Figure 1.
PMNs induce mucoid conversion independent of bacterial uptake and the oxidative burst response.
A, opsonized PAO1algD-cat was incubated with either Hank's Buffered Saline (HBSS), HL-60 cells, or PMNs isolated from healthy human donors or CF patients followed by determination of the mucoid conversion frequency as described in Figure S1. B, phagocytosis of healthy PMNs was blocked by either physical separation from bacteria with transwells or treatment with cytochalasin D. In C and D, the PMN oxidative burst response was blocked by pretreatment of healthy PMNs with diphenyleneiodonium (DPI) (or DMSO, vehicle control) prior to incubation with PAO1algD-cat. The kinetic oxidative burst response of PMNs was measured by luminol (relative luminescent units (RLU)) in C, and the mucoid conversion frequency determined in B and D. All experiments were performed in triplicate on four to five independent occasions. Values are mean +/− standard error of the mean (SEM). For the statistical analysis of the kinetic oxidative burst response (C), the area under each curve was calculated ((RLU*min)2) for each treatment. Statistical analysis was carried out using an unpaired two-tailed student's t-test in B and C and Mann-Whitney test in A and D. (* p≤0.05, **p≤0.001).
Figure 2.
Non-oxidative PMN pathways promote mucoid conversion.
The mucoid conversion frequency was determined upon treatment of PAO1algD-cat with PMN lysates or granule preparations (A), sub-inhibitory concentrations (0.25 µM) of LL-37, human beta defensin 1 and 2 (hBD1/2) and human neutrophil peptide 1 (HNP1) (B), or sputum isolated from CF patients (C). In C, sputum was immune-depleted with a monoclonal LL-37 antibody or mouse IgG1 isotype control antibody. All experiments were performed in triplicate on four independent occasions. Values are mean +/− SEM. Statistical analysis was carried out comparing HBSS or 10 mM sodium phosphate buffer (pH 6.2) (SPB) to PMN component treated in A and B and anti-LL-37 to isotype control treated sputum in C, using an unpaired two-tailed Mann-Whitney test (* p≤0.05, **p≤0.001).
Figure 3.
LL-37 induces bacterial mutagenesis in a DinB-dependent manner.
A, mucA was expressed in PAO1algD-cat LL-37 1.1, 1.2 and, 2.1 on an arabinose-inducible plasmid (pHERD). Alginate was isolated from strains containing either empty pHERD (vector) or pHERDmucA (mucA+), grown on 0.5% arabinose, and the amount of alginate produced measured by a carbazole assay and compared to the parental non-mucoid PAO1algD-cat and mucoid PDO300. B, The frequency of rifampin resistance (RifR) of non-mucoid PAO1algD-cat (P. aeruginosa) and UTI89 (E. coli) following treatment with sub-inhibitory doses of LL-37 or H2O2 was determined. C, Non-mucoid PAO1algD-cat, PAO1algD-catΔdinB, PAO1algD-catΔmutS and PAO1algD-catΔmutSΔdinB were treated with sub-inhibitory LL-37 or SPB and the mucoid conversion frequency determined. Experiments were performed in triplicate on four independent occasions. Values are mean +/− SEM. Statistical analysis was carried out using an unpaired two-tailed student's t-test in A and C and Mann-Whitney test in B. (* p≤0.05, **p≤0.001).
Figure 4.
Sub-inhibitory concentrations of LL-37 enter the bacterial cytosol and interact with DNA.
Visualization of LL-37 localization in P. aeruginosa cells using confocal (A) and transmission electron microscopy (TEM, B). Bacteria were treated +/− sub-inhibitory LL-37, fixed and in A, permeabilized (top and bottom panels), stained for LL-37 (AlexaFluor647, red) (indicated by white arrows) and visualized by 100× objective. The percentage of cells counted on triplicate coverslips with LL-37 labeling inside the cell is indicated. In B, bacteria were treated with sub-inhibitory LL-37 (middle and bottom panels) or untreated (top) and cells were labeled with anti-LL-37 antibody conjugated to Protein G colloidal gold (20 nm). In the bottom panel, cells were additionally labeled with anti-dsDNA antibody conjugated to Protein G colloidal gold (10 nm). C: cytosol, M: membrane and E: extracellular. White arrows: cytosolic LL-37, Black arrows: membrane LL-37. In the bottom panel, boxed images are 2× magnified and LL-37 labeling is indicated by *. Twenty random, blind images were taken for each condition and the percentage of LL-37 labeling in the membrane and in the cytosol is indicated.
Figure 5.
LL-37 DNA binding promotes mucoid conversion.
A, homology modeling of LL-37 bound to B-DNA was performed manually on the basis of the backbone atomic coordinates of the homologous protein, sterol regulatory element binding protein, bound to DNA. In B, amino acid sequences of native LL-37 and synthetic derivatives are represented and the putative DNA binding region is indicated in yellow. Red: positive residues, blue: negative residues. In C, the percent of DNA bound by LL-37 derivatives was calculated from electrophoretic mobility shift assays (representative images in Figure S4), where densitometry was performed on each image using ImageJ. D represents the mucoid conversion frequency after treatment with LL-37 derivatives. Values are mean +/− SEM. Experiments were performed in triplicate on three independent occasions and statistical analysis was carried out using an unpaired two-tailed student's t-test (C) or Mann-Whitney test (D). (* p≤0.05).
Figure 6.
Alginate provides P. aeruginosa protection from lethal concentrations of LL-37.
The survival of P. aeruginosa strains (non-mucoid parental PAO1algD:cat (PAO1), mucoid PDO300, mucoid strains derived from LL-37 (1.2 and 2.1) or H2O2 (1.1) treatment (A), FRDΔalgD (non-mucoid) and FRD1 (mucoid) (B)) was determined following treatment with lethal concentrations of LL-37 (6.25 µM) and is represented as a fold increase from the non-mucoid isogenic strain. Experiments were performed in triplicate on three or four independent occasions. The mean +/− SEM is indicated. Statistical analysis was carried out to compare the survival of mucoid isolates compared to the non-mucoid isogenic strain (PAO1 or FRDΔalgD), using an unpaired two-tailed student's t-test to compare in (* p≤0.05).
Figure 7.
Proposed model of LL-37 induced mutagenesis.
At sub-inhibitory concentrations, LL-37 can penetrate P. aeruginosa cells and enter the bacterial cytosol, where LL-37 dimers then bind to DNA. DNA binding by LL-37 then promotes DinB-dependent replication, which potentiates mutations in mucA leading to mucoid conversion. Alginate overproducing bacteria are then protected from lethal concentrations of LL-37 and mucoid variants are selected for and persist in CF.