Role of the two-component system AmgRS in early resistance of Pseudomonas aeruginosa to cinnamaldehyde

Abstract

Exposure of Pseudomonas aeruginosa to cinnamaldehyde (CNA), a natural electrophilic antimicrobial often used as self-medication to treat mild infections, triggers overproduction of the MexAB-OprM efflux system, leading to multidrug resistance. In this study, we demonstrate that CNA exposure induces expression of genes regulated by the two-component system AmgRS. AmgRS activates MexAB-OprM production, independent of repressors MexR and NalD. In addition to the essential role played by the NalC-ArmR pathway in this adaptive process, AmgRS is critical for the survival of P. aeruginosa challenged with CNA. Altogether, these data suggest that efflux-dependent and -independent mechanisms are activated in the early phase of CNA exposure, allowing for progressive enzymatic reduction of the biocide to non-toxic cinnamic alcohol.IMPORTANCEExposure of Pseudomonas aeruginosa to cinnamaldehyde (CNA), an antimicrobial used in self-medication, induces overproduction of the MexAB-OprM efflux system, leading to multidrug resistance. Our study demonstrates that the AmgRS two-component system aids in the survival of P. aeruginosa strain PA14 under CNA exposure through both MexAB-OprM-dependent and -independent mechanisms until the enzymatic reduction of CNA into the less toxic cinnamic alcohol. This discovery highlights the pivotal role of AmgRS in mediating defense against aldehyde biocides, emphasizing its significance in the persistence of P. aeruginosa, a pathogen associated with hospital-acquired infections and cystic fibrosis, and underscores the potential impact on clinical treatment strategies.

Keywords: Pseudomonas aeruginosa; antibiotic resistance; cinnamaldehyde; cross resistance antibiotics essential oils; efflux pumps; two-component regulatory systems.

Fig 1

Regulation of operon mexAB-oprM expression. (A) In the absence of inducer (e.g., CNA), repressor NalC prevents the production of the MexR antirepressor ArmR, thus allowing MexR to negatively control mexAB-oprM expression. In the presence of sub-Minimal Inhibitory Concentration (MIC) CNA, mexAB-oprM is upregulated by both AmgR and ArmR/NalC (black arrow). AmgRS is also known to indirectly regulate operon mexXY expression through the activity of membrane protease HtpX and protein PA5528 (10). (B) Map of the mexR-mexA intergenic region showing the binding sites of regulators NalD, MexR, CpxR, and AmgR, respectively, as determined by electromobility shift assays (9, 11–13). The following repeated DNA motifs GNAANANNNNGNAANA and GTAAANNNNNGTAAAN have been found by in silico analysis upstream of the genes regulated by AmgR and CpxR, respectively (7). Alignment of the AmgR DNA motif with the region upstream of mexAB-oprM identified a sequence (indicated in yellow), overlapping the distal promoter of the operon and the DNA site recognized by MexR (The MEME Suite, meme-suite.org).

Fig 2

Relative expression levels of gene PA5528. The relative expression levels of PA5528 in P. aeruginosa strains PA14 and PA14∆amgRS were determined at t₀ (white bars), t_{30 min} (light gray), t_1h (dark gray), and t_2h (black) post-exposure to 350 µg/mL CNA in exponentially growing bacteria (A_{600 nm} of 0.75). The amounts of specific cDNA were assessed on a Rotor Gene RG6000 instrument (Qiagen, Courtaboeuf, France) by using the QuantiTect SYBR green PCR kit (Qiagen). The values from three biological replicates were averaged for each strain, normalized to that of the housekeeping gene rpsL, and finally expressed as a ratio to the transcript levels of untreated wild-type PA14 strain cultured in Mueller-Hinton broth (MHB) supplemented with 0.3% dimethylsulfoxide (DMSO), the solvent used for CNA solubilization. An analysis of variance test was performed on the data, followed by a Dunnett test comparing each time to t0 in PA14 and in PA14ΔamgRS, and comparing PA14ΔamgRS to PA14 at each selected time. A Friedman test followed by a Wilcoxon test were made for both strains at t2h. *P-value <0.05; **P-value <0.01.

Fig 3

Susceptibility of strain PA14 and derivative mutants to CNA. (A) Spot test where 5 µL volumes of serial dilutions of log-phase bacteria (from ca. 10⁸ to 10⁵ CFU/mL) were deposited on the surface of a Mueller-Hinton agar (MHA) plate containing 500 µg/mL CNA, and then incubated for 18 h at 37°C. Growth controls (T+) were spotted on an MHA without CNA. (B) Bacterial suspensions adjusted to an A_{600 nm} of 0.01 were cultured in MHB in 96-well microplates for 18 h at 37°C, with (dashed lines) or without (solid lines) 500 µg/mL CNA. Bacterial density was assessed at A_{600 nm} in a spectrophotometer Spark (Tecan, Männedorf, Switzerland) equipped with a humidity cassette to prevent desiccation. (C) Bioluminescence (expressed in Log RLU, Relative Light Units) of bacteria developing in MHB with (dashed lines) or without (solid lines) 700 µg/mL CNA was monitored over a 4-h time course in white 96-well assay plates (Corning, NY, USA) by using a Synergy H1 microplate reader (Biotek Instrument, Winooski, USA) set at a gain value of 160, read height of 7 mm, and integration time of 2 s. The initial inoculum was adjusted to an A_{600 nm} equal to 0.8. Growth (B) and survival (C) curves are in black for PA14-lux, blue for PA14∆armR, green for PA14∆amgRS, pale blue for PA14∆mexAB, and red for PA14∆amgRS∆armR.

Fig 4

Relative expression levels of genes mexB (A) and mexR (B). Log-phase suspensions of strains PA14, PA14∆armR, PA14∆amgRS, and PA14∆amgRS∆armR (PA14∆∆) were diluted in MHB to an A_{600 nm} of 0.03 and then exposed to 350 µg/mL CNA when the absorbance reached 0.75 units. Total RNA was extracted at t₀ (white bars), t_{30 min} (light gray), t_1h (dark gray), and t_2h (black) of culture at 37°C, and mRNA levels were then quantified by RT-qPCR on a Rotor Gene RG6000 instrument (Qiagen, Courtaboeuf, France) by using the QuantiTect SYBR green PCR kit (Qiagen) on three biological replicates, averaged for each strain, normalized to that of housekeeping gene rpsL, and finally expressed as a ratio (Log fold change between values) to the transcript levels of untreated wild-type PA14. As established previously, strains were considered to significantly upregulate mexB when the gene expression level was ≥2.5-fold higher than that of PA14 (16). An analysis of variance test was performed on the data, followed by a Dunnett test comparing each time to t0, and comparing each strain to PA14 at each selected time. A Friedman test followed by a Wilcoxon test were made for PA14 and PA14ΔarmR strain for mexR. *P-value <0.05; **P-value < 0.01; ***P_value < 0.001.