The Aspergillus fumigatus Sialidase (Kdnase) Contributes to Cell Wall Integrity and Virulence in Amphotericin B-Treated Mice

Abstract

Aspergillus fumigatus is a filamentous fungus that can cause a life-threatening invasive pulmonary aspergillosis (IPA) in immunocompromised individuals. We previously characterized an exo-sialidase from A. fumigatus that prefers the sialic acid substrate, 2-keto-3-deoxy-D-glycero-D-galacto-nononic acid (Kdn); hence it is a Kdnase. Sialidases are known virulence factors in other pathogens; therefore, the goal of our study was to evaluate the importance of Kdnase in A. fumigatus. A kdnase knockout strain (Δkdnase) was unable to grow on medium containing Kdn and displayed reduced growth and abnormal morphology. Δkdnase was more sensitive than wild type to hyperosmotic conditions and the antifungal agent, amphotericin B. In contrast, Δkdnase had increased resistance to nikkomycin, Congo Red and Calcofluor White indicating activation of compensatory cell wall chitin deposition. Increased cell wall thickness and chitin content in Δkdnase were confirmed by electron and immunofluorescence microscopy. In a neutropenic mouse model of invasive aspergillosis, the Δkdnase strain had attenuated virulence and a significantly lower lung fungal burden but only in animals that received liposomal amphotericin B after spore exposure. Macrophage numbers were almost twofold higher in lung sections from mice that received the Δkdnase strain, possibly related to higher survival of macrophages that internalized the Δkdnase conidia. Thus, A. fumigatus Kdnase is important for fungal cell wall integrity and virulence, and because Kdnase is not present in the host, it may represent a potential target for the development of novel antifungal agents.

Keywords: Kdn; cell wall integrity; chitin; invasive aspergillosis; sialidase.

FIGURE 1

Comparative growth of wild type and Δkdnase strains on minimal media supplemented with varying carbon sources. Wild type and Δkdnase strains were plated on solid Kafer’s minimal medium (fungal media), supplemented with various carbon sources. Sialic acid is N-acetylneuraminic acid (Neu5Ac). Equivalent numbers of conidia were inoculated into the center of each plate and plates were incubated for 72 h at 37°C. Growth differences between the parental strain and the knockout are noted with boxes. Hygromycin resistance is the Δkdnase selectable marker and the mutant but not the wild type grew in the presence of hygromycin, as expected. Three other Δkdnase mutants displayed the same phenotype (data not shown).

FIGURE 2

Growth of the Δkdnase strain is inhibited compared to the wild type strain when grown on solid medium supplemented with 1 M sorbitol. Left: YAG plates (with or without 1 M sorbitol) were inoculated with 10⁵ conidia, in a volume of 5 μL in the center of each plate and incubated at 37°C for 48 h. (A) WT on YAG+sorbitol; (B) Δkdnase on YAG+sorbitol; (C) WT on YAG; (D) Δkdnase on YAG. Right: Slide cultures were prepared for each strain on YAG ± sorbitol and cultures were grown for 20 h at 37°C. All DIC images were captured at 1000× magnification. The same letters correspond to the samples on the plate cultures shown on the Left.

FIGURE 3

Transmission electron microscope images show that sorbitol increases cell wall thickness of the Δkdnase strain. (A) Each row represents one strain/growth condition at lower (Left) and higher (Right) magnification. (A,B) - Δkdnase knockout grown in YAG; (C,D) - Δkdnase knockout in YAG + sorbitol; (E,F) - WT in YAG; (G,H) - WT in YAG + sorbitol. Note the difference in cell wall thickness in (D) and (H). Scale bar is below each image. (B) From the TEM images of hyphae, cell wall thickness was quantified with ImageJ using the measure function. Cell wall thickness was measured 5 times in each image at coordinates determined by random number generation. The number of separate images analyzed for each condition is as follows: WT – 17, Δkdnase (KO) – 17, WT + 1M sorbitol – 12, Δkdnase + 1M sorbitol – 24. Whiskers represent the 5–95% confidence interval, the dots indicate outliers, and the + indicates the median. The KO plus sorbitol group was significantly difference from all other groups at p < 0.001 (ANOVA followed by multiple comparison test with the Bonferroni correction).

FIGURE 4

The Δkdnase knockout strain is resistant to the growth inhibitory effects of Congo Red and Calcofluor White. (A) Wild type or Δkdnase conidia were plated in a series of 10-fold dilutions in a total volume of 2 μl on YAG containing 500 μg/mL of Congo Red. (B) Conidia (∼5000) from each strain were plated on solid YAG media containing 0.25% Calcofluor White and colony diameter was measured after growth at 37°C. - wild type; - Δkdnase knockout strain; and - Δkdnase^R strain. Values represent the mean ± SD of three experiments.

FIGURE 5

Hyperosmotic stress increases cell wall α(1,3)-glucan and chitin levels in the kdnase strain. Cell wall polysaccharides (A) β(1,3)-glucan; (B) α(1,3)-glucan; and (C) chitin were determined in hyphal tips grown in slide culture on YAG medium with or without sorbitol (1 M). Live cells were labeled with fluorescent antibodies (A,B) or the chitin probe (C), imaged with confocal microscopy and the fluorescence yield was quantified with Volocity software. Representative images of chitin staining are shown in Supplementary Figure S6. WT, wild type; KO, Δkdnase strain and Res, Δkdnase^R strain.

FIGURE 6

Susceptibility of WT and kdnase strains to various antifungal agents. Wild type and Δkdnase conidia (2 × 10⁵ conidia/mL) were inoculated into 96-well microdilution plates containing RPMI supplemented with resazurin with increasing concentrations of one of the following antifungal agents: (A) nikkomycin, (B) voriconazole, (C) caspofungin and (D) amphotericin B. Plates were incubated at 37°C for up to 46 h. Growth was quantified by fluorescence (excitation = 560 nm; emission = 590 nm). Control wells contained no antibiotic or no conidia. Error bars indicate the 95% confidence interval of 6 replicates per condition. Significant differences in growth between the wild type and Δkdnase strains are indicated by ^∗∗∗∗(p < 0.0001), ^∗∗∗(p < 0.001) or ^∗(p < 0.05). Each antibiotic was tested at least 2 times with similar results. Black bars - wild type; white bars - Δkdnase.

FIGURE 7

Kaplan-Meier survival curves of immunosuppressed mice exposed to wild type (WT), Δkdnase (KO), or Δkdnase^R (R) conidia with or without AmBisome (AB) treatment. All mice were immunocompromised with cyclophosphamide and cortisone. On day 0, mice in all treatment groups (7 per group) received 10⁶ conidia intranasally of the appropriate strain, and indicated groups were also treated with i.v. AB (5 mg/kg), 4 h after exposure to conidia. Control mice that received saline intranasally (3) or saline intranasally followed by i.v. AB (4) had 100% survival (not shown). Comparison of survival curves with the log rank test (Mantel-Cox) was significant at p < 0.0007. ^∗Indicates p < 0.04 for KO+AB versus WT+AB.

FIGURE 8

Lung fungal burden. (A) DNA was extracted from homogenized lung samples (n = 7 except KO+AB was n = 6) and fungal DNA was quantified by TaqMan qPCR. Fungal DNA content was determined by comparison to a standard curve and normalized to lung wet weight. Statistical analysis was carried out using a one-way ANOVA followed by a Games Howell test for differences between groups. Groups with the same letters are not significantly different at p < 0.05. Control animals received saline only and represent the pooled values from animals with and without AB treatment. (B) Percent coverage of GMS stained lung sections by fungal elements (conidia or hyphae). Mean percent fungal element was quantified from 3 complete lung sections from 7 WT AB mice and 6 KO AB mice taken at endpoint using the color threshold tool in ImageJ; error bars represent the S.E.M. ^∗∗∗ KO AB was significantly different from WT AB at p < 0.0009 using ANOVA followed by Tukey’s test on log-transformed data. All CON AB values were 0 (not shown). Abbreviations are the same as in the legend to Figure 7. Representative images of GMS-stained lung sections are shown in Figure 9.

FIGURE 9

Representative images of lung histology. Fixed lung sections were stained with hematoxylin and eosin (H&E) to visualize cell infiltrate or Grocott’s methenamine silver (GMS) to stain fungal elements (fungi appear black against the green host tissue). All images were obtained at 400× magnifications. Abbreviations are the same as described in the legend to Figure 7. Images from lung sections taken from two different mice in the KO+AB group are shown.

FIGURE 10

Leukocyte counts in lung sections. Whole lung sections (3 per animal) were fixed and stained to detect neutrophils (MPO⁺) cells (A), or macrophages (Mac3+) (B). Data represent the mean of n = 7 (WT AB), n = 6 (KO AB) and n = 3 (CON AB) ± SEM. (A) MPO⁺ cells/mm² were not significantly different in lungs of WT AB and KO AB animals; ^∗∗CON AB values were significantly lower than WT AB and KO AB at p < 0.01 (ANOVA followed by Tukey’s multiple comparison test). (B) The number of Mac3+ cells was significantly higher in KO AB samples compared to CON AB ^∗(p < 0.05) whereas Mac3+ values for WT AB were not significantly different from KO AB or CON AB (one-way ANOVA followed by Tukey’s multiple comparisons test). Abbreviations are the same as in the legend of Figure 7. Sample images of Mac3+ staining are shown in Supplementary Figure S9.

FIGURE 11

Uptake of conidia by cultured murine macrophages. (A) Microscopic analysis of the uptake of the fungal conidia by J774 macrophages after 1 h of co-incubation at 37°C. Green fluorescence from FITC shows all conidia (470 nm) whereas only external conidia were stained by the antibody (red – 590 nm). (B) Proportion of bound conidia internalized by cultured macrophages was equivalent in all three strains. Data show the mean ± SEM from a minimum of 28 images per strain. The experiment was repeated twice with the same result. Abbreviations are the same as those described in the legend Figure 7.