Identification of a New Class of Antifungals Targeting the Synthesis of Fungal Sphingolipids

Abstract

Recent estimates suggest that >300 million people are afflicted by serious fungal infections worldwide. Current antifungal drugs are static and toxic and/or have a narrow spectrum of activity. Thus, there is an urgent need for the development of new antifungal drugs. The fungal sphingolipid glucosylceramide (GlcCer) is critical in promoting virulence of a variety of human-pathogenic fungi. In this study, we screened a synthetic drug library for compounds that target the synthesis of fungal, but not mammalian, GlcCer and found two compounds [N'-(3-bromo-4-hydroxybenzylidene)-2-methylbenzohydrazide (BHBM) and its derivative, 3-bromo-N'-(3-bromo-4-hydroxybenzylidene) benzohydrazide (D0)] that were highly effective in vitro and in vivo against several pathogenic fungi. BHBM and D0 were well tolerated in animals and are highly synergistic or additive to current antifungals. BHBM and D0 significantly affected fungal cell morphology and resulted in the accumulation of intracellular vesicles. Deep-sequencing analysis of drug-resistant mutants revealed that four protein products, encoded by genes APL5, COS111, MKK1, and STE2, which are involved in vesicular transport and cell cycle progression, are targeted by BHBM.

Importance: Fungal infections are a significant cause of morbidity and mortality worldwide. Current antifungal drugs suffer from various drawbacks, including toxicity, drug resistance, and narrow spectrum of activity. In this study, we have demonstrated that pharmaceutical inhibition of fungal glucosylceramide presents a new opportunity to treat cryptococcosis and various other fungal infections. In addition to being effective against pathogenic fungi, the compounds discovered in this study were well tolerated by animals and additive to current antifungals. These findings suggest that these drugs might pave the way for the development of a new class of antifungals.

FIG 1

BHBM and D0 inhibit the synthesis of fungal but not mammalian glucosylceramide. (A) Thin-layer chromatography analysis of the synthesis of glucosylceramide (GlcCer) upon in vivo labeling of C. neoformans (Cn) or J774 cells with [³H]palmitate and treated with BHBM or D0 at the indicated concentrations. (B) Structure of N′-(3-bromo-4-hydroxybenzylidene)-2-methylbenzohydrazide (BHBM) and 3-bromo-N′-(3-bromo-4-hydroxybenzylidene) benzohydrazide (D0). The MIC and minimum fungicidal concentration (MFC) are shown. Representative data of three independent experiments are shown.

FIG 2

Killing activity of BHBM and D0. Killing activity was determined using an in vitro killing assay in which the compounds were added to C. neoformans cells, which were then incubated at 37°C, 5% CO₂, and pH 7.4. The number of CFU were counted during the 96-h incubation. (A and B) Both BHBM (A) and D0 (B) showed fungicidal activity. BHBM showed concentration-dependent killing, whereas D0 showed time-dependent killing. D0 is more effective in killing C. neoformans (100% dead cells within 24 h), and the killing activity does not occur earlier than 24 h with higher doses. BHBM kills more slowly, requiring at least 72 h of incubation to kill about 50% of the cells. Values that are significantly different are indicated as follows: *, P < 0.05 comparing treated cells versus untreated cells (no drug); #, P < 0.001 comparing treated cells versus untreated cells (no drug). (C) Intracellular activity of BHBM was assessed by incubating macrophages with internalized C. neoformans cells with different concentrations of BHBM in the absence of opsonins. Values that are significantly different are indicated as follows: *, P < 0.05 comparing extracellular or intracellular treated (0.25, 1, or 4 µg/ml) versus extracellular or intracellular untreated (0 µg/ml), respectively. Statistical analysis was performed using analysis of variance (ANOVA). Data were compiled from three independent experiments.

FIG 3

Effects of BHBM and D0 on survival of mice upon infection with C. neoformans, P. murina, or C. albicans. (A) Cryptococcosis. Mice were infected intranasally with C. neoformans (Cn) and received 1.2 mg/kg/day of either BHBM or D0 intraperitoneally (10 mice per group). Values that are significantly different are indicated as follows: *, P < 0.001 comparing BHBM- or D0-treated mice versus untreated mice (no drug). (B) Cryptococcosis. Survival of mice infected intravenously with C. neoformans (8 mice per group) and treated intraperitoneally either with BHBM, D0, fluconazole (FLC), or amphotericin B (AMB). Values that are significantly different are indicated as follows: ^, P < 0.05 comparing BHBM- or D0-treated mice versus untreated mi8ce (solvent); $, P < 0.005 comparing FLU-treated mice versus untreated mice and AMB-treated mice versus untreated mice. (C) Pneumocystosis. Survival of corticosteroid-immunosuppressed (Cort) or CD4-depleted (CD4d) mice infected intranasally with P. murina after 13 days of intraperitoneal treatment with BHBM (3.2 mg/kg/day), D0 (1.25 mg/kg/day), or trimethoprim-sulfamethoxazole (T/S) (12 mice per group). Values that are significantly different are indicated as follows: &, P < 0.05 comparing BHBM- or D0-treated mice versus respective untreated controls. (D) Candidiasis. Survival of mice infected with C. albicans SC 5314 (Ca) intravenously received no drug or 1.2 mg/kg/day of either BHBM or D0 intraperitoneally (8 mice per group). Values that are significantly different are indicated as follows: §, P < 0.01 comparing BHBM- or D0-treated mice versus untreated mice. Statistical analysis for survival studies was performed using Kruskal-Wallis test and by Student-Newman-Keuls t test for multiple comparisons using INSTAT.

FIG 4

Measurements of sphingolipids upon treatment with BHBM. (a) Thin-layer chromatography analysis of sphingolipids (see diagram) isolated from untreated or BHBM-treated C. neoformans cells after in vivo labeling with [³H]palmitate (³H-Palm). (b to h) Lipid analysis by LC-MS. (b) C18 hydroxy (Δ8) 9 methyl-glucosylceramide (GlcCer); (c) C18 dihydroceramide (C18 dhCer); (d) sphingosine and dihydrosphingosine (SPHs); (e) sphingosine-1-phosphate and dihydrosphingosine-1-phosphate (SPH-1-P); (f) C18 hydroxyceramide (C18 OH-Cer); (g) C18 hydroxy Δ8-ceramide (C18 OH-Δ8-Cer); and (h) C18 hydroxy Δ8, 9 methyl-ceramide (C18 OH-Δ8, 9Me-Cer). Values that are significantly different are indicated as follows: *, P < 0.05 comparing treated to untreated (no drug). Statistical analysis was performed using analysis of variance (ANOVA) test. Statistical significance is accepted at a P value of <0.05. Data were compiled from three independent experiments.

FIG 5

Effect of BHBM on Golgi morphology in C. neoformans and yeast-to-hyphal differentiation in C. albicans. (A) Control or BHBM-treated cells were stained with NBD C6-ceramide (Golgi and Golgi-derived compartments, green fluorescence). DIC, differential interference contrast; NBD-Cer, NBD 6-ceramide. Bars, 5 µm. (B) FACS analysis shows NBD C6-ceramide accumulation in C. neoformans. The gray histogram corresponds to C. neoformans yeast in the absence of NBD C6-ceramide. CT, control (no BHBM treatment). (C) Percentage of yeast and hyphal forms under control conditions and after BHBM treatment. Statistical analysis was performed using analysis of variance (ANOVA) test. Statistical significance is accepted at a P value of <0.05. The images in panels A and B are representative of three separate experiments. The data in panel C were compiled from three independent experiments.

FIG 6

Transmission electron microscopy examination of high-pressure frozen and freeze substituted C. neoformans yeast cells. (A, D, and E) Cells treated with 4 µg/ml of BHBM for 6 h, showing accumulation of intracellular vesicles (arrows) and regions containing a “reticulum” presumably formed by the fusion of intracellular vesicles (arrowheads). M, mitochondrion. (B) Control cells showing a regular aspect. (C) Morphometric analysis results showing the volumetric density of the intracellular vesicles in control and treated cells. (E) Accumulation of cytoplasm-rich vesicles in the periplasmic space. Images and data are representative of the results of three separate experiments.

FIG 7

Effects of BHBM, fluconazole, and methyl methane sulfonate (MMS) on wild-type BY4741 and Δapl5, Δcos111, Δmkk1, and Δste2 deletion strains. Relative growth inhibition was calculated by the average rate after normalizing the OD₆₀₀ values in drug-treated wells against the DMSO control wells on each assay plate. The mutant strains show increased resistance to BHBM but not to fluconazole or MMS. Results are from two independent growth assays.