The KdmB-EcoA-RpdA-SntB chromatin complex binds regulatory genes and coordinates fungal development with mycotoxin synthesis

Abstract

Chromatin complexes control a vast number of epigenetic developmental processes. Filamentous fungi present an important clade of microbes with poor understanding of underlying epigenetic mechanisms. Here, we describe a chromatin binding complex in the fungus Aspergillus nidulans composing of a H3K4 histone demethylase KdmB, a cohesin acetyltransferase (EcoA), a histone deacetylase (RpdA) and a histone reader/E3 ligase protein (SntB). In vitro and in vivo evidence demonstrate that this KERS complex is assembled from the EcoA-KdmB and SntB-RpdA heterodimers. KdmB and SntB play opposing roles in regulating the cellular levels and stability of EcoA, as KdmB prevents SntB-mediated degradation of EcoA. The KERS complex is recruited to transcription initiation start sites at active core promoters exerting promoter-specific transcriptional effects. Interestingly, deletion of any one of the KERS subunits results in a common negative effect on morphogenesis and production of secondary metabolites, molecules important for niche securement in filamentous fungi. Consequently, the entire mycotoxin sterigmatocystin gene cluster is downregulated and asexual development is reduced in the four KERS mutants. The elucidation of the recruitment of epigenetic regulators to chromatin via the KERS complex provides the first mechanistic, chromatin-based understanding of how development is connected with small molecule synthesis in fungi.

Figure 1.

Discovery of the novel chromatin modifier KdmB-EcoA-RpdA-SntB (KERS) complex. (A) Selected interactors of KdmB, EcoA, RpdA and SntB, as identified with TAP/GFP/HA fusions by immunoprecipitation coupled with LC–MS/MS analysis. Fungal mycelia were harvested for immunoprecipitation at the end of 24 h vegetative culture at 37ºC. (B) Domain architecture of KERS complex members. PHD: plant homeodomain, ZF: zinc finger, A/B: arid/bright domain, BAH: Bromo-adjacent homology domain, SANT: Swi3, Ada2, N-Cor and TFIIIB domain, ATase: acetyltransferase. Numbers signify positions of the domains. (C) Orthologue analysis of the KERS complex among various organisms. KERS components are conserved from yeast to human. Trees were created using baker's yeast, Aspergillus, Drosophila, mouse and human homologs. (D) Bimolecular fluorescence complementation analysis with yellow fluorescent protein (YFP), verifying subcellular KdmB-EcoA, KdmB-RpdA and KdmB-SntB binary interactions. The images were captured by confocal microscopy at the end of 16 h incubation at 30 ºC under illumination. Nuclei were visualized in red by a monomeric red fluorescent protein fused to histone 2A (mRFP-H2A). (E) Cellular localization of KERS complex subunits. Confocal microscopy analysis of KdmB, EcoA, RpdA, SntB fused to GFP in a mRFP-H2A expressing strain. Strains were incubated and observed as above. (F) Identification of KERS subcomplexes in the absence of KdmB or SntB. LC–MS/MS analysis of EcoA, RpdA and SntB TAP purifications in kdmBΔ and sntBΔ genetic background, respectively. Growth conditions are the same as in (A).

Figure 2.

Cellular levels and enzymatic activities of KERS complex. Subcellular localizations of KERS GFP fusions in (A) kdmB, (B) sntB and (C) kdmB/sntB double mutants. (D) Cellular levels of KdmB-EcoA-RpdA-SntB in the WT and mutants. Fungal cultures were grown in submerged liquid GMM for 24 h at 37 ºC. α-HA monoclonal mouse and generic α-SkpA polyclonal rabbit antibodies were used to detect HA-fused protein and SkpA as a loading control, respectively. Crude extract (100 μg total protein) was used for immunodetection. Proteasome inhibitor epoxomicin (EPOX, 20 μM) was supplemented at the end of 20 h vegetative growth for a further 4 h of incubation. (E) Quantification of the EcoA^HA fusion in different backgrounds or treatments. EcoA^HA/SkpA ratio was used for quantification in biological triplicates. (F) LC–MS/MS analysis of phosphorylated residues on EcoA in WT, kdmBΔ and in kdmBΔ/sntBΔ strains. The numbers indicate the position of amino acids. Red stars represent phosphorylated residues. Serines are depicted in bold. Underlined amino acids represent peptide coverage in MS analysis with high confidence. (G) Expression of EcoA^HA and its mutant versions of S41A, S45A, S41A/S45A, S41D, S45D, S41D/S45D in kdmBΔ and WT. (H) Comparison of expression levels of EcoA^HA and S41A/S45A mutant of EcoA^HA left (blot) and right (quantification) from three biological triplicates. (I) Acetylation of SudA^K106/107 (yeast Smc3 homolog) cohesin by EcoA and loss of acetylation in ecoA^TetON. Red stars and bold letters represent acetylated residues. Kit-enriched acetylated peptides from immunoprecipitation of SudA^HA were analysed in LC–MS/MS. (J) Verification of LC–MS/MS acetylation levels of immunoprecipitated SudA^HA with Lys-Acetylation specific antibodies. α-Lys-Ac/α-HA signal ratio was used for quantification of acetylation from four biological replicates. (K) In vitro histone deacetylase activity (HDAC) of purified KERS complexes. T represents TAP of the corresponding proteins in WT and mutants. Enzymatic activity of enriched RpdA complexes was measured in triplicates using [H-3]-acetate prelabelled chicken histones (52). Blank levels were subtracted and all samples were calculated with respect to RpdA-TAP activity adjusted to 100%. HDAC activity assay is the average of three independent biological replicates. Significance relative to D193A as calculated by Student's t-test is indicated as asterisks: P < 0.05 (*) and P < 0.01 (**).

Figure 3.

KdmB, EcoA, RpdA and SntB co-localized at thousands of promoters (A) A VENN diagram showing the number of common and unique genes between KdmB^HA, EcoA^HA, RpdA^HA and SntB^HA. (B) Genome-browser screenshots showing the binding location of KdmB^HA, EcoA^HA, RpdA^HA and SntB^HA. (C) Heatmaps displaying binding locations, signals and intensities of KdmB^HA, EcoA^HA, RpdA^HA and SntB^HA over their overall target genes (n = 4165). (D) A pie chart showing the distribution of the common KERS binding sites at distal promoters, proximal promoters, genic regions (exon/intron) and gene ends. (E, F) Heatmaps displaying ChIPseq signals of (E) KdmB^HA, histone H3 lysine 4 trimethylation (H3K4me3), histone H3 lysine 9 acetylation (H3K9ac), histone H3 lysine 27 acetylation, histone H3, (F) TBP^HA and TFIIB^HA at the –3 kb to + 3 kb region with respect to the start codon (ATG) of KERS target genes. (G) A line graph showing the median binding signals of KdmB^HA, EcoA^HA, RpdA^HA, SntB^HA, TBP^HA and TFIIB^HA at the –2 kb to + 2 kb region with respect to the start codon (ATG) of KERS target genes. (H) A scatter plot showing the correlation between KdmB bindings at promoters and Pol II occupancies at coding regions. (I) A line plot showing the median Pol II ChIPseq signals between 2 kb upstream (–2 kb) of the translational start codon (ATG) to 1 kb downstream (+1 kb) of the translational stop codon of expressed genes in the WT and mutant strains deleted (Δ) or depleted (TetON) for the indicated subunit. The asterisk (*) indicates strains grown under the same experimental conditions for protein expression shutdown by the TetON system. (J) Heatmap plots showing binding locations, signals and intensities of KdmB^HA, EcoA^HA, RpdA^HA and SntB^HA in WT, kdmBΔ or sntBΔ background. (K) Genome-browser screenshots showing binding of KdmB^HA, EcoA^HA, RpdA^HA and SntB^HA in WT (+), kdmBΔ or sntBΔ background.

Figure 4.

The KERS complex regulates diverse physiological pathways in development and the carcinogen sterigmatocystin biosynthesis. (A) A heat box plot showing the correlations of transcriptional effects (measured by Pol II ChIPseq signals at coding regions in the respective mutant versus WT) between the KERS mutants grown under primary (20 h) and secondary (48 h) growth phases. (B) PCA analysis on the transcriptional effects (measured by Pol II ChIPseq signals at coding regions in the respective mutant versus WT) of the KERS mutants grown under primary (20 h) and secondary (48 h) growth phases. (C) The number of up- and down- regulated genes (Pol II ChIPseq) during transition from primary (20 h) to secondary (48 h) growth phase in WT and the kdmBΔ mutant. (D) An overview of GO enrichment analysis for genes whose response to primary-to-secondary growth phase transition differs between WT and the kdmBΔ mutant. (E) Functions of the KERS complex components on fungal development and light responses. Growth of KERS mutants (deletions or TetON strains) on GMM plates at 37 ºC for 5 days both under continuous white light and dark conditions. (F, G) Quantification of asexual (conidia) and sexual (fruiting bodies) development from (E). Significance as calculated by Student's t-test is indicated as asterisks: P < 0.05 (*), P < 0.01 (**), P < 0.001 and P < 0.0001 (****). Asexual sporulation and sexual development were calculated relative to WT-light conidiation and WT-dark sexual development, respectively. (H) Heatmaps displaying binding locations, signals and intensities of KdmB^HA, EcoA^HA, RpdA^HA and SntB^HA over regulatory genes of the ST BGC and development. (I) A heatmap showing expression level of genes associated with positive regulation of SM biosynthesis in the KERS mutants at primary and secondary growth phases. Expression is presented as log₂ fold change with respect to values in WT. Asterisks (*) indicate non-detectable Pol II ChIPseq signal in both WT and mutant strains, and hence fold change cannot be calculated. (J) A heatmap showing expression (as Pol II ChIPseq) levels of the ST BGC and border genes at secondary growth phase. (K) Validation of a few key genes from (H) by Northern blotting. Strains were grown for 48 h (Secondary Metabolism) in GMM. 10 μg of total RNA was used per lane. rRNA signals and gpdA served as loading control.

Figure 5.

Chromatin level coordination of fungal development and mycotoxin production by the KERS complex. (A) Molecular mechanisms of KERS assembly. Two heterodimers, KdmB-EcoA, SntB-RpdA establish the KERS core complex. EcoA acetylates the cohesin SudA at two residues. KdmB protects EcoA from proteasomal degradation by preventing its phosphorylation and subsequent ubiquitination presumably by the ring finger E3 ligase SntB. (B) The KERS binds to more than 1,600 promoters including many regulatory target genes and activates the transcription of these genes. These regulatory genes, particularly, velvet complex genes whose expression depends on the KERS complex, control production of mycotoxin and its coordination with fungal development.