Fig 1.
The MAC and lysozyme trigger E. coli cell wall degradation in human serum.
A) Flow cytometry plots (FSC/SSC) of the number of E. coli particles in 10 μl after exposure to buffer, 5% serum or Δlysozyme serum with or without 5 μg/ml lysozyme for 60 minutes at 37°C. A SSC threshold was set based on untreated bacteria to filter out background noise and to determine changes in FSC/SSC upon treatment with different serum conditions. B) Cell count in 10 μl incubation volume of E. coli treated with a concentration range of nhs or Δlysozyme serum with or without 5 μg/ml lysozyme for 60 min 37°C. The number of cells represent the events in 10 μl sample that were measured within the conditions depicted in A. C) Cell count in 10 μl incubation volume of E. coli treated with buffer or 5% Δlysozyme serum in the presence or absence of 20 μg/ml OmCI and a titration of lysozyme or heat inactivated (HI) lysozyme for 60 min 37°C. D) Flow cytometry histograms of outer membrane damage (left: mCherry) and inner membrane damage (right: Sytox blue) of bacteria treated with buffer, 1% serum with or without 20 μg/ml OmCI or 1% Δlysozyme serum. A, D) Flow cytometry plots and histograms represent data of three independent experiments. B, C) Data represent mean ±SD of 3 independent experiments.
Fig 2.
The MAC and lysozyme in serum trigger alterations in the morphology of E. coli.
A) Schematic representation of Cy3-labeling of E. coli LPS via click chemistry. Bacteria are incubated with KDO-azide (green), which is incorporated into LPS. A DBCO (blue) linked to the fluorophore Cy3 (red) can subsequently react with the azide group via click chemistry. OM = outer membrane, PG = peptidoglycan, IM = inner membrane. B) The number of Cy3-positive E. coli cells in 10μl sample after exposure to a concentration range of serum or Δlysozyme serum with or without 5 μg/ml lysozyme. Samples were measured without a SSC threshold. C) Flow cytometry plots (FSC/SSC) of Cy3-positive E. coli particles in 10 μl sample after exposure to buffer, 5% nhs, or 5% Δlysozyme serum with or without 5 μg/ml lysozyme for 60 minutes at 37°C. D) Cy3-labeled E. coli was treated with a concentration range of serum or Δlysozyme serum with or without 5 μg/ml lysozyme for 60 min 37°C. For each condition, the number of Cy3-positive particles within the FSC/SSC gate of untreated bacteria was quantified in 10μl. B-D) The lysozyme concentration that was used exceeds the concentration in 5% serum by a factor 100. B, D) Data represent mean ±SD of 3 independent experiments. C) Flow cytometry plots represent data of three independent experiments.
Fig 3.
The combination of the MAC and lysozyme alters the cell shape of E. coli from rod-shaped to spherical.
Confocal microscopy images of PerimCherry/cytoGFP E. coli bacteria that were immobilized onto poly-L-lysin coated coverslips and treated with A) 5% serum or B) 5% Δlysozyme serum with (bottom) or without (top) 0.5 μg/ml lysozyme. All incubations were done in the presence of To-pro-3 as a readout for inner membrane damage. Images were taken after A) 0, 10 and 25 minutes or B) 0 and 25 minutes at 37°C. In C), bacteria were treated with 5% Δlysozyme serum with 0.5 μg/ml lysozyme in the presence of 20 μg/ml OmCI to block MAC formation. B, C) The lysozyme concentration that was used exceeds the concentration in 5% serum by a factor 10. A-C) Scale bars: 20 μm. Images represent data of three independent experiments.
Fig 4.
A completely assembled MAC pore sensitizes E. coli for cell wall degradation by lysozyme and hGIIA.
A) Flow cytometry plots (FSC/SSC) and B) particle count of PerimCherry/cytoGFP E. coli that was pre-treated with buffer or 10% ΔC9 serum and, after washing (indicated with an @), exposed to buffer, 100 nM C9 and/or a concentration range of lysozyme for 60 minutes. In A, the flow cytometry plots of 0 and 5 μg/ml lysozyme are depicted. B) A gate was set on untreated bacteria as depicted in A, after which the number of particles was counted within those gates. C) Outer membrane damage (left: mCherry signal (relative to the mean of the mCherry signals of the 0 nM C9 controls)) and particle count (right) of PerimCherry/cytoGFP E. coli that were treated with 10% ΔC9 serum, and after washing, exposed to a concentration range of C9 in the presence or absence of 5 μg/ml lysozyme for 60 minutes. Particle count represents the number of particles in the FSC/SSC gate of untreated bacteria. D) GFP intensity of PerimCherry/cytoGFP E. coli that was pre-treated with 10% ΔC9 serum and, after washing, exposed to a concentration range of C9 in the absence (control) or presence of 5 μg/ml lysozyme and/or 1 μg/ml hGIIA for 30 minutes. The GFP intensity (Geomean) of the total number of events is presented. A) Flow cytometry plots represent data of three independent experiments. B-D) Data represent mean ±SD of 3 independent experiments.
Fig 5.
The MAC sensitizes E. coli to killing by neutrophils.
A) Confocal microscopy images of PerimCherry/cytoGFP E. coli (green) treated with 5% Δlysozyme serum for 60 minutes at 37°C. After washing, bacteria were incubated with buffer (left) or neutrophils (red, right) for 20 minutes at 37°C, fixed in 1.5% paraformaldehyde and imaged. B) Confocal images of PerimCherry/cytoGFP E. coli (green) that were pre-labeled with 10% ΔC5 serum (for deposition of C5 convertases), washed and exposed to buffer or C5-C9 for 30 minutes at 37°C. 10 nM C5 and C6, 20 nM C7 and C8 and 100 nM C9 was used. After washing, bacteria were exposed to neutrophils for 20 minutes at 37°C, fixed in 1.5% paraformaldehyde and imaged. A, B) Neutrophils membranes (red) were stained with Alexa-647 labeled Wheat Germ Agglutinin. C) Relative GFP intensity of neutrophils after phagocytosis of PerimCherry/cytoGFP E. coli. Bacteria were pre-labeled with 10% ΔC8 serum and, after washing, exposed to a concentration range of C8 in the presence or absence of 2.5 nM C9 for 30 minutes at 37°C. After washing, bacteria were incubated with neutrophils for 20 minutes at 37°C, after which the GFP signal of the neutrophils was analyzed by flow cytometry. GFP intensity was normalized against the GFP intensity of bacteria that were treated with ΔC8 serum only. Neutrophils without bacteria (green dot) served as basal value for the GFP signal. D) Bacterial viability (CFU/ml) of E. coli that was pre-treated with MAC components similar to C, and exposed to buffer or neutrophils. After 20 minutes, neutrophils were lysed for 15 minutes in MQ, after which bacterial survival was determined (CFU/ml). C, D) Bacteria/neutrophil ratio used: 10/1. A, B) Images represent two (A) or four (B) independent experiments in which the 488-laser settings were adjusted to the GFP intensity. C, D) Data represent mean ±SD of 3 independent experiments. D) Statistical analysis was done using a ratio paired t-test in which each condition was compared to the buffer control (no C9 and no neutrophils) within the same C8 concentration. Significance was displayed only when significant as *P ≤ 0.05 or **P≤ 0.01.