A Novel Mycovirus Evokes Transcriptional Rewiring in the Fungus Malassezia and Stimulates Beta Interferon Production in Macrophages

Abstract

Mycoviruses infect fungi, and while most persist asymptomatically, there are examples of mycoviruses having both beneficial and detrimental effects on their host. Virus-infected Saccharomyces and Ustilago strains exhibit a killer phenotype conferring a growth advantage over uninfected strains and other competing yeast species, whereas hypovirus-infected Cryphonectria parasitica displays defects in growth, sporulation, and virulence. In this study, we identify a double-stranded RNA (dsRNA) mycovirus in five Malassezia species. Sequence analysis reveals it to be a totivirus with two dsRNA segments: a larger 4.5-kb segment with genes encoding components for viral replication and maintenance, and a smaller 1.4-kb segment encoding a novel protein. Furthermore, transcriptome sequencing (RNA-seq) of virus-infected versus virus-cured Malassezia sympodialis revealed an upregulation of dozens of ribosomal components in the cell, suggesting the virus modifies the transcriptional and translational landscapes of the cell. Given that Malassezia is the most abundant fungus on human skin, we assessed the impact of the mycovirus in a murine epicutaneous infection model. Although infection with virus-infected strains was not associated with an increased inflammatory response, we did observe enhanced skin colonization in one of two virus-infected M. sympodialis strains. Noteworthy, beta interferon expression was significantly upregulated in bone marrow-derived macrophages when challenged with virus-infected, compared to virus-cured, M. sympodialis, suggesting that the presence of the virus can induce an immunological response. Although many recent studies have illuminated how widespread mycoviruses are, there are relatively few in-depth studies about their impact on disease caused by the host fungus. We describe here a novel mycovirus in Malassezia and its possible implications in pathogenicity.IMPORTANCEMalassezia species represent the most common fungal inhabitant of the mammalian skin microbiome and are natural skin commensal flora. However, these fungi are also associated with a variety of clinical skin disorders. Recent studies have reported associations of Malassezia with Crohn's disease and pancreatic cancer, further implicating this fungal genus in inflammatory and neoplastic disease states. Because M. sympodialis has lost genes involved in RNA interference (RNAi), we hypothesized Malassezia could harbor dsRNA mycoviruses. Indeed, we identified a novel mycovirus of the totivirus family in several Malassezia species and characterized the MsMV1 mycovirus of M. sympodialis We found conditions that lead to curing of the virus and analyzed isogenic virus-infected/virus-cured strains to determine MsMV1 genetic and pathogenic impacts. MsMV1 induces a strong overexpression of transcription factors and ribosomal genes, while downregulating cellular metabolism. Moreover, MsMV1 induced a significantly higher level of beta interferon expression in cultured macrophages. This study sheds light on the mechanisms of pathogenicity of Malassezia, focusing on a previously unidentified novel mycovirus.

Keywords: Malassezia; beta interferon; mycoviruses; skin.

FIG 1

Malassezia species harbor dsRNA segments. (A) dsRNA from representative M. sympodialis strains. (B) dsRNA from additional Malassezia species. Strain names can be found in Table 1. (C) dsRNA (upper panel) and ssRNA (lower panel) from M. sympodialis strains passaged at 30°C, at 37°C, and on mDixon medium containing 10 μg/ml biphenyl. All dsRNA and ssRNA extracts were visualized by electrophoresis in 0.8% agarose gels with 0.5 μg/ml ethidium bromide.

FIG 2

dsRNA segments code for viral protein products. (A) Representative region of the large dsRNA segment illustrating the two open reading frames for the capsid and Rdp1 and the site of the −1 ribosomal frameshift. Stop codon for capsid highlighted in red. Start codon for Rdp1 highlighted in green. Frameshift site highlighted in blue. (B) Phylogenetic analysis of mycovirus Rdp1 sequences. Scale bar represents the number of substitutions per site. Blue boxes on protein diagrams represent the reverse transcriptase (RT) domain of the Rdp1 protein. (C) Protein sequence of the small dsRNA segment. The signal peptide is underlined, and hydrophobic residues are indicated with stars. (D) Pictorial representation of predicted domains of the small dsRNA segment.

FIG 3

RNA-seq reveals transcriptional rewiring in virus-infected M. sympodialis strain. (A) Dot plot of log₂ fold change in gene expression in the virus-infected strain KS012. Red and blue dots represent upregulated and downregulated genes that pass a false-discovery rate (FDR) of 5%, respectively. Gray dots indicate genes that do not meet the FDR. The virus-cured strain SEC494 was used as the baseline for gene expression, and a gene is marked as differentially expressed in the virus-infected strain KS012 if expression differs from SEC494. (B) Number of genes upregulated (red) and downregulated (blue) categorized by gene ontology. (C) Heatmap of the three RNA-seq technical replicates of virus-infected and virus-cured strains. Green represents upregulated genes, and red represents downregulated genes. (D) Gene network map of upregulated genes in virus-infected strain. Genes are clustered by color based on the Markov-Cluster (MCL) inflation parameter.

FIG 4

MsMV1 dsRNA virus induces beta interferon (IFN-β) expression in cultured macrophages in a TLR3-independent manner. (A) IFN-β expression in bone marrow-derived macrophages (BMDMs) after infection with the indicated virus-infected and virus-cured strains at an MOI of 5 for 24 h. Bars are the mean + SEM for four replicate samples with each being the pool of two separate wells. Data are pooled from two independent experiments. (B) IFN-β expression by wild-type (solid bars) and Tlr3^−/− (striped bars) BMDMs after infection with the indicated virus-infected and virus-cured strains at an MOI of 5 for 24 h. Bars are a pool of two wells. Data are representative of one out of two independent experiments. Poly(I:C) was included as a positive control; US, unstimulated. The dotted line indicates the expression levels under unstimulated conditions. Statistics were calculated using unpaired Student’s t test. *, P < 0.05; **, P < 0.01. Black bars below the x axis labels group virus-infected/virus-cured strain pairs. “+” indicates virus infected; “−” indicates virus cured.