Virulence Characteristics and an Action Mode of Antibiotic Resistance in Multidrug-Resistant Pseudomonas aeruginosa

Abstract

Pseudomonas aeruginosa displays intrinsic resistance to many antibiotics and known to acquire actively genetic mutations for further resistance. In this study, we attempted to understand genomic and transcriptomic landscapes of P. aeruginosa clinical isolates that are highly resistant to multiple antibiotics. We also aimed to reveal a mode of antibiotic resistance by elucidating transcriptional response of genes conferring antibiotic resistance. To this end, we sequenced the whole genomes and profiled genome-wide RNA transcripts of three different multi-drug resistant (MDR) clinical isolates that are phylogenetically distant from one another. Multi-layered genome comparisons with genomes of antibiotic-susceptible P. aeruginosa strains and 70 other antibiotic-resistance strains revealed both well-characterized conserved gene mutations and distinct distribution of antibiotic-resistant genes (ARGs) among strains. Transcriptions of genes involved in quorum sensing and type VI secretion systems were invariably downregulated in the MDR strains. Virulence-associated phenotypes were further examined and results indicate that our MDR strains are clearly avirulent. Transcriptions of 64 genes, logically selected to be related with antibiotic resistance in MDR strains, were active under normal growth conditions and remained unchanged during antibiotic treatment. These results propose that antibiotic resistance is achieved by a "constitutive" response scheme, where ARGs are actively expressed even in the absence of antibiotic stress, rather than a "reactive" response. Bacterial responses explored at the transcriptomic level in conjunction with their genome repertoires provided novel insights into (i) the virulence-associated phenotypes and (ii) a mode of antibiotic resistance in MDR P. aeruginosa strains.

Figure 1

Alignment of PAO1, Y31, Y71, Y82, and Y89 whole genomes with the Mauve program. Strain names are indicated on the left in different colors. The Y89 strain harbors a plasmid, and its sequence is shown at the end (red arrow). The PAO1 genome is displayed as a collection of squared regions above the center line. Regions with homologous sequences are shown with the same-colored squares. Squares that lie above the center line are aligned in the same orientation, while squares below the line are aligned in the inverse orientation, relative to the corresponding regions of the PAO1 genome. Each squared region consists of a collection of vertical lines, and the height of each line is drawn proportional to the similarity between the sequences. Thus, a white region inside squared regions represents a sequence that is exclusively present on a given genome.

Figure 2

Transcriptome comparison between MDR versus antibiotic-susceptible strains. (A) Seventy-eight genes that satisfied two parameters (fold change >2, FDR < 0.05) were selected among the antibiotic-susceptible (PAO1, Y31) and MDR strains (Y71, Y82, Y89) by the edgeR tool. These genes were lined up next to the heat map. TMM-normalized read counts of samples were converted to log2. After each gene was averaged from five samples, the average was subtracted from the read count of each gene. If the last calculated value was high, it was displayed in the heat map in red; when it was low, it was displayed in blue, as shown in the Color Key (top left). Hierarchical clustering was performed on the basis of genes and samples. (B) 64 genes that exhibited decreased expression (>2-fold) in the MDR strains were used for a functional gene network using STRING. The box drawn with the solid line was the group related to quorum sensing, and box drawn with the dashed line was the group related to the type VI secretion system.

Figure 3

Elevated expressions of mexAB-oprM genes in Y71 and Y89. TMM-normalized read counts of the mexAB-oprM operon (A) and nalC gene (B) in RNASeq results with bacteria of the exponential phase. RNA extractions and sequence analysis were performed as described in Experimental Procedures. (C) NalD protein sequences of PAO1, Y31, Y71, Y82, and Y89 were compared with CLUSTALW.

Figure 4

Decreased QS-mediated virulence and in vivo infectivity in MDR strains. (A) Qualitative analysis of 3-oxo-C12-HSL production. An E. coli reporter strain harboring the pKDT17 plasmid was incubated with overnight-grown supernatants of each bacterial strain (indicated at the bottom) for one hour and then subjected to β-galactosidase assays. Overnight-grown culture supernatants of each strain were used for elastase (B) and pyocyanin (D) assays. (C) LasR protein sequences of PAO1, Y31, Y71, Y82, and Y89 were compared with CLUSTALW. (E) Eight-week-old BALB/C mice (n = 6) were infected with approximately 2.5 × 10⁷ bacterial cells of indicated strains. Mouse survival rates were monitored following infection. (F) After the death of mouse or 42 hours of infection, mouse lung was homogenized in PBS and serially diluted for enumerating bacterial cells. Diluted lung homogenates were spotted on Pseudomonas Isolation Agar. ***p < 0.001 vs. the CFU of the PAO1-infected samples.

Figure 5

Decreased Type VI secretion system swarming activities and iron acquisition activities in MDR strains. (A) Bacterial competition assays with Vibrio cholerae V52 and P. aeruginosa. V. cholerae V52 as prey of P. aeruginosa strains, PAO1 and clinical isolates were incubated overnight in LB. V52 alone and mixtures of P. aeruginosa and V52 in a 1:1 ratio were grown on LB agar for 5 hr. Bacteria grown on LB agar plates were scraped, serially diluted, spread on TCBS agar for selection of V52, and incubated overnight at 37 °C for enumeration. NC is CFU of V52 alone for 5 hr. ***p < 0.001 vs. V52 CFUs after incubation with PAO1. (B) Overnight-grown P. aeruginosa strains were spot-inoculated on 0.5% agar plate composed of the 0.8% nutrient broth and 0.5% glucose and incubated overnight at 30 °C. (C) CAS assay for iron acquisition activity. The 10-fold diluted supernatants from overnight cultures were reacted with CAS solution for 2 hr, and the OD of the resulting solution was measured at 630 nm. *p < 0.05 vs. iron acquisition activity of PAO1 and Y31.

Figure 6

Distribution of 64 ARGs in the genomes of 70 different MDR P. aeruginosa strains. Protein sequences encoded by the 64 ARGs (far left column) were aligned against all protein sequences deduced from genomes of 70 different MDR P. aeruginosa strains (top line), with several thresholds of similarity (S) and length coverage (L); (S > 95% and L > 95%, S > 90% and L > 90%, S > 80% and L > 80%, S > 70% and L > 70%). Seventy MDR strains are listed in Table S4 with detailed information. Presence of 64 genes in each genome is indicated with color-coded square boxes. The heat map was constructed by applying the color that satisfied the highest threshold. The blue-colored and red-colored genes were designated as gene group 1 (GG1) and gene group 2 (GG2), respectively.

Figure 7

Transcriptional responses of 64 ARGs to the treatment with antibiotic cocktail (AC) and constitutive function of the ARGs to endure antibiotic stresses. (A) The mRNA transcript of each of the 64 genes was calculated before and after AC treatment in each MDR strain and the ratio of their average values is represented as a color-coded square. Red and blue squares indicate >2-fold increase or >2-fold decrease in expression upon AC treatment, respectively. The genes with fold changes between 0.5 and 2 are shown in black squares. The names of the representative genes were taken from the strain Y82 that has the largest genome among the MDR strains. Among the MDR strains (Y71, Y82, and Y89), 64 ARGs presented homology of more than 90%. (B) Potential working models by which ARGs respond to antibiotic stresses. Pacman shapes represent the expressions of ARGs, and fan shapes represent various antibiotics. The thick cell wall means that the ARG-expressing bacterial cell is ready to withstand antibiotic stresses.