Exploring aggregation genes in a P. aeruginosa chronic infection model

Abstract

Bacterial aggregates are observed in both natural and artificial environments. In the context of disease, aggregates have been isolated from chronic and acute infections. Pseudomonas aeruginosa (Pa) aggregates contribute significantly to chronic infections, particularly in the lungs of people with cystic fibrosis (CF). Unlike the large biofilm structures observed in vitro, Pa in CF sputum forms smaller aggregates (~10-1,000 cells), and the mechanisms behind their formation remain underexplored. This study aims to identify genes essential and unique to Pa aggregate formation in a synthetic CF sputum media (SCFM2). We cultured Pa strain PAO1 in SCFM2 and LB, both with and without mucin, and used RNA sequencing (RNA-seq) to identify differentially expressed genes. The presence of mucin revealed 13 significantly differentially expressed (DE) genes, predominantly downregulated, with 40% encoding hypothetical proteins unique to aggregates. Using high-resolution microscopy, we assessed the ability of mutants to form aggregates. Notably, no mutant exhibited a completely planktonic phenotype. Instead, we identified multiple spatial phenotypes described as "normal," "entropic," or "impaired." Entropic mutants displayed tightly packed, raft-like structures, while impaired mutants had loosely packed cells. Predictive modeling linked the prioritized genes to metabolic shifts, iron acquisition, surface modification, and quorum sensing. Co-culture experiments with wild-type PAO1 revealed further spatial heterogeneity and the ability to "rescue" some mutant phenotypes, suggesting cooperative interactions during growth. This study enhances our understanding of Pa aggregate biology, specifically the genes and pathways unique to aggregation in CF-like environments. Importantly, it provides insights for developing therapeutic strategies targeting aggregate-specific pathways.

Importance: This study identifies genes essential for the formation of Pseudomonas aeruginosa (Pa) aggregates in cystic fibrosis (CF) sputum, filling a critical gap in understanding their specific biology. Using a synthetic CF sputum model (SCFM2) and RNA sequencing, 13 key genes were identified, whose disruption led to distinct spatial phenotypes observed through high-resolution microscopy. The addition of wild-type cells either rescued the mutant phenotype or increased spatial heterogeneity, suggesting cooperative interactions are involved in aggregate formation. This research advances our knowledge of Pa aggregate biology, particularly the unique genes and pathways involved in CF-like environments, offering valuable insights for developing targeted therapeutic strategies against aggregate-specific pathways.

Keywords: CF; Pseudomonas aeruginosa; aggregate; chronic infection.

Fig 1

Differentially expressed genes in aggregates. (a) Volcano plot of differentially expressed genes in SCFM + mucin compared to SCFM-mucin at 8 hours of growth. Upregulated genes are shown in red, downregulated genes are shown in blue, and non-significant genes are shown in gray. 50 most significantly differentially expressed genes are labeled. (b) Differentially expressed genes in SCFM + mucin compared to LB + mucin at 8 hours. (c) Heatmap of genes of interest and associated genes in their operon/pathway at 8 hours. Heatmap also includes several classical biofilm-associated genes such as polysaccharides and QS genes that notably do not show differential expression in aggregates in this study.

Fig 2

Transposon mutant aggregate phenotypes. (a) Examples of aggregation, with WT shown in green and two representative mutants shown in red. Transposon mutant PA2111 is an example of the entropic phenotype, with long chains of stacked cells oriented along the coverslip. Transposon mutant PA0998 is an example of an impaired phenotype, with slow growth and loosely packed cells that become most apparent at later timepoints. Scale bar is 10 µm (b) Average aggregate volume of transposon mutants over time. Mutants are grouped by phenotype. Mutant aggregates differ significantly from WT but vary over time. Three biological replicates ± SEM, significance calculated using Kruskal-Wallis (P value < 0.0001) with multiple comparisons test (alpha 0.05).

Fig 3

Proposed pathways of genes of interest. Hypothetical gene functions and pathways were predicted using the computational pipeline. Genes have been grouped by predicted function. Upregulated functions are shown in green, downregulated functions are shown in red. Gene names within function indicate associated genes that were present in our data set. The figure was created with Biorender.

Fig 4

WT and transposon mutant co-culture. Confocal laser scanning microscopy of WT PAO1 (green) and Tn-mutants (red) co-cultured in SCFM2 at a 1:1 ratio after 5 hours. Images are representative of each mix of which there were three biological replicates. The scale bar is 5 µm.

Fig 5

Comparisons of WT (solid bars) and transposon mutant (striped bars) (a) total biomass (b) average aggregate volume, and total volume of planktonic cells after 15 hours growth in SCFM2. Mutants are grouped by phenotype. Three biological replicates ± SEM, significance of WT vs. mutants was calculated using ordinary two-way ANOVA (P value < 0.0001) with Fischer’s LSD multiple comparisons test (alpha 0.05).