High prevalence of lipopolysaccharide mutants and R2-Pyocin susceptible variants in Pseudomonas aeruginosa populations sourced from cystic fibrosis lung infections

Abstract

Chronic, highly antibiotic-resistant infections in cystic fibrosis (CF) lungs contribute to increasing morbidity and mortality. Pseudomonas aeruginosa, a common CF pathogen, exhibits resistance to multiple antibiotics, contributing to antimicrobial resistance (AMR). These bacterial populations display genetic and phenotypic diversity, but it is unclear how this diversity affects susceptibility to bacteriocins. R-pyocins, i.e. bacteriocins produced by P. aeruginosa, are phage tail-like antimicrobials. R-pyocins have potential as antimicrobials, however recent research suggests the diversity of P. aeruginosa variants within CF lung infections leads to varying susceptibility to R-pyocins. This variation may be linked to changes in lipopolysaccharide (LPS), acting as the R-pyocin receptor. Currently, it is unknown how frequently R-pyocin-susceptible strains are in chronic CF lung infection, particularly when considering the heterogeneity within these strains. In this study, we tested R2-pyocin susceptibility of 139 P. aeruginosa variants from 17 sputum samples of seven CF patients and analyzed LPS phenotypes. We found that 83% of sputum samples did not have R2-pyocin-resistant variants, while nearly all samples contained susceptible variants. there was no correlation between LPS phenotype and R2-pyocin susceptibility, though we estimate that about 76% of sputum-derived variants lack an O-specific antigen, 40% lack a common antigen, and 24% have altered LPS cores. The absence of a correlation between LPS phenotype and R-pyocin susceptibility suggests LPS packing density may play a significant role in R-pyocin susceptibility among CF variants. Our research supports the potential of R-pyocins as therapeutic agents, as many infectious CF variants are susceptible to R2-pyocins, even within diverse bacterial populations.

Keywords: Pseudomonas aeruginosa; R-pyocin; bacteriocin; cystic fibrosis; lipopolysaccharide.

Figure 1.. R2-pyocin inhibition zones for multiple CF variants from different patient sputum samples.

The inhibition zones caused by R2-pyocin applied to soft agar overlays containing clinical variants from the CF sputum samples listed on the left. PAO1ΔrmlC serves as an R2susceptible control, and PAO1 wt as an R2-resistant control. Red boxes indicate patients with populations containing resistant variants. Up to ten variants per sputum sample were tested. An inhibition zone indicates R2-pyocin susceptibility, while the absence of a visible zone indicates resistance.

Figure 2.. R2-pyocin susceptibility varies among clinical P. aeruginosa populations and changes over time.

Relative susceptibility to R2-pyocins differs among variants in each population and across populations. Susceptibility heterogeneity is observed across all five patients and within the same patient’s longitudinal samples. There is no trend of strains becoming more resistant to R2-pyocins in longitudinally sampled populations, up to 2 years (consecutive numbers with collection number in parentheses). Two sputum samples contained at least one variant resistant to R2-pyocins (Relative Susceptibility = 0). Relative Susceptibility was determined by counting the number of serial dilutions (including neat) where inhibition zones appeared after R2-pyocins were applied to agar overlays. Means are shown with 95% confidence interval error bars.

Figure 3.. LPS phenotype heterogeneity and lack of correlation with R2-susceptibility.

A one-tailed Spearman correlation analysis for each phenotype revealed no correlations between R2-pyocin susceptibility and the absence of (A) common antigen (CPA), (C) O-specific antigen (OSA), or (E) an altered core (p > 0.05). CPA was absent in 42.73% of variants (B), OSA was absent in 76.36%, and (F) 27.27% of variants had altered LPS cores. Each data point represents the average Relative Susceptibility of a variant, and variants are colored based on their longitudinal collection.

Figure 4.. No correlation between combined LPS deficiency phenotypes and R2-pyocin susceptibility.

An LPS Score of 0 to 3 was assigned to each variant from each sputum collection to assess the correlation between overall LPS phenotypes and R2-pyocin susceptibility. Variants from each population are shown as data points of the same color. No correlation was found between LPS Score and R2-pyocin susceptibility (p > 0.05).