Evolutionary trade-offs associated with loss of PmrB function in host-adapted Pseudomonas aeruginosa

Abstract

Pseudomonas aeruginosa colonises the upper airway of cystic fibrosis (CF) patients, providing a reservoir of host-adapted genotypes that subsequently establish chronic lung infection. We previously experimentally-evolved P. aeruginosa in a murine model of respiratory tract infection and observed early-acquired mutations in pmrB, encoding the sensor kinase of a two-component system that promoted establishment and persistence of infection. Here, using proteomics, we show downregulation of proteins involved in LPS biosynthesis, antimicrobial resistance and phenazine production in pmrB mutants, and upregulation of proteins involved in adherence, lysozyme resistance and inhibition of the chloride ion channel CFTR, relative to wild-type strain LESB65. Accordingly, pmrB mutants are susceptible to antibiotic treatment but show enhanced adherence to airway epithelial cells, resistance to lysozyme treatment, and downregulate host CFTR expression. We propose that P. aeruginosa pmrB mutations in CF patients are subject to an evolutionary trade-off, leading to enhanced colonisation potential, CFTR inhibition, and resistance to host defences, but also to increased susceptibility to antibiotics.

Fig. 1

P. aeruginosa pmrB mutants are resistant to host defence molecules and lysozyme killing. Supernatants from phorbol 12-myristate 13-acetate (PMA)-stimulated (a) HL-60 or (b) mouse bone marrow neutrophil cultures were transferred to mid-log growth P. aeruginosa. Colony forming units (CFU) as a percentage of control. Control bacteria were exposed to supernatant from non-activated (no PMA) HL-60 cultures. c Optical density at 600 nm of P. aeruginosa grown in the presence of lysozyme (mean ± s.d.). Results are expressed as a percentage of the no-lysozyme control for each isolate and are a composite of 3 biological replicates per isolate, each containing 3 technical replicates. Asterisks (*) represent significant differences in ANOVA performed on CFU counts (a, b) or OD readings, with lysozyme as a covariant (c), with Dunnett’s multiple comparison test vs. LESB65. *p < 0.05, **p < 0.01, ***p < 0.005

Fig. 2

Principal component biplot of LESB65 and ΔpmrB proteomics data showing proteins with significantly different abundance between samples. Five biological replicates were performed for each isolate. Unit variance scaling is applied to rows; SVD with imputation is used to calculate principal components. X axis and Y axis show principal component 1 and principal component 2 that explain 59.6% and 16.7% of the total variance, respectively. Prediction ellipses are such that with probability 0.95, a new observation from the same group will fall inside the ellipse. PCA analysis of the full dataset, with all detected proteins, can be found in Supplementary Fig. 4

Fig. 3

ΔpmrB downregulates CFTR expression on airway epithelial cells. a CFTR expression on uninfected (red), LESB65 infected (black) and ΔpmrB infected (blue) A549 airway epithelial cells as measured by flow cytometry and compared to isotype control staining (grey). CFTR alexa fluor 488 median fluorescent intensity (MFI) on stained (b) A549 and (c) Detroit cells. Data in (b) are a composite of two independent experiments with two-three biological replicates per condition in each experiment. Data in (c) are from a single experiment containing 4 biological replicates. Asterisks (*) represent significant differences in one-way ANOVA with Dunnett’s multiple comparison test. *p < 0.05, ***p < 0.001, ****p < 0.0001

Fig. 4

Cif and Ivy contribute to establishment of lung infection. Mice were infected by intranasal instillation of 1 × 10⁶ colony forming units (CFU) P. aeruginosa PAO1, or PAO1 with transposon insertions in cif or ivy. Mice were culled immediately after infection (n = 4 per group) or at 1, 3 and 7 days post-infection (n = 6 per group) and CFU in (a) lung and (b) nasopharynx determined by plating of tissue homogenates onto selective agar. Data are from a single experiment. *p < 0.05, **p < 0.01, ***p < 0.001 in two-way ANOVA with Dunnett’s multiple comparison test vs. PAO1

Fig. 5

Pili-dependent processes in LESB65 and ΔpmrB. a Adhesion and (b) invasion of LESB65 and ΔpmrB to A549 human alveolar epithelial cells (lung) and human Detroit pharyngeal cells (pharynx). Results are expressed as a percentage of total bacterial numbers per well. Data are a composite of four (lung) or two (pharynx) independent experiments, each containing 5 replicate wells per experimental group. c Swarming motility of PAO1, ΔpmrB, and LESB65 on semisolid agar. (d) Swarm diameter of isolates grown on semisolid agar. Data points shown are from three independent experiments. *p < 0.05, **p < 0.01 and ns non-significant in unpaired t-test with Welch’s correction (a, b) or one-way ANOVA with Tukey post-test (d)

Fig. 6

Principal component biplot of LESB65, pmrB SNP and ΔpmrB. Unit variance scaling is applied to rows; SVD with imputation is used to calculate principal components. X axis and Y axis show principal component 1 and principal component 2 that explain 45% and 25.3% of the total variance, respectively. Prediction ellipses are such that with probability 0.95, a new observation from the same group will fall inside the ellipse. N = 15 data points

Fig. 7

pmrB mutations are found in patient isolates and are associated with lysozyme resistance. Three isolates from one patient (a) and seven from another (b) were compared for lysozyme resistance. Isolates with at least one pmrB mutation are in red (49194 = R202H, 49137 = I73L, E82K, E225Q). (a, b) Growth in the presence of lysozyme expressed as a percentage of growth in LB (OD₆₀₀). Data are presented as mean ± s.d. and are a composite of three independent experiments with each isolate run in triplicate. **p < 0.01, ***p < 0.005, ns non-significant, vs. 49194 (a) or 49137 (b) in ANOVA analysis performed on OD₆₀₀ with lysozyme as a covariant

Fig. 8

Proposed mechanism for enhanced infection potential of pmrB mutants. Protein expression changes associated with pmrB mutation may confer an ability to reproduce the phenotypic features of the CF lung via inhibition of host CFTR. Increased expression of type IV pili aids colonisation and adhesion to airway epithelium. Resistance to lysozyme associated with increased ivy production enables persistence in the face of immune pressure. The airway surface liquid is depicted in green for wild type and yellow for the pmrB mutant to denote dehydration resulting from CFTR inhibition