Developmental cell fate and virulence are linked to trehalose homeostasis in Cryptococcus neoformans

Abstract

Among pathogenic environmental fungi, spores are thought to be infectious particles that germinate in the host to cause disease. The meningoencephalitis-causing yeast Cryptococcus neoformans is found ubiquitously in the environment and sporulates in response to nutrient limitation. While the yeast form has been studied extensively, relatively little is known about spore biogenesis, and spore germination has never been evaluated at the molecular level. Using genome transcript analysis of spores and molecular genetic approaches, we discovered that trehalose homeostasis plays a key role in regulating sporulation of C. neoformans, is required for full spore viability, and influences virulence. Specifically, we found that genes involved in trehalose metabolism, including a previously uncharacterized secreted trehalase (NTH2), are highly overrepresented in dormant spores. Deletion of the two predicted trehalases in the C. neoformans genome, NTH1 and NTH2, resulted in severe defects in spore production, a decrease in spore germination, and an increase in the production of alternative developmental structures. This shift in cell types suggests that trehalose levels modulate cell fate decisions during sexual development. We also discovered that deletion of the NTH2 trehalase results in hypervirulence in a murine model of infection. Taken together, these data show that the metabolic adaptations that allow this fungus to proliferate ubiquitously in the environment play unexpected roles in virulence in the mammalian host and highlight the complex interplay among the processes of metabolism, development, and pathogenesis.

FIG 1

Trehalose metabolism gene transcripts are overrepresented in spores. (A) Heat map of genes in the KEGG database starch and sucrose metabolism pathway that were spore enriched. Relative fold change is represented from −4 (blue) to +4 (yellow). Each column represents a time point of germination in hours. Dormant spore transcripts are represented in 0 h. Yeast transcripts are represented in 10 h. Each row represents a gene. (B) A schematic of genes predicted to act in the trehalose and glycogen metabolic pathway, with those found to be spore enriched in red. Only one gene, NTH1, was found not to be spore enriched. (C) A schematic of predicted C. neoformans trehalases, Nth1 and Nth2, and their functional domain content, including the family 37 glycosyl hydrolase α,α-trehalase domain (GH37; blue), calcium binding regulatory domain (Ca; green), and predicted signal peptide (purple).

FIG 2

NTH1 and NTH2 are differentially expressed. (A) NTH2 is induced during sexual development. Lanes: 1 and 2, a and α yeast grown in rich medium (YPD liquid); 3, a and α yeast mixed under cross conditions (V8 agar, pH 7, at 22°C for 72 h); 4, spores from a cross after 72 h. Top, middle, and bottom images reflect probes to the transcripts of interest: NTH2, NTH1, and GPD1, respectively. (B) NTH2 is induced on V8 agar. Lanes: 1 and 2, a and α yeast grown on rich medium (YPD agar); 3 and 4, a and α yeast grown individually on V8 agar for 48 h; 5, a and α yeast grown in mixed culture on V8 agar for 48 h. Probes used were to NTH2 and GPD1. GPD1 was used as a loading and hybridization control.

FIG 3

Deletion of NTH1 and NTH2 causes a severe synthetic defect in trehalose utilization. (A) Wild-type a, wild-type α, a nth1Δ, α nth1Δ, a nth2Δ, α nth2Δ, a nth1Δnth2Δ, and α nth1Δnth2Δ strains (left) were struck onto synthetic defined media containing either 2% glucose (middle) or 2% trehalose (right) as a sole carbon source and grown for 2 days at 30°C. (B) Trehalose was quantified in wild-type and mutant crosses after 5 days on V8, pH 7.0. Mean trehalose values from three biological replicates were compared using one-way ANOVA statistics, and those crosses that differed significantly (P < 0.05) from the wild type (a × α) are marked with an asterisk.

FIG 4

Deletion of NTH1 and NTH2 leads to aberrant sexual development. (A) The wild-type, nth1Δ, nth2Δ, and nth1Δ nth2Δ crosses after 5 days all show normal filament formation. Magnification, ×100; bar, 100 μm. (B) nth1Δ nth2Δ crosses form large, ball-like structures in place of basidia and do not produce spore chains. Magnification, ×400; bar, 10 μm. (C) After 7 days of development, nth1Δ nth2Δ crosses produced numerous large, swollen nodules on filaments, giving them a beads-on-a-string appearance (black arrows), consistent with previous descriptions of chlamydospores. Magnification, ×200; bar, 50 μm.

FIG 5

Trehalase-deficient spores are defective for growth. Spores from the wild type and nth1Δ, nth2Δ, and nth1Δ nth2Δ crosses were isolated, counted, and cultured on rich medium for 3 days at 30°C. The x axis represents the genotype of spores. The y axis represents the percentage of spores plated that produced a colony. Resulting mean germination frequencies from three independent biological replicates were compared using ANOVA, with only nth1Δ nth2Δ spores having a significant reduction (P < 0.05) in germination.

FIG 6

Strains harboring nth2Δ are hypervirulent in mice. The x axis shows time, in days, after tail vein infection with 1 × 10⁶ α wild-type (diamonds), α nth1Δ (squares), α nth2Δ (triangles), or α nth1Δ nth2Δ (cross hatches) yeast. The y axis shows percent survival of five BALB/c mice per strain. The difference in average time to death between nth2Δ and nth1Δ nth2Δ strains versus wild-type and nth1Δ strains is 7 days and is both statistically significant (P < 0.05 by Wilcoxon rank-sum test) and reproducible between experiments.