Figure 1.
Multi-Drug–Resistant P. aeruginosa Clinical Isolates Display Unusual Appendage-Like Structures on the Cell Surface
Transmission electron microscopy (TEM) images of MDR virulent clinical isolates at magnification of 15,000 × 1.4.
(A) Strain MDR1.
(B) Strain MDR13.
(C) Strain MDR25.
(D) Strain MDR26.
Novel appendage-like structures are shown by black arrows. For comparison, flagella are indicated by grey arrows.
Table 1.
Virulence Characteristics of MDR P. aeruginosa Clinical Isolates Assessed by Their Ability to Adhere to (% Adhesion) and Alter the Transepithelial Electrical Resistance (% Decrease in TER) of Caco-2 Cells [2]
Figure 2.
Identification of Appendage-Like Structures
(A, B) Fractionation of sheared cell surface proteins by 10% Tris-Glycine SDS-PAGE and identification of abundant proteins by N-terminal sequencing.
(A), Lane 1, molecular weight standards (Bio-Rad, broad range, pre-stained); lane 2, strain MDR25; lane 3, strain MDR1. Black arrows show proteins of strain MDR1 subjected to N-terminal sequencing.
(B) Lane 1, molecular weight standards (Invitrogen, SeeBlue Plus 2 pre-stained); lane 2, strain 13. Black arrow shows protein subjected to N-terminal sequencing.
(C) Immunoblot analysis of cell surface sheared proteins isolated from strains MDR1, MDR13, MDR25, and MPAO1 by using affinity purified polyclonal anti-PA5369 antibody.
(D) Immunoblot analysis of pellet and sheared surface proteins from strains MPAO1, MPAO1 derivative PA5369 mutant, and clinical strain MDR25. Colonies were collected and suspended in 200 μl PBS up to final density of 5.0 (600 nm) followed by vortexing for 2 min and centrifuging at 5,000g, 5 min. Sheared material in 10 μl of supernatant was denatured by boiling with 10 μl sample buffer and 10 μl was loaded on gel wells to be separated by 10% glycine SDS-PAGE. The remaining pellet was resuspended in 50 μl of sample buffer, boiled, and 2 μl were loaded on gel wells. PstS protein was identified by immunoblot using anti-PA5369 antibody.
(E) Enzyme-linked immunosorbent assay (ELISA) to detect PA5369-like protein on cell surfaces of MPAO1 and MDR clinical P. aeruginosa strains. Data are mean ± SD (n = 3).
(F, G) TEM images of strain MDR25 grown on (F) PIA supplemented with 1 mM K-phosphate buffer and (G) on PIA.
(H, I) Immunoblot of (H) sheared surface proteins and (I) cell pellet proteins in strain MDR25 grown on PIA (low phosphate) versus PIA supplemented with 1 mM K-phosphate buffer (high phosphate).
(J) Immuno-electron microscopy of strain MDR25. Gold spots identified with arrows indicate the protein recognized by anti-PA5369 antibody on cell surface structures.
Figure 3.
PA5369 Is Involved in Adhesiveness and Ability of MDR25 and MPAO1 to Disrupt Epithelial Integrity
(A) Anti-PA5369 antibodies decrease the adhesiveness of strain MDR25 to Caco-2 monolayers. Strain MDR25 was grown on PIA plates for 2 days, bacteria were harvested, resuspended in PBS to a cell density of 5.0, and 2 μl was added to the apical surface of confluent Caco-2 cells. The Fab fragments of anti-PA5369 antibody at dilution 1:100 were included in the DMEM media (when needed) at the point of inoculation. Following 1 hour of incubation at 37°C, 5% CO2, the percentage of adherent cell was calculated. Data are mean ± SD (n = 3), p < 0.001 (Student t-test).
(B) Anti-PA5369 antibodies attenuate the ability of strain MDR25 to disrupt Caco-2 monolayer integrity. Strain MDR25 was grown on PIA plates for 2 days, bacteria were harvested, resuspended in PBS to a cell density of 5.0, and 2 μl was added to the apical surface of confluent Caco-2 cells grown in transwells. The Fab fragments of anti-PA5369 antibody at dilution 1:100 were included in the DMEM media at the point of inoculation. Transepithelial resistance (TER) was measured dynamically as previously described [2]. The control group reflects the TER of Caco-2 cells in the absence of MDR25. Data are mean ± SD (n = 3), p < 0.005 (Student t-test).
(C) Detection of surface associated PstS by ELISA in strain MPAO1 and derivative strains. Data are mean ± SD (n = 3), p < 0.001 (Student t-test).
(D) PA5369 is involved in the adhesiveness of MPAO1 to Caco-2 cell monolayers. MPAO1 and its derivative strains were grown in TSB overnight, and 2 μl added to the apical surface of confluent Caco-2 cells. Following 1 hour of incubation at 37°C, 5% CO2, the percentage of adherent bacteria was calculated as previously described [2]. Data are mean ± SD (n = 3), p < 0.005 (Student t-test).
(E) PA5369 is involved in the ability of MPAO1 to disrupt Caco-2 monolayer integrity. MPAO1 and its derivative strains were grown in TSB overnight, and 2 μl added to the apical surface of confluent Caco-2 cells grown in transwells. Transepithelial resistance (TER) of Caco-2 cells was measured after 7 hours of co-incubation with P. aeruginosa strains. Data are mean ± SD (n = 3), p < 0.005 (Student t-test)
Figure 4.
Colony Morphology Variation in MDR25 Correlates with Production of Outer Surface-Exposed PstS and Virulence
(A) The appearance of smooth edged colonies as shown by black arrows in MDR25 grown on PIA, Gm 50 μg/ml.
(B) A single smooth colony was isolated, cultivated in PB, Gm 50 μg/ml, and plated on PIA, Gm 50 μg/ml. The appearance of rough edged colony is shown by the white arrow.
(C) Biofilm production by rough (MDR25R) and smooth (MDR25S) strains grown in PB, Gm 50 μg/ml.
(D) Expression of outer surface PstS in MDR25R and MDR25S grown on PIA, Gm 50 μg/ml.
(E) Kaplan-Meier survival curves of mice intestinally inoculated with MDR25S or MDR25R demonstrating attenuated killing effect of MDR25S. n = 10/group, p = 0.021.
Figure 5.
Intestinal Phosphate Supplementation Prevents the Lethality of MDR Strains in Mice
Seven-week-old male C57BL6 mice were given either water (no phosphate supplementation), or 2 mM PBS solution (2 mM phosphate supplementation), or 10 mM PBS solution (10 mM phosphate supplementation) for 36 hours prior to hepatectomy. P. aeruginosa MDR25 was grown overnight in Pseudomonas broth (PB) containing Gm, 50 μg/ml, cells were collected by centrifugation, resuspended in either water containing 10% glycerol (group 1, n = 8), or 2 mM PBS (group 2, n = 8) or 10 mM PBS (group 3, n = 5) to final OD600 nm 0.2, and 200 μl of bacterial solutions were injected into cecum immediately after hepatectomy. The abdomen was closed and mice were allowed to drink either water (group 1) or 2 mM PBS (group 2) or 10 mM PBS (group 3) and followed for mortality. Data were analyzed using Kaplan-Maier survival curves with SPSS software and significance tested by Log-rank (Mantel-Cox) p = 0.004. Reiterative experiments were performed with the clinical isolate MDR1 with the exception that only 10 mM of phosphate supplementation was tested (n = 10/group, p = 0.001).
Figure 6.
Alternative Type II Secretion System Is Involved In Outer Surface Expression of PstS
(A) Clustering of pstS pa5369 to the P. aeruginosa PAO1 genome.
(B) Clustering of pstS pa0688 to the P. aeruginosa PAO1 genome.
(C) Clustering of pstS pa14–55410 to P. aeruginosa PA14 genome.
(D) Immunoblot analysis of pellet and sheared surface proteins from strains MPAO1, MPAO1 derivative PA0686 (ΔHxcR) mutant, and ΔHxcR complemented mutant. Strains were grown on PIA, and colonies collected and suspended in 200 μl PBS containing protease inhibitor cocktail (Sigma) up to final density of 5.0 (600 nm) followed by vortexing for 2 min and centrifuging at 5,000g, 5 min. Sheared material in 10 μl of supernatant was denatured by boiling with 10 μl sample buffer and 10 μl was loaded on gel wells to be separated by 10% glycine SDS-PAGE. The pellet was resuspended in 50 μl of PBS containing protease inhibitor cocktail, and 150 μl of BugBuster Master Mix (Novagen) was added to lyze the bacterial cells. Debris was removed by centrifugation, and 10 μl of supernatant was denatured by boiling with 10 μl sample buffer and 10 μl was loaded on gel. PstS protein was identified by immunoblot using anti-PA5369 antibody.
(E) Expression of pstS and pstS-related genes under phosphate limitation determined by Real-Time RT-PCR. Gene expression was performed in triplicate with data representing the mean, and three independent experiments run showing similar expression patterns. The (-RT) controls were determined to be at the level of blank control (no template included).