Pyoverdine-Dependent Virulence of Pseudomonas aeruginosa Isolates From Cystic Fibrosis Patients

Abstract

The development of therapies that modulate or prevent pathogen virulence may be a key strategy for circumventing antimicrobial resistance. Toward that end, we examined the production of pyoverdine, a key virulence determinant, in ∼70 Pseudomonas aeruginosa isolates from pediatric cystic fibrosis patients. Pyoverdine production was heterogeneous and showed a clear correlation with pathogenicity in Caenorhabditis elegans and an acute murine pneumonia model. Examination showed pyoverdine accumulation in host tissues, including extrapharyngeal tissues of C. elegans and lung tissues of mice, where accumulation correlated with host death. Many of the isolates tested were resistant to multiple antimicrobials, so we assayed the ability of pyoverdine inhibitors to mitigate virulence and rescue pyoverdine-mediated host pathology. Representatives from three different classes of pyoverdine inhibitors (gallium, fluoropyrimidines, and LK11) significantly improved survival. Our findings highlight the utility of targeting virulence factors in general, and pyoverdine in particular, as a promising method to control bacterial pathogenesis as the utility of antimicrobials continues to diminish.

Keywords: Caenorhabditis elegans; Pseudomonas aeruginosa; antimicrobial resistance; antivirulence; cystic fibrosis; murine infection; pyoverdine.

FIGURE 1

Pyoverdine production is necessary for P. aeruginosa virulence in a murine acute pneumonia model. (A) Survival advantage over sham (PA14: saline) in mice treated with 5-fluorocytosine (5-FC) or those inoculated with PA14pvdF. All drug and saline doses contained 5% DMSO. Data were pooled from two replicates. p-Values were calculated using Kruskal–Wallis one-way ANOVA and Dunn’s multiple comparison test. ^#p < 0.01, ^∗p < 0.0001 compared to saline control. (B) Pyoverdine content in mouse lung homogenates. Pyoverdine fluorescence was measured after treatment with 8-hydroxyquinoline to remove iron from ferripyoverdine. Error bars represent sample standard deviation. p-Values were calculated using Student’s t-test. ^∗p < 0.01.

FIGURE 2

Pyoverdine production correlates with virulence against Caenorhabditis elegans. (A) Pyoverdine production in 69 P. aeruginosa isolates from CF patients (black bar: PA14). Error bars represent SEM from two biological replicates. (B) Correlation between pyoverdine production (from bacteria statically grown in M9 media) and C. elegans death after P. aeruginosa exposure (in Liquid Killing conditions) – see section “Materials and Methods” for details. Data were normalized to PA14. Black diamonds and triangles represent the high- and low-virulence isolates selected for further study. Each point represents the average of at least two biological replicates.

FIGURE 3

Pyoverdine from virulent isolates translocates into C. elegans and disrupts host iron homeostasis. (A) Confocal laser-scanning micrographs of C. elegans pharynx in worms treated with media or pyoverdine-rich bacterial filtrates from PA2-61 and PA2-72. Pyoverdine fluorescence is shown in blue. (B) Pyoverdine accumulation in extralumenal pharyngeal tissue after exposure to filtrate from PA2-61 or the same filtrate pre-saturated with ferric iron. Blue indicates fluorescence from iron-free pyoverdine, red staining shows the pharyngeal lumen. (C) C. elegans iron content after exposure to highly virulent (PA2-61, PA2-72, PA14) and avirulent (PA2-88, PA3-22) isolates or media in the absence of pathogen. Host iron content was measured via inductively coupled plasma mass spectrometry (ICP-MS) (see Supplementary Materials and Methods). Error bars represent SEM between three biological replicates. p-Values were calculated using Student’s t-test. ^#p < 0.05, ^∗p < 0.01.

FIGURE 4

In vivo pyoverdine production correlates with virulence in murine hosts. (A) Mice survival after intranasal inoculation with strains that were highly virulent (PA2-61, PA2-72) or avirulent (PA2-88, PA 3-22) against C. elegans. (B) Pyoverdine content in mouse lung homogenates. Pyoverdine fluorescence was measured after 8-hydroxyquinoline treatment to remove iron from ferripyoverdine. Error bars represent sample standard deviation. Data were pooled from two replicates. p-values in (A) were calculated using Kruskal–Wallis one-way ANOVA and Dunn’s multiple comparison test. p < 0.01 for PA2-61 vs. PA2-88 or PA3-22, and PA2-72 vs. PA2-88 or PA3-22; p < 0.05 for PA2-61 vs. PA2-72. p-Values in (B) were calculated using Student’s t-test. ^∗p < 0.01.

FIGURE 5

Conventional antibiotics display reduced efficacy against MDR isolates in a C. elegans model. (A) Susceptibility of representative virulent (PA14, PA2-61, PA2-72) and avirulent isolates (PA2-88, PA3-22) to various antibiotics, including AMK, amikacin; AMP, ampicillin; AMS, ampicillin-sulbactam; AZM, aztreonam; Cfz, cefazolin; Cpe, cefepime; Cfx, cefoxitin; Caz, ceftazidime; Cax, ceftriaxone; CIP, ciprofloxacin; Etp, ertapenem; GEN, Gentamicin; Imp, imipenem; Lvx, levofloxacin; Mer, meropenem; F/M, nitrofurantoin; TZP, piperacillin-tazobactam; TET, tetracycline; TOB, tobramycin; T/S, trimethoprim-sulfamethoxazole. Based on CLSI MIC cutpoints, S, I, and R indicate sensitivity, intermediate resistance, and complete resistance to the tested antimicrobial, respectively. P. aeruginosa is considered sensitive to CIP if the MIC is 1 μg/mL (∼2.6 μM) or less and sensitive to GEN if the MIC is 4 μg/mL (∼8.4 μM) or less. Most strains of P. aeruginosa are considered resistant to TET, but Staphylococcus aureus is considered sensitive if the MIC is 4 μg/mL (∼8.4 μM) or less. C. elegans survival after exposure to P. aeruginosa PA14, PA2-61, or PA2-72 in the presence of (B) gentamicin, (C) tetracycline, or (D) ciprofloxacin. Error bars represent SEM between three biological replicates. p-values were calculated using Student’s t-test. ^∗p < 0.01.

FIGURE 6

Pyoverdine inhibitors effectively mitigate P. aeruginosa pathogenesis. (A) C. elegans survival after exposure to P. aeruginosa PA14, PA2-61, or PA2-72 in the presence of 100 μM gallium. (B) C. elegans survival after exposure to pyoverdine-rich bacterial filtrates pre-saturated with gallium. (C,D) C. elegans survival after exposure to P. aeruginosa in the presence of (C) 50 μM 5-fluorocytosine (5-FC) and 10 μM 5-fluorouridine (5-FUR), or (D) 100 μM LK11. Error bars in (A,C,D) represent SEM between three biological replicates. Data presented in (B) are one representative result from three biological replicates; error bars represent SEM from 16 technical replicates. p-values were calculated using Student’s t-test. ^∗p < 0.01.

FIGURE 7

Antimicrobial resistance negatively correlates with pyoverdine production. (A) Susceptibility of 69 P. aeruginosa isolates to various antibiotics, including AMK, amikacin; AMP, ampicillin; AMS, ampicillin-sulbactam; AZM, aztreonam; Cfz, cefazolin; Cpe, cefepime; Cfx, cefoxitin; Caz, ceftazidime; Cax, ceftriaxone; CIP, ciprofloxacin; Etp, ertapenem; GEN, gentamicin; Imp, imipenem; Lvx, levofloxacin; Mer, meropenem; F/M, nitrofurantoin; TZP, piperacillin-tazobactam; TET, tetracycline; TOB, tobramycin; T/S, trimethoprim-sulfamethoxazole. (B,C) Histogram of pyoverdine production (normalized to that of PA14) in isolates that are susceptible or resistant to (B) ciprofloxacin, gentamicin, or meropenem, (C) aztreonam, or ceftazidime. p-values were calculated using Student’s t-test. ^#p < 0.05.