Candida albicans biofilm extracellular vesicles deliver candidalysin to epithelial cell membranes and induce host cell responses

Abstract

Extracellular vesicles (EVs) are heterogeneous particles encapsulated with a phospholipid bilayer membrane. EVs have evolved diverse biological functions, serving mainly as prominent mediators and regulators of cell-cell communication. This study investigated whether candidalysin, a key virulence factor in Candida albicans infections, is present within EVs derived from C. albicans biofilms and retains activity by inducing host immune responses. We found that biofilm EVs contain candidalysin and can permeabilize planar lipid bilayer membranes in a dose-dependent manner. However, biofilm EVs were unable to damage oral epithelial cells (OECs) but were able to induce cytokine responses. Notably, EVs obtained from biofilms cultured for 24 h and 48 h exhibited differences in cargo composition and their ability to activate OECs. This study highlights the potential of biofilm EVs as a toxin delivery system during C. albicans infection and identifies temporal differences in the ability of EVs to activate epithelial cells.

Keywords: Candida albicans; candidalysin; cell membranes; extracellular vesicles; fungal infection; host response; host-pathogen interactions; membrane transport.

Fig 1

Candidalysin-containing EVs are unable to cause cellular damage. A scanning electron microscopy (SEM) image of EVs (green) and extracellular matrix (pink) on the surface of C. albicans growing as yeast (a) and biofilm (b). Scale bars represent 200 nm. (c) Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis of EVs obtained from C. albicans WT and ece1Δ/Δ biofilms cultured for 24 h and 48 h. Average peptide spectrum match (PSM) values (n = 4 biological repeats from WT-EVs and n = 1 from ece1Δ/Δ-EVs) are shown with standard deviation. No candidalysin was detected from ece1Δ/Δ-EVs. (d) Epithelial cell damage was monitored with biofilm EVs. WT-24H, WT-48H, ece1Δ/Δ-24H, and ece1Δ/Δ-48H EVs of 3 × 10¹² p/mL were applied to oral epithelial cells for 24 h and lactate dehydrogenase activity was quantified as a marker of cellular damage. As a positive control, epithelial cells were treated with candidalysin peptides of 70 µM. Data are shown as the mean ± standard deviation of n = 3 biological repeats. Statistical significance was calculated by unpaired t-test compared to vehicle-treated cells (P **** <0.0001).

Fig 2

Candidalysin-containing EVs permeabilize DPhPC bilayer membranes in a concentration-dependent manner. Changes in electrical current across a DPhPC bilayer were monitored following the addition of (a) WT-24H EVs, (b) WT-48H EVs, (c) ece1Δ/Δ-24H EVs, and (d) ece1Δ/Δ-48H EVs (1 × 10¹², 3 × 10¹², and 5 × 10¹² p/mL). Vesicle fusion and membrane permeabilization were monitored (see zoomed-in box). (e) The membrane permeabilization kinetics, defined as the latency until membrane permeabilization, are presented as a bar chart (see Table S1). Concentration-dependent membrane permeabilization was observed based on data from at least six independent experiments. Mean values of latency and their corresponding standard deviations are displayed. Statistical significance was determined using a one-way analysis of variance (P **** <0.0001, ns = non-significant).

Fig 3

EVs from C. albicans biofilms induce cytokine secretion. Oral epithelial cells were exposed to C. albicans biofilm EVs for 24 h, and secretion of cytokines was quantified. Observed concentrations (Obs. conc.; pg/mL) are shown. Treatments with synthetic candidalysin peptide (7 µM and 15 µM) were included as a positive control. WT-24H EVs, WT-48H EVs, and ece1Δ/Δ-48H were able to induce some cytokine release including (a) G-CSF, (b) GM-CSF, and (c) IL-1β. Data are shown as the mean ± standard deviation of n = 3–4 biological repeats. Statistical significance was calculated using an unpaired two-tailed t-test in comparison with the vehicle (P * <0.05, P ** <0.01, and P *** <0.001).

Fig 4

Candidalysin-containing EVs exhibit differences in detergent susceptibility. The radius of EVs obtained from wild-type C. albicans biofilms cultured for 24 h and 48 h was monitored in response to treatment with Triton X-100 using dynamic light scattering. (a) The bar chart represents the mean ± standard deviation of n = 3 independent experiments. Statistical analysis was performed using an unpaired two-tailed t-test (P * <0.05, ns = non-significant). (b) Hydrodynamic radius distribution by intensity for WT-24H EVs without (red) and with Triton treatment (purple). The reduction in radius after detergent treatment indicates membrane-ruptured EVs. (c) Similar analysis of WT-48H EVs, without (light blue) and with Triton treatment (blue) shows that the radius of WT-48H EVs remained unchanged, indicating the resistance to detergent-induced membrane rupture.

Fig 5

Comparative lipidomics of C. albicans biofilm EVs. The circular Voronoi treemaps reflect absolute amounts of distinct lipid classes detected in EVs isolated from 24 h and 48 h old biofilms, respectively. The coloration of individual clusters is based on their classification and shows polar lipids in gray-blue, sphingolipids in rainforest green, and neutral lipids in putty yellow. Data shown are the mean of n = 7 independent experiments. Polar lipids: PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphoethanolamine; PG, phosphatidylglycerol; PI, phosphoinositol; lysoPC, lyso phosphatidylcholine; lysoPE, lysophosphoethanolamine; PS, phosphatidylserine. Sphingolipids: IPC, inositol-P-ceramides; MIPC, mannose-inositol-P-ceramides; MI(IP)₂C, mannosyl-di-(inositolphosphoryl)-ceramide. Neutral lipids: DAG, diacylglycerols; TAG, triacylglycerols; ERG, ergosterol esters.