Transcription of genes in the biosynthetic pathway for fumonisin mycotoxins is epigenetically and differentially regulated in the fungal maize pathogen Fusarium verticillioides

Abstract

When the fungal pathogen Gibberella moniliformis (anamorph, Fusarium verticillioides) colonizes maize and maize-based products, it produces class B fumonisin (FB) mycotoxins, which are a significant threat to human and animal health. FB biosynthetic enzymes and accessory proteins are encoded by a set of clustered and cotranscribed genes collectively named FUM, whose molecular regulation is beginning to be unraveled by researchers. FB accumulation correlates with the amount of transcripts from the key FUM genes, FUM1, FUM21, and FUM8. In fungi in general, gene expression is often partially controlled at the chromatin level in secondary metabolism; when this is the case, the deacetylation and acetylation (and other posttranslational modifications) of histones are usually crucial in the regulation of transcription. To assess whether epigenetic factors regulate the FB pathway, we monitored FB production and FUM1, FUM21, and FUM8 expression in the presence of a histone deacetylase inhibitor and verified by chromatin immunoprecipitation the relative degree of histone acetylation in the promoter regions of FUM1, FUM21, and FUM8 under FB-inducing and noninducing conditions. Moreover, we generated transgenic F. verticillioides strains expressing GFP under the control of the FUM1 promoter to determine whether its strength under FB-inducing and noninducing conditions was influenced by its location in the genome. Our results indicate a clear and differential role for chromatin remodeling in the regulation of FUM genes. This epigenetic regulation can be attained through the modulation of histone acetylation at the level of the promoter regions of the key biosynthetic genes FUM1 and FUM21, but less so for FUM8.

Fig 1

Effects of the HDAC inhibitor TSA on FUM gene expression in wild-type VP2, FR3, and GE1 strains of F. verticillioides. The strains were cultivated in the non-FB-inducing Czapek medium in the absence or in the presence of 1 μM TSA. After 7 days of growth, the amounts of transcripts of FUM1 (A), FUM21 (B), and FUM8 (C) were estimated by the absolute quantification method in RT-qPCR. For each gene, expression was normalized to TUB2 transcript abundance. Mean and standard error were calculated with data from three biological and three analytical replicates. Asterisks indicate a P value of <0.05.

Fig 2

HPLC quantification of FB1 production in wild-type VP2, FR3, and GE1 strains of F. verticillioides. Values are the means of three independent culture batches for each strain, grown for 7 days in liquid Czapek medium (40 ml, inoculated with 10⁵ spores). The role of HDACs in FB1 production was tested by adding the HDAC inhibitor TSA to the medium to a final concentration of 1 μM. Means and standard deviations were calculated with data from three biological replicates.

Fig 3

Relative acetylation levels of the FUM1 (A), FUM21 (B), and FUM8 (C) gene promoters under different conditions. Strain VP2 was cultured for 7 days in Czapek medium, in Czapek medium containing 1 μM TSA, an HDAC inhibitor, or in fructose-enriched, FB-inducing medium. Acetylation levels were analyzed by ChIP with antibodies specific to hyperacetylated histone H4 (Penta). The abundance of target DNA was quantified by the comparative C_T method (Applied Biosystems) with TUB2 as the endogenous reference for normalization. Error bars indicate the standard errors of two biological and three technical repetitions. Different letters indicate significant differences (P < 0.01). −IND, non-FB-inducing medium; +IND, FB-inducing medium.

Fig 4

(A) Fluorescence micrographs of the F. verticillioides transformed strain Fv-P_FUM1-1 grown in FB-inducing GYAM (pH 3) for 1 week. Conidia, conidiophores, and phialides are the most intensely labeled (white arrows). The lower panels are the corresponding bright-field images. (B) Roots of etiolated maize seedlings colonized by GFP-expressing hyphae; confocal images were taken under UV light 3 weeks after sowing of Fv-P_FUM1-1-infected seeds. Blastospores budding from a growing hyphal tip in a xylem vessel are indicated by a white arrow. Plant cell walls fluoresce in red, mycelium in green (total depth, 235 μm; resolution, 0.36 × 0.36 × 1.5 μm).

Fig 5

FUM1 and/or GFP transcript abundance in the Fv-P_FUM1-1 and -3 transformants and in the Fv-P_TOXA∷GFP strain grown for 1 week in Czapek medium (non-FB-inducing) (A) and grown for 20 days in fructose-enriched, FB-inducing medium (B). For each gene, transcript amounts were estimated by the absolute quantification method in RT-qPCR after normalization to TUB2 transcripts. Means and standard errors were calculated with data from three biological and three analytical replicates. Asterisks indicate a P value of <0.01.