Extracellular DNA release, quorum sensing, and PrrF1/F2 small RNAs are key players in Pseudomonas aeruginosa tobramycin-enhanced biofilm formation

Abstract

Biofilms are structured microbial communities that are the leading cause of numerous chronic infections which are difficult to eradicate. Within the lungs of individuals with cystic fibrosis (CF), Pseudomonas aeruginosa causes persistent biofilm infection that is commonly treated with aminoglycoside antibiotics such as tobramycin. However, sublethal concentrations of this aminoglycoside were previously shown to increase biofilm formation by P. aeruginosa, but the underlying adaptive mechanisms still remain elusive. Herein, we combined confocal laser scanning microscope analyses, proteomics profiling, gene expression assays and phenotypic studies to unravel P. aeruginosa potential adaptive mechanisms in response to tobramycin exposure during biofilm growth. Under this condition, we show that the modified biofilm architecture is related at least in part to increased extracellular DNA (eDNA) release, most likely as a result of biofilm cell death. Furthermore, the activity of quorum sensing (QS) systems was increased, leading to higher production of QS signaling molecules. We also demonstrate upon tobramycin exposure an increase in expression of the PrrF small regulatory RNAs, as well as expression of iron uptake systems. Remarkably, biofilm biovolumes and eDNA relative abundances in pqs and prrF mutant strains decrease in the presence of tobramycin. Overall, our findings offer experimental evidences for a potential adaptive mechanism linking PrrF sRNAs, QS signaling, biofilm cell death, eDNA release, and tobramycin-enhanced biofilm formation in P. aeruginosa. These specific adaptive mechanisms should be considered to improve treatment strategies against P. aeruginosa biofilm establishment in CF patients' lungs.

Keywords: Antimicrobials; Biofilms; Pathogens.

Fig. 1

Effect of sub-MICs of tobramycin on biofilm formation by P. aeruginosa. a CLSM images of 24-h-old biofilms as a function of different concentrations of tobramycin. For each concentration, a 3D view along the x, y and z axes is displayed. Images show representative data from at least three independent biofilm assays. Scale bars = 20 µm. b COMSTAT image analyses were performed to determine maximum thicknesses (μm), average thicknesses (μm), and total biovolumes (μm³ μm⁻²). The error bars represent the standard error of the means (SEMs) and are the result of the analysis of three views of each of the three independent biological assays. Statistics were achieved by a two-tailed t test: ★★★, P = 0.0001 to 0.001; ★★, P = 0.001 to 0.01; ★, P = 0.01 to 0.05; NS (not significant), P ≥ 0.05

Fig. 2

Sub-MIC of tobramycin leads to increased eDNA release and cell death in P. aeruginosa biofilm. a Representative CLSM 3D-top and side views of eDNA accumulation in 24-h-old P. aeruginosa biofilms exposed to tobramycin (0.8 μg ml⁻¹) alone, DNase I (100 μg ml⁻¹) alone or tobramycin and DNase I simultaneously, compared to untreated biofilms. Prior to image acquisition by CLSM, P. aeruginosa biofilm cells were labeled in green with SYTO 9 and the eDNA was stained in red with DDAO. Scale bars = 20 µm. CLSM images were analyzed using the COMSTAT software to quantify b the eDNA relative abundances relatively to biofilm biovolume values. Error bars represent standard error of the means (n = 3) and c the biofilm biovolumes. Error bars represent standard error of the means (n = 3). d CLSM micrographs of P. aeruginosa biofilms grown in the absence of tobramycin (left panel) and in the presence of drug (right panel) stained using the LIVE/DEAD^® BacLight^TM Bacterial Viability Kit. Green fluorescent cells are viable, whereas red fluorescent cells have compromised cell membranes. Scale bars = 20 µm. e The cell death in biofilms was determined by COMSTAT images analyses. Values of nonviable biovolumes were normalized to total biovolumes. Error bars represent standard error of the means (n = 3). Statistics were achieved by a two-tailed t test: ★★, P = 0.001 to 0.01; ★, P = 0.01 to 0.05

Fig. 3

Protein−protein interaction network for highly differentially accumulated proteins in the P. aeruginosa biofilm cultures exposed to sub-MIC level of tobramycin. Nodes are colored according to the protein abundance (fold change). Nodes highlighted in green color correspond to the overaccumulated proteins whereas the nodes in red color represent the underaccumulated proteins. Edges indicate protein−protein interactions

Fig. 4

Tobramycin-enhanced biofilm formation is associated with increased production of QS molecules. a Quantification of the two main AHLs (C₄-HSL and 3-oxo-C₁₂-HSL) of Rhl and Las QS systems by LC-MS/MS analysis. Error bars represent standard error of the means (n ≥ 3). b Quantification of HAQ molecules (HHQ, PQS, and HQNO) by LC-MS/MS analysis. Error bars represent standard error of the means (n ≥ 3). c Representative CLSM 3D-top micrographs of 24-h-old biofilms of P. aeruginosa H103 and ΔpqsA mutant strains exposed to tobramycin (0.8 μg ml⁻¹) compared to untreated biofilms. Scale bars = 20 µm. CLSM images were analyzed using COMSTAT software to quantify d biofilm biovolumes. Error bars represent standard error of the means (n = 3) and e eDNA relative abundances. Error bars represent standard error of the means (n = 3). eDNA values were normalized to biofilm biovolumes. Asterisks indicate a significant difference as determined by a two-tailed t test: ★★★, P = 0.0001 to 0.001; ★★, P = 0.001 to 0.01; ★, P = 0.01 to 0.05; NS (not significant), P ≥ 0.05

Fig. 5

Role of PrrF1/F2 sRNAs in tobramycin-enhanced biofilm formation. a Relative prrF mRNA levels in P. aeruginosa biofilms exposed to tobramycin (brown bars) compared to the relative expression of mRNA levels in the control condition (green bars), after 24 h of growth. Quantifications have been obtained from at least three independent experiments and rpoD was used as a control housekeeping gene. Error bars represent standard error of the means (n = 3). b Representative CLSM 3D micrographs of 24-h-old biofilms P. aeruginosa H103 and ΔprrF mutant strains exposed to tobramycin (0.8 μg ml⁻¹) compared to untreated biofilms. Scale bars = 20 µm. CLSM images were analyzed using COMSTAT software to quantify c biofilm biovolumes. Error bars represent standard error of the means (n = 3) and d eDNA relative abundances. eDNA values are normalized to biofilm biomass. Error bars represent standard error of the means (n = 3). e HAQs levels of the ΔprrF mutant and its isogenic parent strain H103 grown with or without tobramycin supplementation, as determined by LC-MS/MS. Error bars represent standard error of the means (n = 3). Statistics were achieved by a two-tailed t test: ★★, P = 0.001 to 0.01; ★, P = 0.01 to 0.05; NS (not significant), P ≥ 0.05

Fig. 6

Sub-MIC of tobramycin enhance iron/heme uptake strategies in P. aeruginosa biofilm cultures. Relative pvdS, pvdH, hasI, phuR, and feoB mRNA levels in P. aeruginosa biofilm cultures exposed to sub-MIC of tobramycin (green bars) compared to the relative mRNA levels in the control condition (brown bars), after 24 h of growth. Quantifications have been obtained from at least three independent experiments and rpoD was used as a control housekeeping gene. Error bars represent standard error of the means (n = 3). Statistics were achieved by a two-tailed t test: ★★★, P = 0.0001 to 0.001; ★★, P = 0.001 to 0.01; ★, P = 0.01 to 0.05