The Protective Role of 1,8-Dihydroxynaphthalene-Melanin on Conidia of the Opportunistic Human Pathogen Aspergillus fumigatus Revisited: No Role in Protection against Hydrogen Peroxide and Superoxides.

Previously, 1,8-dihydroxynaphthalene (DHN)-melanin was described to protect Aspergillus fumigatus against hydrogen peroxide (H2O2), thereby protecting this opportunistic human pathogen from reactive oxygen species generated by the immune system. This was based on the finding that the ATCC 46645 mutant with mutations in the pksP gene of the DHN-melanin synthesis pathway showed increased sensitivity to reactive oxygen species compared to the wild type. Here, it is shown that deletion of the pksP gene in A. fumigatus strain CEA10 did not affect sensitivity for H2O2 and superoxide in a plate stress assay. In addition, direct exposure of the dormant white conidia of the pksP deletion strains to H2O2 did not result in increased sensitivity. Moreover, complementation of the ATCC 46645 pksP mutant strain with the wild-type pksP gene did result in pigmented conidia but did not rescue the H2O2-sensitive phenotype observed in the plate stress assay. Genome sequencing of the ATCC 46645 pksP mutant strain and its complemented strain revealed a mutation in the cat1 gene, likely due to the UV mutagenesis procedure used previously, which could explain the increased sensitivity toward H2O2. In summary, DHN-melanin is not involved in protection against H2O2 or superoxide and, thus, has no role in survival of conidia when attacked by these reactive oxygen species. IMPORTANCE Opportunistic pathogens like Aspergillus fumigatus have strategies to protect themselves against reactive oxygen species like hydrogen peroxides and superoxides that are produced by immune cells. DHN-melanin is the green pigment on conidia of Aspergillus fumigatus and more than 2 decades ago was reported to protect conidia against hydrogen peroxide. Here, we correct this misinterpretation by showing that DHN-melanin actually is not involved in protection of conidia against hydrogen peroxide. We show that UV mutagenesis that was previously used to select a pksP mutant generated many more genome-wide mutations. We discovered that a mutation in the mycelial catalase gene cat1 could explain the observed phenotype of increased hydrogen peroxide sensitivity. Our work shows that UV mutagenesis is not the preferred methodology to be used for generating mutants. It requires genome sequencing with single-nucleotide polymorphism analysis as well as additional validations to discard unwanted and confirm correct phenotypes.

Correction: Variation of virulence of five Aspergillus fumigatus isolates in four different infection models.

[This corrects the article DOI: 10.1371/journal.pone.0252948.].

Variation of virulence of five Aspergillus fumigatus isolates in four different infection models.

Conidia of Aspergillus fumigatus are inhaled by humans on daily basis. As a consequence, these conidia can cause infections that differ in severity ranging from allergic bronchopulmonary aspergillosis to invasive aspergillosis. In this study we compared virulence of five A. fumigatus isolates in four different infection models to address the predictive value of different model systems. Two of the A. fumigatus strains were isolated from dogs with a non-invasive sino-nasal aspergillosis (DTO271-B5 and DTO303-F3), while three strains were isolated from human patients with invasive aspergillosis (Af293, ATCC46645 and CEA10). Infection models used encompassed cultured type II A549 lung epithelial cells, Protostelium aurantium amoeba, Galleria melonella larvae and zebrafish embryos. No major differences in virulence between these five strains were observed in the lung epithelial cell model. In contrast, strain ATCC46645 was most virulent in the amoeba and zebrafish model, whereas it was much less virulent in the Galleria infection model. DTO303-F3 was most virulent in the latter model. In general, reference strain Af293 was less virulent as compared to the other strains. Genome sequence analysis showed that this latter strain differed from the other four strains in 136 SNPs in virulence-related genes. Together, our results show that virulence of individual A. fumigatus strains show significant differences between infection models. We conclude that the predictive value of different model systems varies since the relative virulence across fungal strains does not hold up across different infection model systems.

The sino-nasal warzone: transcriptomic and genomic studies on sino-nasal aspergillosis in dogs.

We previously showed that each dog with chronic non-invasive sino-nasal aspergillosis (SNA) was infected with a single genotype of Aspergillus fumigatus. Here, we studied the transcriptome of this fungal pathogen and the canine host within the biofilm resulting from the infection. We describe here transcriptomes resulting from natural infections in animal species with A. fumigatus. The host transcriptome showed high expression of IL-8 and alarmins, uncontrolled inflammatory reaction and dysregulation of the Th17 response. The fungal transcriptome showed in particular expression of genes involved in secondary metabolites and nutrient acquisition. Single-nucleotide polymorphism analysis of fungal isolates from the biofilms showed large genetic variability and changes related with adaptation to host environmental factors. This was accompanied with large phenotypic variability in in vitro stress assays, even between isolates from the same canine patient. Our analysis provides insights in genetic and phenotypic variability of Aspergillus fumigatus in biofilms of naturally infected dogs reflecting in-host adaptation. Absence of a Th17 response and dampening of the Th1 response contributes to the formation of a chronic sino-nasal warzone.