Distinct Roles of Two DNA Methyltransferases from Cryphonectria parasitica in Fungal Virulence, Responses to Hypovirus Infection, and Viral Clearance

Abstract

Two DNA methyltransferase (DNMTase) genes from Cryphonectria parasitica have been previously identified as CpDmt1 and CpDmt2, which are orthologous to rid and dim-2 of Neurospora crassa, respectively. While global changes in DNA methylation have been associated with fungal sectorization and CpDmt1 but not CpDmt2 has been implicated in the sporadic sectorization, the present study continues to investigate the biological functions of both DNMTase genes. Transcription of both DNMTases is regulated in response to infection with the Cryphonectria hypovirus 1 (CHV1-EP713). CpDmt1 is upregulated and CpDmt2 is downregulated by CHV1 infection. Conidium production and response to heat stress are affected only by mutation of CpDmt1, not by CpDmt2 mutation. Significant changes in virulence are observed in opposite directions; i.e., the CpDmt1-null mutant is hypervirulent, while the CpDmt2-null mutant is hypovirulent. Compared to the CHV1-infected wild type, CHV1-transferred single and double mutants show severe growth retardation: the colony size is less than 10% that of the parental virus-free null mutants, and their titers of transferred CHV1 are higher than that of the wild type, implying that no defect in viral replication occurs. However, as cultivation proceeds, spontaneous viral clearance is observed in hypovirus-infected colonies of the null mutants, which has never been reported in this fungus-virus interaction. This study demonstrates that both DNMTases are significant factors in fungal development and virulence. Each fungal DNMTase affects fungal biology in both common and separate ways. In addition, both genes are essential to the antiviral responses, including viral clearance which depends on their mutations.IMPORTANCE Although relatively few in number, studies of DNA methylation have shown that fungal DNA methylation is implicated in development, genome integrity, and genome defense. While fungal DNMTase has been suggested as playing a role in genome defense, studies of the biological function of fungal DNMTase have been very limited. In this study, we have shown distinct biological functions of two DNA methyltransferases from the chestnut blight fungus C. parasitica We have demonstrated that DNMTases are important to fungal development and virulence. In addition, these genes are shown to play an important role in the fungal response to hypoviral CHV1 infection, including severely retarded colonial growth, and in viral clearance, which has never been previously observed in mycovirus infection. These findings provide a better understanding of the biological functions of fungal DNA methyltransferase and a basis for clarifying the epigenetic regulation of fungal virulence, responses to hypovirus infection, and viral clearance.

Keywords: Cryphonectria parasitica; DNA methyltransferase; fungal growth and development; hypovirulence; viral clearance.

FIG 1

Expression analyses of CpDmt1 and CpDmt2. (A and B) qRT-PCR results of expression levels of CpDmt1 and CpDmt2 during cultivation in standard liquid EP complete medium. (C and D) qRT-PCR results of the expression levels of CpDmt1 and CpDmt2 during cultivation on tannic acid (TA)-supplemented medium. The numbers of days (d) and hours (h) in liquid culture and TA-supplemented medium are shown below the bars in panels A to D. The strains are indicated in the top leftmost corner of each panel, and media with (+) or without (–) TA supplementation are indicated. At least three individual experiments were performed. Values are means plus standard deviations (SD) (error bars). Student’s t test was used to compare data between two groups (**. P < 0.01). Strains used were virus-free wild-type EP155/2 and its isogenic CHV1-infected UEP1 strain.

FIG 2

Colony morphology and conidium production of the mutant strains. (A) Colonies are shown after 14 days of cultivation on PDAmb. (B) Analysis of conidium production. Strains are indicated below the photographs and include the wild-type EP155/2, CpDmt1-null mutant (TdDMT1), a sectored progeny of TdDMT1 (TdDMT1-S), CpDmt2-null mutant (TdDMT2), complementing strains of TdDMT1 and TdDMT2 (TcDMT1 and TcDMT2, respectively), two independently isolated double mutants constructed through CpDmt2-null mutation from TdDMT1 (TdDMT1/2-1 and -2), and two independently isolated double mutants constructed through CpDmt1-null mutation from TdDMT2 (TdDMT2/1-1 and -2). The mean comparisons of conidia harvested from at least three individual experiments were analyzed using one-way ANOVA with Duncan’s method. The means with a common letter are not significantly different (**, P < 0.01). Values are means plus SD (error bars).

FIG 3

Colony morphology of DNMTase mutant strains in response to temperature stress conditions. Temperature stress was induced by incubating the plates at high (30°C) and low (20°C) temperatures relative to the standard 25°C condition.

FIG 4

Effect of hypovirus infection on the mutant strains. (A) Colony morphology of CHV1-free (top) and CHV1-infected (bottom) mutant strains. Virus-infected isogenic strains (V⁺) are indicated. (B) qRT-PCR analysis of viral single-stranded RNA (ssRNA) accumulation. The levels of ssRNA accumulation of indicated strains are represented as the fold change relative to that of the UEP1. At least three individual experiments were performed. Values are means ± SD (error bars). Student’s t test was used to compare data between two groups (**, P < 0.01). Virus-free, virus-transferred, and cured strains are indicated by (0), (+), and (−), respectively.

FIG 5

Virulence assays. (A) Bavendamm’s assay for polyphenol oxidase activity of strains. The level of brown coloration correlates with polyphenol oxidase activity of each strain. (B) Stromal pustule eruption on chestnut tree stems of strains. (C) Virulence assay using excised chestnut tree bark. Lesion measurement values are shown in square millimeters. Different letters indicate significant differences between treatments according to Duncan’s multiple range test (**, P < 0.01). Values are mean plus SD (error bars). DNMTase-null mutant strains containing CHV1 (V⁺) are indicated.

FIG 6

Colony morphology of CHV1-infected, curing, and cured mutant strains and expression analyses of antiviral genes CpDcl2 and CpAgl2. (A) Strains indicated at the left of the panel are the wild-type EP155/2, CpDmt1-null mutant (TdDMT1), and the CpDmt2-null mutant (TdDMT2). (B) Virus-infected isogenic strains of panel A are shown, and UEP1, which is virus infected and isogenic to the wild-type EP155/2, is shown as a control. (C) Colony morphology of mutant strains showing the curing of CHV1 are represented as fast growing and well pigmented mycelial areas. (D) Colony morphology of CHV1-cured strains. (E and F) qRT-PCR analysis results for CpDcl2 and CpAgl2. Changes in expression of CpDcl2 and CpAgl2 among mutant strains relative to the level of gpd are shown. Strains are indicated below the bars. Virus-free, -transferred, and cured strains are indicated by (0), (+), and (–), respectively. Different letters indicate significant differences between treatments according to Duncan’s multiple range test (**, P < 0.01).

FIG 7

Colony morphology during hypoviral transfer through hyphal anastomosis. Hypoviral transfers during cocultivation from CHV1-infected UEP1 to strains related to CpDmt1-null mutation (top row) and CpDmt2-null mutation (bottom row) are shown. D(V), R, and R(V) indicate the virus-donor, virus-recipient, and putative virus-transferred recipient, respectively.