Potent Antifungal Activity of Penta- O-galloyl-β-d-Glucose against Drug-Resistant Candida albicans, Candida auris, and Other Non- albicans Candida Species

Abstract

Among fungal pathogens, infections by drug-resistant Candida species continue to pose a major challenge to healthcare. This study aimed to evaluate the activity of the bioactive natural product, penta-O-galloyl-β-d-glucose (PGG) against multidrug-resistant (MDR) Candida albicans, MDR Candida auris, and other MDR non-albicans Candida species. Here, we show that PGG has a minimum inhibitory concentration (MIC) of 0.25-8 μg mL^-1 (0.265-8.5 μM) against three clinical strains of C. auris and a MIC of 0.25-4 μg mL^-1 (0.265-4.25 μM) against a panel of other MDR Candida species. Our cytotoxicity studies found that PGG was well tolerated by human kidney, liver, and epithelial cells with an IC₅₀ > 256 μg mL^-1 (>272 μM). We also show that PGG is a high-capacity iron chelator and that deletion of key iron homeostasis genes in C. albicans rendered strains hypersensitive to PGG. In conclusion, PGG displayed potent anti-Candida activity with minimal cytotoxicity for human cells. We also found that the antifungal activity of PGG is mediated through an iron-chelating mechanism, suggesting that the compound could prove useful as a topical treatment for superficial Candida infections.

Keywords: candidiasis; iron chelation; mechanism of action; natural product.

Figure 1

Chemical structure of penta-O-galloyl-β-d-glucose, PGG.

Figure 2

Penta-O-galloyl-β-d-glucose displays broad spectrum anti-Candida activity. Growth inhibition of PGG against (A) C. albicans, (B) C. glabrata, (C) C. parapsilosis, and (D) C. auris. Solid lines represent PGG. Dashed lines represent the positive control fluconazole. The data are plotted as the mean % growth inhibition ± SD as compared to the DMSO vehicle from two independent experiments. The horizontal dotted line represents 90% growth inhibition. See Supplemental Figures S8–S11 for expanded graphs including additional positive controls. See Table 2 for strain details.

Figure 3

PGG sensitivity in C. albicans is not modulated by efflux pumps. The cdr1Δ/Δ, cdr2Δ/Δ, mdr1Δ/Δ, flu1Δ/Δ, and wild-type strains were grown overnight in YPD and subject to dose–response assays in RPMI-1640 medium with twofold dilution gradients of PGG. Fluconazole was included as a positive control. Growth (OD_600nm) was measured after 72 h of incubation at 30 °C, averaged between technical duplicates, and normalized to the drug-free wild-type control. Results are presented in a heat-map format with a scale bar provided at the bottom of the figure. All assays were performed in biological duplicates.

Figure 4

Growth inhibition of PGG against multiple Candida species is abrogated by iron supplementation, and PGG binds multiple iron ions. (A) C. albicans 90028, (B) C. glabrata CDC325, and (C) C. auris CDC381 were treated with PGG or Amphotericin B at varying concentrations (4, 8, and 16 μg/mL) and media was supplemented with a 10 mM solution of ferrous sulfate (Fe²⁺) or ferric sulfate (Fe³⁺) and % growth inhibition measured after 48 h. The data are plotted as the mean % growth inhibition ± SD as compared to the DMSO vehicle from two independent experiments. The horizontal dotted line represents 90% growth inhibition compared to the vehicle. Statistical analysis was done using two-way ANOVA with multiple comparisons. (D) PGG or EGCG was allowed to react with varying concentrations of iron (III) in solution at pH 2.0. The bipy solution was added to the mixtures and the formation of the [Fe(bipy)₃]²⁺ complex was measured at 520 nm. [PGG]_total, [EGCG]_total = 10 μM. The data are plotted as the mean absorbance ± SD from two independent experiments.

Figure 5

Iron homeostasis genes are required for PGG tolerance in C. albicans. The ire1Δ/ire1Δ, hap43Δ/hap43Δ, and rim101Δ/rim101Δ strains and their respective wild-type controls were grown overnight in YPD and subject to dose–response assays in RPMI-1640 medium with twofold dilution gradients of PGG. The metal chelator BPS was included as a positive control. After 48 h incubation at 30 °C, the relative viable cell number was assessed. Data were averaged between technical duplicates and normalized to wild-type drug-free control wells. Results are presented in a heat-map format with a color scale bar presented at the bottom of the figure. All assays were performed in biological duplicates.