Fig 1.
Identification of biofilms at the inner surfaces of the catheters removed from candidemia patients.
The catheters were taken for cross-section imaging by bright field microscopy. Magnified images of the inner and outer surfaces of the catheters were shown in red and blue boxes as indicated. Scale bar, 500 μm for whole catheters, and 20 μm for magnified inner and outer surfaces.
Table 1.
Clinical summary of the patients.
Fig 2.
Detection of fungal cells at the inner surfaces of the catheters.
The catheters were fixed with paraformaldehyde followed by paraffin embedding and sectioning to 10 μm thickness. (A) Periodic acid-Schiff (PAS) staining was performed to detect fungal cells in the catheters. Fungal cells are indicated by yellow arrows. (B) Hematoxylin and eosin staining (H&E) was applied for detection of human cell components. Polymorphonuclear cells associated with a fibrous structure are indicated by red arrows. Scale bar, 500 μm for whole catheters, and 50 μm for magnified images at the inner surfaces of the catheters.
Fig 3.
Infiltration of human immune cells and their association with Candida cells at the inner surfaces of the catheters.
Whole mount catheters (2 mm in length) were fixed and stained for the presence of Candida cells and human immune cells, followed by observation under a two-photon microscope. Single xy-plane images are shown. Candida cells and human CD45+ F-actin+ immune cells associated with Candida antigens are indicated by white and magenta arrows, respectively. CD45+ F-actin- immune cells showing actin cytoskeleton disassembly are indicated by yellow arrows. Scale bar, 20 μm.
Fig 4.
Neutrophil extracellular traps were associated with Candida cells in the catheters.
The catheters were processed as described in Fig 3. Antibodies against citrullinated histone H3 (C-H3) were used to detect neutrophil extracellular traps (NETs). Candida cells were stained by Calcofluor white. Immune cells were identified by F-actin staining with phalloidin. (A) Single xy-plane and reconstituted 3D (z = 60 μm) images are shown. Candida cells associated with NETs are indicated by magenta arrows, while Candida cells free from NETs are indicated by white arrows. Scale bar, 20 μm. (B) Violin plot for comparison of NETs quantified by the volume of C-H3 signals in the stacked images obtained from the catheters. n = 3 to 9. Each point represents one stacked image.
Fig 5.
Morphology of the catheter isolates cultured in RPMI or TPN.
C. albicans cells were cultured in RPMI medium (A) or TPN (B) for 24 hours at 37°C, 180 rpm. The cells were fixed by 4% paraformaldehyde followed by Concanavalin A (Con A) staining to detect mannan. Represented images are shown. BF, bright field. Scale bar: 20 μm.
Fig 6.
The C. albicans catheter isolates exhibited reduced expression of specific filamentation-associated genes.
C. albicans yeast cells were cultured in RPMI (A) or TPN (B) at 37°C for 24 hours. Gene expression was quantified by qRT-PCR and normalized to ACT1 expression. Data represent mean with SD from three independent experiments. Statistical analysis was performed with the one-way ANOVA, followed by Dunnett’s multiple comparison test as compared to the reference strain SC5314. *, P<0.05. **, P<0.01. ***, P<0.001. ****, P<0.0001. NS, no significant difference.
Fig 7.
The catheter isolates exhibited diverse capacity in resistance to neutrophil-mediated fungal biofilm removal.
C. albicans cells were cultured in TPN for 24 hours to allow biofilm formation on a coverslip. The biofilms developed in TPN were then incubated with mouse bone marrow-derived neutrophils (BMN) in TPN in the absence or presence of candidalysin (CaL, 15 μM) for another 6 hours before fixation and staining. BMN NETosis was observed by Ly6G (cyan), F-actin (white) and citrullinated histone H3 staining (C-H3, red). Candida cells were stained by mDectin-1-Fc (green) that recognizes β-glucan on fungal cell wall. (A) Representative images are shown. (B) Candida biofilms were quantified as the area of C. albicans cells in each field on the coverslips. (C) Infiltration of BMNs (Ly6G+) into the surface-associated biofilms of each isolate is shown. (D and E) The percentage of NETotic BMNs (C-H3+ Ly6G+ cells/ total Ly6G+ cells) (D) and viable non- NETotic (F-actin+ C-H3- Ly6G+ cells/ total Ly6G+ cells) (E) in the biofilms is shown. All the images were analyzed with Imaris (Bitplane). (B-E) Results shown are mean with SD from 3 independent experiments, 2 images from each coverslip. Statistical analysis was performed with the one-way ANOVA, followed by Tukey’s multiple comparison test (B) or a two-tailed unpaired Student T test (C-E). *, P<0.05. **, P<0.01. ***, P<0.001. ****, P<0.0001. NS, no significant difference. Scale bar, 50 μm.
Fig 8.
Candidalysin induced NETosis of biofilm-infiltrating neutrophils and removal of Candida biofilms.
Bofilm formation assay of the reference strain SC5314, the candidalysin deficient isogenic mutants (ece1Δ/Δ, ece1Δ/Δ+ECE1Δ184–279 [I] and ece1Δ/Δ+ECE1Δ184–279 [II]), and the ECE1 revertant strains (ece1Δ/Δ+ECE1 [I] and ece1Δ/Δ+ECE1 [II]) was performed as described in Fig 7. (A) Representative images are shown. (B) Candida biofilms were quantified as the area of C. albicans cells (Fc-mDectin-1+) in each field on the coverslips. (C) Infiltration of BMNs (Ly6G+) into the surface-associated biofilms of each isolate is shown. (D and E) The percentage of NETotic BMNs (C-H3+ Ly6G+ cells/ total Ly6G+ cells) (D) and viable non- NETotic (F-actin+ C-H3- Ly6G+ cells/ total Ly6G+ cells) (E) in the biofilms is shown. All the images were analyzed with Imaris (Bitplane). (B-E) Results shown are mean with SD from 3 independent experiments, 2 images from each coverslip. Statistical analysis was performed with the one-way ANOVA, followed by Tukey’s multiple comparison test (B) or a two-tailed unpaired Student T test (C-E). **, P<0.01. ***, P<0.001. ****, P<0.0001. NS, no significant difference. Scale bar, 50 μm.
Fig 9.
A model for the function of candidalysin in inducing NETosis and C. albicans biofilm elimination in an intravascular catheter.
The biofilms of C. albicans clinical isolates exhibited different ability in promoting NETosis. Supplementation of candidalysin induced NETosis and facilitated C. albicans biofilm elimination Whereas, ablation of candidalysin expression in the high candidalysin-producing strain reduced NETosis and increased C. albicans persistence on a surface. Created with Biorender.com.
Table 2.
Primers used in this study.