The Cryptococcus neoformans catalase gene family and its role in antioxidant defense

Abstract

In the present study, we sought to elucidate the contribution of the Cryptococcus neoformans catalase gene family to antioxidant defense. We employed bioinformatics techniques to identify four members of the C. neoformans catalase gene family and created mutants lacking single or multiple catalase genes. Based on a phylogenetic analysis, CAT1 and CAT3 encode putative spore-specific catalases, CAT2 encodes a putative peroxisomal catalase, and CAT4 encodes a putative cytosolic catalase. Only Cat1 exhibited detectable biochemical activity in vitro, and Cat1 activity was constitutive in the yeast form of this organism. Although they were predicted to be important in spores, neither CAT1 nor CAT3 was essential for mating or spore viability. Consistent with previous studies of Saccharomyces cerevisiae, the single (cat1, cat2, cat3, and cat4) and quadruple (cat1 cat2 cat3 cat4) catalase mutant strains exhibited no oxidative-stress phenotypes under conditions in which either exogenous or endogenous levels of reactive oxygen species were elevated. In addition, there were no significant differences in the mean times to mortality between groups of mice infected with C. neoformans catalase mutant strains (the cat1 and cat1 cat2 cat3 cat4 mutants) and those infected with wild-type strain H99. We conclude from the results of this study that C. neoformans possesses a robust antioxidant system, composed of functionally overlapping and compensatory components that provide protection against endogenous and exogenous oxidative stresses.

FIG. 1.

Phylogenetic analysis of the C. neoformans catalases. The phylogenetic tree of fungal catalases and selected animal, protist, bacterial, and archaeal catalases is rooted with two plant catalases. Homologs were identified with BLASTP searches of C. neoformans Cat1, Cat2, Cat3, and Cat4 proteins against the nonredundant protein database from NCBI. A multiple sequence alignment was performed automatically with MUSCLE, and the tree was constructed via NJDIST and PROTML (available in MOLPHY). Numbers on branches indicate the bootstrap values produced by PROTML running with the -R option and starting with an input neighbor-joining tree calculated from NJDIST. Some bootstrap values were removed at the tips of the tree for clarity in visualizing the tree. There are four distinct clades of fungal catalases: clade P, the peroxisomal catalases; clade C, the cytoplasmic catalases; clade A, spore-specific catalases; and clade B, primarily secreted catalases. C. neoformans possesses catalases in three of the four clades.

FIG. 2.

Constructionof catalase null mutant strains. Overlap PCR was performed to create the cat1::URA5, cat2::NAT, cat3::URA5, and cat4::URA5 deletion constructs. Allele-specific integration of constructs at the native catalase loci resulted in the deletion of 39%, 50%, 54%, and 74% of the Cat1, Cat2, Cat3, and Cat4 catalase domains, respectively.

FIG. 3.

Southern blot analysis confirmed deletion of the native catalase alleles. (A to D) Genomic DNA was isolated from cat1, cat2, cat3, and cat4 mutant strains and wild-type strain H99 and used to perform Southern blot analysis, as described previously. Southern blot analysis confirmed that a single allele-specific integration event occurred at each catalase locus. As shown in Fig. 2, restriction enzymes were chosen so that one of the two restriction sites used to digest genomic DNA was outside of the deletion construct. WT, wild type.

FIG. 4.

Cat1 is the sole catalase with activity in vitro. (A) Cell lysates from S. cerevisiae (S. c) and C. neoformans wild-type (WT) strain H99 (C. n) grown at 30°C in YPD medium were separated on a 10% acrylamide gel under nondenaturing conditions. Catalase activity was visualized by potassium ferricyanide-negative staining. The two bands in lane 1 correspond to the S. cerevisiae catalases Ctt1 and Cta1. (B) Native polyacrylamide gel electrophoresis of protein extracts from C. neoformans cat1, cat2, cat3, and cat4 mutant strains and wild-type strain H99 grown at 30°C in YPD medium. A single activity band was observed for lysates from all strains except the cat1 mutant strain. (C) Native polyacrylamide gel electrophoresis was performed with protein extracts and cell culture supernatants of C. neoformans wild-type strain H99 cells that were either treated with hydrogen peroxide (0.5 mM or 1.0 mM), grown at an elevated temperature (37°C), or grown in yeast nitrogen base (YNB) medium with 2% glucose. Each lane was loaded with 20 μg of total protein. Each gel represents one of at least three independent experiments.

FIG. 5.

Catalase mutant strains do not exhibit an oxidative-stress phenotype. Sensitivity to oxidative stress was assessed by a disc diffusion assay. Sterile discs were saturated with 10 μl of 8% (A and C) or 16% (B, D, and E) hydrogen peroxide. Plates were incubated at either 30°C (A and B) or 37°C (B, D, and E). No differences in the diameters of zones of inhibition were observed among any of the catalase mutant strains compared to that of the wild-type (WT) strain. Results represent mean values ± the standard errors of the means for three or more experiments.

FIG. 6.

Role of catalases in C. neoformans sexual differentiation. (A) Cat1 and Cat3 are not required for sexual differentiation in C. neoformans. Wild-type MATα and cat1α and cat3α mutant strains were each cocultured with a MATa wild-type tester strain on V8 mating medium. In addition, the cat3α mutant strain was cocultured with the cat1a mutant strain. (B) Cat4 appears to contribute to the sexual differentiation of C. neoformans. MATα and MATa pairs of each catalase mutant strain were cocultured on V8 mating medium. (C) A quadruple catalase mutant is not impaired for sexual differentiation. The MATα quadruple strain was cocultured with each MATa catalase mutant strain on V8 mating medium. Plates were incubated for 2 weeks. Representative crosses were photographed at a magnification of ×90.

FIG. 7.

Cat1 activity is not required for virulence. Groups of 10 A/Jcr mice were anesthetized by intraperitoneal injection of pentobarbital and infected with 5 × 10⁵ CFU of cat1 (A) and cat1 cat2 cat3 cat4 (B) mutant strains (two independent mutants) and wild-type strain H99. All mice infected with the catalase mutant strains and wild-type strain H99 succumbed to infection and died by day 25. The absence of differences in the mean times to death demonstrates that catalase does not contribute to the virulence composite.