The antimicrobial activity of ETD151 defensin is dictated by the presence of glycosphingolipids in the targeted organisms

Abstract

Fungal infections represent a significant global health concern, with a growing prevalence of antifungal drug resistance. Targeting glucosylceramides (GlcCer), which are functionally important glycosphingolipids (GSL) present in fungal membranes, represents a promising strategy for the development of antifungal drugs. GlcCer are associated with the antifungal activity of certain plant and insect defensins. The 44-residue ETD151 peptide, optimized from butterfly defensins, is active against several fungal pathogens. ETD151 has been shown to induce a multifaceted mechanism of action (MOA) in Botrytis cinerea, a multiresistant phytopathogenic fungus. However, the target has yet to be identified. Our findings demonstrate that the presence of GlcCer in membranes determines the susceptibility of Pichia pastoris and Candida albicans toward ETD151. To ascertain whether this is due to direct molecular recognition, we demonstrate that ETD151 selectively recognizes liposomes containing GlcCer from B. cinerea, which reveals a methylated-sphingoid base structure. The dissociation constant was estimated by microscale thermophoresis to be in the µM range. Finally, fluorescence microscopy revealed that ETD151 localizes preferentially at the surface of B. cinerea. Furthermore, the majority of prokaryotic cells do not contain GSL, which explains their resistance to ETD151. We investigated the susceptibility of Novosphingobium capsulatum, one of the rare GSL-containing bacteria, to ETD151. ETD151 demonstrated transient morphological changes and inhibitory growth activity (IC₅₀ ~75 µM) with an affinity for the cell surface, emphasizing the critical importance of GSL as target. Understanding the MOA of ETD151 could pave the way for new perspectives in human health and crop protection.

Keywords: Botrytis cinerea; Novosphingobium capsulatum; antifungal defensin; glucosylceramide; molecular affinity.

Fig. 1.

Antifungal activity of ETD151 defensin on wt yeasts C. albicans, P. pastoris, and their mutants lacking GlcCer (Δgcs). Dose–response curves illustrating the growth of yeasts in the presence of ETD151 and the radish defensin, RsAFP2, used as a positive control. Peptide activity is expressed as % growth relative to their concentrations in μg/mL (log scale from 0.1 to 200). The study was performed in triplicate, and results are represented as mean ± SD.

Fig. 2.

Structural analysis of GlcCer obtained from B. cinerea B05.10 mycelia. (A) Isolation of glycosphingolipid-enriched fraction of crude total lipid extracted from B. cinerea using HPTLC. HPTLC migration was performed with dichloromethane/methanol/ammonia 2 M (65:25:4, vol/vol/vol), and lipid band was visualized under UV light after primulin staining. GlcCer from Soybean (GlcCer_Soybean) was used as a standard. (B) ESI–MS spectra of purified glycosphingolipid-enriched fraction. (C) ESI–MS/MS of molecular precursor ions at m/z 726.4 and 728.5 observed in ESI–MS. (D) GC–MS analysis of the glycosphingolipid-enriched fraction after extraction, methanolysis, and trimethylsilyl (TMS) derivatization. Total ion chromatograms (TIC) for glucose and galactose as standards and glycosphingolipid-enriched fraction from B. cinerea. (E) Proposed structure for the major GlcCer species found in B. cinerea. By analogy to plant GlcCer, the double bound of the fatty acid was proposed at C3-C4 position.

Fig. 3.

ETD151 preferentially binds to liposomes containing fGlcCer. (A) Binding affinities by MST assay of Atto647-ETD151 to (Right) GlcCer-containing liposomes, GlcCer extracted from B. cinerea (GlcCer_{B. cinerea}) were incorporated into PC liposomes with a molar ratio at 10/90 (GlcCer/PC) and (Left) GlcCer-free liposomes, made with only PC, were used as a control. Kd is the dissociation constant given as mean of three independent experiments ± SD. All experiments were done with a constant concentration of Atto647-ETD151 (50 nM) at 22 °C. Autodetect excitation power and 40% MST power were used. MST buffer was phosphate buffer (10 mM, pH 5.8). (B) Liposome pulldown assay comparing the binding of ETD151 to PC only and PC: fGlcCer (90:10 molar ratio) liposomes. Bound (B) and unbound (U) fractions were subjected to SDS-PAGE analysis and colloidal Coomassie staining. ETD151 alone was used as loading control (LC). The intensities of the bands were calculated using ImageJ software. Averages of the ratio of intensities (Bound/Unbound) were then calculated for n = 2 experiments (shown in the gel). Data represent mean ± SEM, n = 2.

Fig. 4.

BODIPY-ETD151 localization in B. cinerea cells. (A) Internalization of the BODIPY-ETD151 peptide (0.78 µM) on B. cinerea. Confocal microscopy images of B. cinerea (10⁴ sp/mL in a 25% PDB medium) by 63x optical magnification. Hyphae examined in transmitted light and fluorescence at T0 and 24 h after addition of BODIPY-ETD151 (Scale bar, 10 μm). The white frames indicate the staining of the membranes around intracellular organelles by BODIPY-ETD151. (B) Subcellular localization of BODIPY-ETD151 (0.19 µM) on hyphae of the B. cinerea. Confocal observations made on B. cinerea (10⁴ sp/mL in a 25% PDB medium) by 100x optical magnification. Hyphae examined by colabeling 30 min (Scale bar, 5 μm) and 24 h (Scale bar, 10 μm) in the presence of 0.19 μM BODIPY-ETD151 (green), Calcofluor (blue), and FM4-64 (red), respectively. (C) Signal intensities taken from regions of interest ROI.01, ROI.02, and ROI.03 from the merged image after 30-min treatment of mycelia B. cinerea (Fig. 4B).

Fig. 5.

Effect of ETD151 on glycosphingolipid-containing bacterium N. capsulatum. (A) Antibacterial activity of ETD151 against N. capsulatum was determined using the resazurin cell viability assay. Different concentrations of ETD151 (25 to 200 µM) were incubated with N. capsulatum for 24 h. Fluorescence intensity was measured after 2 h of adding resazurin (10%, v/v). The metabolically active cells can reduce nonfluorescent resazurin to fluorescent form. Data are represented as means ± SEM for triplicate measurements. (B) Representative microscopy images of N. capsulatum treated after 24 h with two concentrations of ETD151 (12.5 and 100 µM). Untreated cells, showing normal rod-shaped cells. At 12.5 µM, ETD151 induces bacterial cell shape changes, showing spherical cells of N. capsulatum. At 100 µM, ETD151 appears to have any effect on cellular morphology. (Scale bar, 2 µm.) (C) Cytoplasmic membrane permeabilization of N. capsulatum treated with 100 µM ETD151 determined via flow cytometry with 30 µM PI (PI membrane impermeable) stain. The proportion of permeabilized cells was detected by fluorescence due to binding of PI fluorescent probe with DNA after 30 and 60 min of incubation with ETD151. Untreated cells and cells treated with Triton-X 100 (1%, v/v) were used as negative and positive controls, respectively. Data are means ± SEM for triplicate measurements. (D) Subcellular localization of Atto647-ETD151 in N. capsulatum. Left: Confocal microscopy images of N. capsulatum cells simultaneously exposed to 2 µM Atto647-ETD151 (blue) and 10 nM Sytox orange (red) for 30 min. (Scale bar, 1 µm.) Right: Signal intensity (Arbitrary Units) taken from region of interest (ROI) of each cell.