner functions necessary for increased tolerance in the absence of RavA-ViaA, in Vibrio cholerae. We show that membrane stress response systems Cpx and Zra2 are required in the absence of RavA-ViaA, for the tolerance to AGs and for outer membrane integrity. In the absence of these systems, the ∆ravvia strain's membrane becomes permeable to external agents such as the antibiotic vancomycin.

Keywords: Vibrio cholerae; antibiotic resistance; antibiotic tolerance; bacterial membrane; ravA viaA; sub-MIC antibiotics.

Fig 1

Effect of RavA-ViaA on fitness and tolerance to aminoglycosides. (A) Competition experiments mixing ravvia deletion (∆) and extra-copy (OE+) mutants and WT in indicated conditions. NT, non-treated; TOB, tobramycin; GEN, gentamicin; CM, chloramphenicol; CIP, ciprofloxacin. The y-axis represents log₂ of competitive index value calculated as described in the methods. A competitive index of 1 (i.e., log₂ value of 0) indicates equal growth of both strains. For statistical significance calculations, we used one-way ANOVA. **** means P < 0.0001, *** means P < 0.001, ** means P < 0.01, * means P < 0.05. Only significant P values are shown. Number of replicates for each experiment: 3 < n < 8. Concentrations are indicated in µg/mL. (B) Survival of indicated strain to lethal tobramycin treatment. V. cholerae WT and deletion mutant cultures were grown without antibiotics up to early exponential phase, and serial dilutions were plated on MH medium without antibiotics. Exponential phase cultures were then treated with antibiotics at lethal concentrations for the indicated times. At each time point, dilutions were spotted on MH. Y-axis shows survival calculated as a number of colonies at time TN divided by the initial number of colonies before antibiotic treatment. TOB: tobramycin 5 or 10 µg/mL.

Fig 2

Effect of RavA-ViaA on AG uptake and membrane potential. (A) Intracellular level of neomycin coupled to the fluorophore Cy5 measured by fluorescence-associated flow cytometry. Error bars represent standard deviation. (B) Quantification of changes in PMF using Mitotracker Red fluorescence measured by flow cytometry. Representative acquisitions are shown: fluorescence is represented in the x-axis (FITC channel), the y-axis represents the number of events corresponding to the number of cells, normalized to height (same number of total cells for both conditions). Each plot represents one experiment. (C) Competition experiments of V. cholerae WT and indicated mutants. MH: no antibiotic treatment (black). TOB: tobramycin 0.6 µg/mL (blue). The y-axis represents log₂ of competitive index value calculated as described in the methods. A competitive index of 1 (i.e., log₂ value of 0) indicates equal growth of both strains. For statistical significance calculations, we used one-way ANOVA. **** means P < 0.0001, *** means P < 0.001, ** means P < 0.01, * means P < 0.05. ns, non-significant. Number of replicates for each experiment: n = 3. Only significant P values are shown.

Fig 3

High-throughput methods identify factors that are differentially detected in ∆ravvia compared to WT. Pie-chart for RNA-seq recapitulates differentially (up- or down-) expressed gene categories in ∆ravvia, as shown in Table S1. Pie-chart for TN-seq recapitulates gene categories for differentially detected transposon insertions in ∆ravvia compared to WT, in all conditions shown in Tables S2 to S4.

Fig 4

Cpx and Zra-like two-component envelope stress response systems are involved in fitness increase of ∆ravvia with TOB and TOB tolerance. ABC. Competitions. The effect of deletion of cpx (A), or zra-like (B), or both (C) on competitive index in MH without and with TOB, where MH is the untreated growth medium. In vitro competition experiments of V. cholerae WT and indicated mutants in specified media: in black: MH: no antibiotic treatment. In blue: TOB: tobramycin 0.6 µg/mL. The y-axis represents log₂ of competitive index value calculated as described in the methods. A competitive index of 1 (i.e., log₂ value of 0) indicates equal growth of both strains. (D) Competitive index in MH without and with TOB, of the triple mutant (indicated as ∆∆∆) ∆ravvia ∆cpx ∆zra against ∆ravvia ∆cpx. (E) Tolerance. Cultures were grown to exponential phase in MH medium. Survival of WT and ∆ravvia to 3 h treatment with lethal TOB at 5× MIC 5 µg/mL was measured. The y-axis represents log₂ value of survival ratios, calculated as survival of the mutant over survival of the WT. A relative survival ratio of 1 (i.e., log₂ = 0) indicates equal survival as the WT strain. For statistical significance calculations, we used one-way ANOVA. **** means P < 0.0001, *** means P < 0.001, ** means P < 0.01, * means P < 0.05. ns, non-significant. Number of replicates for each experiment: 3 < n < 8.

Fig 5

Effect of zinc supplementation. ABC. Competitions. Cultures were grown to exponential phase in MH medium supplemented with zinc during growth. “Zn” stands for ZnCl₂: 1.5 mM. The effect of deletion of cpx (A), or zra-like (B), or both (C) on competitive index in MH without and with TOB, where MH is the untreated growth medium. In vitro competition experiments of V. cholerae WT and indicated mutants In specified media: in black: MH: no antibiotic treatment. In blue: TOB: tobramycin 0.6 µg/mL. The y-axis represents log₂ of competitive index value calculated as described in the methods. A competitive index of 1 (i.e., log₂ value of 0) indicates equal growth of both strains. (D) Survival of WT and ∆ravvia to 3 h treatment with lethal TOB at 5× MIC 5 µg/mL, in the presence of zinc. The y-axis represents log₂ value of survival rates ratios, calculated as survival of the mutant over survival of the WT. A relative survival ratio of 1 (i.e., log₂ = 0) indicates equal survival as the WT strain. (E) Expression of cpx and zra. mRNA levels were measured using digital RT-PCR as explained in Materials and Methods. The y-axis represents the fold change of induction in the presence of zinc divided by the expression in the absence of zinc. (F) Expression from promoter of ravvia was measured using fluorescent transcriptional fusion of gfp expressed from ravvia promoter and quantified using flow cytometry. The y-axis represents the relative fluorescence in the indicated conditions. Zn: zinc 1.5 mM. NT: non-treated. ∆zur: zur deletion strain. WT: wild-type strain. For statistical significance calculations, we used one-way ANOVA. **** means P < 0.0001, *** means P < 0.001, ** means P < 0.01, * means P < 0.05. ns, non-significant. Number of replicates for each experiment: 3 < n < 6.

Fig 6

Low ROS phenotype of ∆ravvia is dependent on the presence of Cpx and Zra-like stress response systems. Quantification of variation of reactive oxygen species using CellRox. The y-axis represents log₂-fold change of detected ROS fluorescence in the indicated strain over the WT strain. Each experiment was performed at least three times, and data and statistical significance are shown in the histograms. For statistical significance calculations, we used one-way ANOVA. **** means P < 0.0001, ** means P < 0.01. ns means non-significant.

Fig 7

Response of ∆ravvia to the membrane targeting antibiotics. (A) Minimum inhibitory concentration of vancomycin. MIC values are represented in the y-axis for the indicated strains. (B) Outer membrane permeability to nitrocefin measured on stationary phase cultures diluted to 5 × 10⁷ cells/mL by measuring the OD at 490 nm for 40 min. (C) AG uptake quantified through Neo-cy5 entry into exponential phase cultures. Intracellular level of neomycin coupled to the fluorophore Cy5 measured by fluorescence-associated flow cytometry. Error bars represent standard deviation. For statistical significance calculations, we used one-way ANOVA. **** means P < 0.0001. Only significant P values are shown.