The Anticancer Drug Bleomycin Shows Potent Antifungal Activity by Altering Phospholipid Biosynthesis

Abstract

Invasive fungal infections are difficult to treat with limited drug options, mainly because fungi are eukaryotes and share many cellular mechanisms with the human host. Most current antifungal drugs are either fungistatic or highly toxic. Therefore, there is a critical need to identify important fungal specific drug targets for novel antifungal development. Numerous studies have shown the fungal phosphatidylserine (PS) biosynthetic pathway to be a potential target. It is synthesized from CDP-diacylglycerol and serine, and the fungal PS synthesis route is different from that in mammalian cells, in which preexisting phospholipids are utilized to produce PS in a base-exchange reaction. In this study, we utilized a Saccharomyces cerevisiae heterologous expression system to screen for inhibitors of Cryptococcus PS synthase Cho1, a fungi-specific enzyme essential for cell viability. We identified an anticancer compound, bleomycin, as a positive candidate that showed a phospholipid-dependent antifungal effect. Its inhibition on fungal growth can be restored by ethanolamine supplementation. Further exploration of the mechanism of action showed that bleomycin treatment damaged the mitochondrial membrane in yeast cells, leading to increased generation of reactive oxygen species (ROS), whereas supplementation with ethanolamine helped to rescue bleomycin-induced damage. Our results indicate that bleomycin does not specifically inhibit the PS synthase enzyme; however, it may affect phospholipid biosynthesis through disruption of mitochondrial function, namely, the synthesis of phosphatidylethanolamine (PE) and phosphatidylcholine (PC), which helps cells maintain membrane composition and functionality. IMPORTANCE Invasive fungal pathogens cause significant morbidity and mortality, with over 1.5 million deaths annually. Because fungi are eukaryotes that share much of their cellular machinery with the host, our armamentarium of antifungal drugs is highly limited, with only three classes of antifungal drugs available. Drug toxicity and emerging resistance have limited their use. Hence, targeting fungi-specific enzymes that are important for fungal survival, growth, or virulence poses a strategy for novel antifungal development. In this study, we developed a heterologous expression system to screen for chemical compounds with activity against Cryptococcus phosphatidylserine synthase, Cho1, a fungi-specific enzyme that is essential for viability in C. neoformans. We confirmed the feasibility of this screen method and identified a previously unexplored role of the anticancer compound bleomycin in disrupting mitochondrial function and inhibiting phospholipid synthesis.

Keywords: Cryptococcus neoformans; PS synthase assay; ROS; antifungal drug; bleomycin; inhibitor; mitochondria; phosphatidylserine synthase.

FIG 1

CnCHO1 expression in Sccho1Δ mutant rescued its growth arrest phenotype. (A) Wild-type (WT) S. cerevisiae strain (Sc WT) grown in a 96-well plate at 30°C for 48 h showed no difference in growth on SD medium, both in presence and absence of ethanolamine. Sccho1Δ mutant failed to grow on SD medium lacking ethanolamine, while it grew on ethanolamine-supplemented medium. Sccho1Δ mutant expressing CnCHO1 (Sccho1Δ+CnCHO1) grew well even in the absence of ethanolamine complementing the growth arrest. (B) Growth was evaluated via absorbance measurements at an optical density of 600 nm (OD₆₀₀) using a plate reader. (C) PS synthesis in fungi and mammals. In yeast de novo pathway of phosphatidylserine (PS) synthesis, endoplasmic reticulum (ER)-localized PS synthase, and Cho1 converting CDP diacylglycerol (CDP-DAG) to PS. PS is subsequently converted to phosphatidylethanolamine (PE) by inner mitochondrial membrane-localized PS decarboxylase 1 (Psd1). PE can be exported to the ER for further conversion to phosphatidylcholine (PC). Ethanolamine (Etn) and choline (Cho) from cytosol can be utilized to form PE and PC, respectively, via the Kennedy pathway. In mammals, PS is produced through a base exchange reaction in which choline in PC and ethanolamine in PE are replaced with serine by the activity of PS synthases Pss1 and Pss2, respectively.

FIG 2

Secondary screening of compound collections identified compounds with some specificity against PS synthase. Primary screening at 5 μM starting concentration identified 46 chemical compounds in the FDA-approved drug collection (A), 6 compounds in the Pathogen Box collection (B), and 7 compounds in the NIH Clinical Collection (NCC) set (C). All chemical compounds from the primary screen were further examined for rescue of growth arrest in the presence of 1 mM ethanolamine (SD+Etn) compared to that in SD medium lacking ethanolamine (SD). Growth of Sccho1Δ+CnCHO1 strain (1,000 cells) in 96-well plates in SD (left panel) and SD+Etn medium (right panel) at final concentrations of chemical compounds of 5, 2.5, 1.25, 0.625, 0.312, and 0 μM DMSO (dimethyl sulfoxide) was evaluated via absorbance measurements at OD₆₀₀ using a plate reader after 48 h of incubation at 30°C. Results are represented as a heat map, with green indicating growth and purple indicating no growth. While most compounds showed similar ranges of growth inhibition, bleomycin and bleomycin sulfate (indicated by black arrowheads at the sides of the heat maps) showed some degree of growth rescue.

FIG 3

Determination of MICs and PS synthase activity. (A) MIC assay was performed on 96-well plate for indicated strains to determine the rescue effect of bleomycin (blm) in the presence of ethanolamine. (B) The lowest concentration that inhibited visible growth after 48 h of incubation is interpreted as the MIC. (C) Structure of bleomycin sulfate. (D) PS synthase activity in S. cerevisiae strains. Compared to the untreated condition, Sc WT and Sccho1Δ+CnCHO1 with addition of 5 mM bleomycin showed no difference in PS synthase activity, indicating that bleomycin is not a specific inhibitor of Cho1. Data are represented as means ± standard deviation and statistical analysis was performed by unpaired two-tailed Student’s t test.

FIG 4

Analysis of mitochondrial function in bleomycin-treated yeast cells. The Sccho1+CnCHO1 strain and C. neoformans WT (H99) cultured in SD and SD+Etn medium were treated for 10 h with 2.5 μM and 5 μM bleomycin, respectively. Cells stained with MitoTracker Deep Red FM and Hoechst 33342 were viewed under a florescence microscope. Under the bleomycin-treated condition, both Sccho1+CnCHO1 (A) and H99 (B) showed accumulation of florescence dye in ethanolamine supplemented-medium, indicating normal mitochondrial membrane function. Scale bar = 10 μm. Flow cytometric analysis of MitoTracker-stained cells revealed higher mitochondrial membrane potential in yeast cells grown in ethanolamine-lacking medium. (C and E) Representative graphs of MitoTracker with flow cytometry for bleomycin-treated Sccho1Δ+CnCHO1 (C) and H99 cells (E), respectively. (D and F) Overlaid fluorescence histogram of MitoTracker for cell populations from bleomycin-treated Sccho1Δ+CnCHO1 (D) and H99 (F) cells. Values represent the average of three independent experiments, and error bars indicate standard deviation. Student’s t test was performed to determine statistical significance. *, P < 0.05; **, P < 0.01; ns, not significant.

FIG 5

Effect of bleomycin treatment on reactive oxygen species (ROS) production in yeast cells. Overnight-incubated yeast cells were transferred to SD and SD+Etn media for 4 h before bleomycin treatment for 10 h. Intracellular ROS accumulation was assayed using 2′,7′-dichlorodihydrofluorescin diacetate (H₂DCFDA) staining and visualized under a fluorescence microscope. Scale bar = 10 μm. Supplementation of ethanolamine decreased ROS accumulation in both strains, Sccho1+CnCHO1 (A) and H99 (B). (C and E) Representative graph of dichlorofluorescein (DCF) with flow cytometry for bleomycin-treated Sccho1Δ+CnCHO1 (C) and H99 (E) cells, respectively. (D and F) Overlaid fluorescence histogram of DCF for cell populations from bleomycin-treated Sccho1Δ+CnCHO1 (D) and H99 (F) cells. (G and H) Intracellular ROS measurement using dihydroethidium (DHE). DHE fluorescence in Sccho1Δ+CnCHO1 (G) and H99 (H) cells was analyzed using a microplate reader. Values represent the average of three independent experiments, error bars indicate standard deviation. Student’s t test was performed to determine statistical significance. *, P < 0.05; ns, not significant.