1. Introduction

Aspergillus flavus is one of the common asexual species, a saprophytic fungus and the second largest pathogenic fungus after Aspergillus fumigatus, widely distributed in soil, air, water, plants, and agricultural products in nature [1]. Aflatoxins produced by A. flavus has strong toxicity, and is extremely harmful to human society. Animal and human health can be negatively affected by aflatoxins, which are carcinogenic, teratogenic, and mutagenic [2]. Among aflatoxins, AFB1 is the most frequently occurring and the most toxic and carcinogenic, which is converted to AFB1-8 and 9-epoxide in the liver and formed adducts with the guanine base of DNA and thus results in acute and chronic diseases in both human and household animals [3]. According to the Food and Agriculture Organization of the United Nations, 25% of the food crops in the world are contaminated with aflatoxins [4], which are often detected in grains, nuts, and spices [5, 6]. It is urgent to control the contamination of A. flavus and its main mycotoxin, AFB1. In recent decades, the biosynthetic pathway of aflatoxins was investigated in detail benefit from the sequence of A. flavus genome [7]. This pathway consists of a complex set of enzymatic reactions [8–11]. In general, these enzymes are encoded by clusters of genes, which are regulated by cluster-specific genes: aflR and aflS [12, 13]. The initial stage of aflatoxins biosynthesis is catalyzed by polyketide synthase (PKSA) to form the polyketone backbone [14]. Aflatoxins synthesis also regulated by environmental stimuli including pH, light, nutrient sources, and oxidative stress response, which may result in the modulation of the expression of genes involved in toxin production [15–17].

Besides the biosynthetic pathway and its internal gene regulation, protein post-translational modifications (PTMs), an important mean of epigenetics, represent an important role in the regulation of aflatoxins synthesis, including 2-hydroxyisobutyrylation, succinylation, acetylation, and methylation [18–24], in which Snt2 (also called sntB, an epigenetic reader) is deeply involved. Despite advancements in the field, our understanding of the molecular mechanisms of aflatoxin production in A. flavus is still fragmentary.

The epigenetic reader encoded by sntB in A. nidulans was identified as a transcriptional regulator of the sterigmatocystin biosynthetic gene cluster and deletion of sntB gene in A. flavus results in loss of aflatoxin production [25], increasing global levels of H3K9K14 acetylation and impairing several developmental processes [20]. The homolog gene in yeast, SNT2 coordinates the transcriptional response to hydrogen peroxide stress [26, 27]. In Penicillium expansum, SntB regulated the development, patulin and citrinin production, and virulence on apples [28]. In A. nidulans, SntB combined with a H3K4 histone demethylase KdmB, a cohesin acetyltransferase (EcoA), and a histone deacetylase (RpdA) to form a chromatin binding complex and bound to regulatory genes and coordinated fungal development with mycotoxin synthesis [29]. sntB also regulated the virulence in Fusarium oxysporum, and respiration in F. oxysporum and Neurospora crassa [30, 31].

However, the specific regulatory mechanism of sntB in A. flavus remains unclear. In this study, we identified the regulatory network of sntB by Chromatin immunoprecipitation sequencing (ChIP-seq) and RNA sequencing (RNA-seq), which shed light on its impact on fungal biology.

Experimental Procedures

A. flavus strains, media, and culture conditions

A. flavus Δku70 ΔpyrG was used as a host strain for genetic manipulations. All strains used in this study are listed in the Table 1. potato dextrose agar (PDA, 39 g/L, BD, Difco, Franklin, NJ, USA), complete medium (CM, 6 g/L tryptone, 6 g/L yeast extract, 10 g/L glucose), and potato dextrose (PDB, 24 g/L, BD, Difco, Franklin, NJ, USA) were used for mycelial growth and sporulation determination, sclerotia production, and mycotoxin production analysis, respectively. All experiments were technically repeated three times and biologically repeated three times.

Table 1.

The construction of mutant strains

All mutant strains, including snt2 and catC gene knock-out strain (Δsnt2 and ΔcatC), the complementation strain for the ΔsntB strain (Com-sntB), and HA tag fused to snt2 strain (snt2-HA), were constructed following the protocol of homologous recombination [32] and the detail protocol was as described in our previous study [33]. The related primers were listed in Table 2. The constructed strains were confirmed by t diagnostic PCRs and by southern bolt [34]. The construction of snt2-HA was further determined by western blot with anti-HA antibody as descripted previously [33].

Table 2.

Phenotypic analysis and aflatoxin analysis

The spores (10⁷ conidia/mL) of wild type (WT), ΔsntB, and Com-sntB strains were used. The details of experiment were according to our previous study [35]. Hyphal septum was stained according to the descripted method [36]. Each fungal strain was evaluated on four plates, and each experiment was repeated three times.

Fungal colonization on crop kernels

According to our previous experimental protocol [35], the colonizing ability of WT, ΔsntB, and Com-sntB fungal strains on peanut and corn kernels was analyzed. The crop kernels were disinfected with 0.05% sodium hypochlorite and soaked for 30 min in a solution containing 10⁵ conidia/mL fungal spores. Afterwards, the seeds were placed in a petri dish and cultured at 29°C for 6 d. Finally, the number of conidia was calculated and AFB1 product was analyzed by TLC.

Animal invasion assay

The animal invasion assay using silkworms was according to previous study [32, 37]. Silkworms (Bombyx mori) were randomly separated into four groups (10 larvae/group) when silkworm larva reaches about 1 g in weight. Each silkworm was injected with 5 µL saline, or 5 µL conidial suspension (10⁶ spores/mL) from WT, ΔsntB, and Com-sntB strains. The survival rate of silkworms was calculated. Dead silkworms were transferred into fresh 9 cm Petri dishes and inoculated for 5 d in the dark. The conidia number and AFB1 production from each group were measured.

RNA-seq analysis

To reveal the potential complex regulatory network of the sntB, RNA-seq analysis was carried out on the WT and ΔsntB strains. RNA-seq analysis was carried out by APPLIED PROTEIN TECHNOLOY, Shanghai (www.aptbiotech.com) [38]. Data processing was according pervious study [39]. Differentially expressed gene (DEGs) were assigned as genes with |log2FoldChange|>1 and adjusted P-adj< 0.05. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were used to analyze the functions of DEGs.

ChIP-seq and data analysis

ChIP-seq analysis was carried out on the WT and sntB-HA strains. The conidia (10⁴/ml) for each strain were inoculated in 100 mL PDB shaking at 180 rpm under 29°C for 72h, and subjected to ChIP-seq analysis by Wuhan IGENEBOOK Biotechnology Co.,Ltd (http://www.igenebook.com). ChIP experiment was according previous study [32]. Raw sequencing with low-quality reads discarded and reads contaminated with adaptor sequences trimmed were filtered by Trimmomatic (v0.36) [40]. The clean reads were mapped to the reference genome of A. flavus by Burrows-Wheeler Alignment tool (BWA, version 0.7.15) [41]. MACS2 (v2.1.1) and Bedtools (v2.25.0) were used for peak calling and peak annotation, respectively. Differential binding peaks were identified by Fisher’s test with q value <0.05. Genes less than 2000 bp away were associated to the corresponding peak. GO and KEGG enrichment analyses of annotated genes were implemented in EasyGO [42] and KOBAS (v2.1.1) [43], with a corrected P-value cutoff of 0.05.

Quantitative RT-PCR (qRT-PCR) analysis

The fungal spores (10⁶/mL) were cultured in PDB medium for 48 h, and then mycelium was ground into powder with liquid nitrogen. Total RNA was prepared by TRIpure total RNA Extraction Reagent (Bestek, China) according to the protocol used by Zhang [44]. qRT-PCR was performed according to previously described [45], and the primers were shown in Table 3.

Table 3.

Oxidative stress assays

To evaluate the role of sntB in fungal resistance to oxidative stress, a series concentration (0, 0.12, 0.24, and 0.36 mM) of menadione sodium bisulfite (MSB) were added to the medium. 10⁶ fungal spores for each strain were inoculated on the medium and cultured in dark at 37℃. The diameters of colonies were measured 3 d after inoculation and the inhibition rate was calculated as previously descripted [33]. The AFB1 product was analyzed by TLC after the strains were cultured in YES medium in dark at 29℃ for 7 d.

Statistical analysis

All data in this study were expressed as mean ± standard deviation. The statistical analysis was performed using the software GraphPad Prism8 (GraphPad Software, La Jolla, CA, USA). The difference was considered to be statistically significant when P<0.05.

Results

The phenotype of SntB in A. flavus

The role of SntB in A. flavus has been previously characterized by analyzing both ΔsntB and overexpression sntB genetic mutants [20]. To further investigate the intrinsic mechanism of this regulator on the development and aflatoxin biosynthesis in A. flavus, the sntB deletion strain (ΔsntB) and the complementary strain (Com-sntB) were constructed by the method of homologous recombination and verified by diagnostic PCR and southern blotting (Figure S1). The expression levels of sntB in WT, ΔsntB, and Com-sntB strains was further detected by qRT-PCR and the result showed that the expression of sntB was absent in the gene-deletion strains, and it fully recovered in the Com-sntB (Figure S1), which reflected that the sntB and Com-sntB had been constructed, and could be used in the subsequent experiments of this study. The phenotype analysis of this study revealed that the deletion of sntB gene significantly inhibited the growth of mycelium, hyphae morphology, the length of fungal cell (between two adjacent septa), the number of conidiation, sclerotium formation, and aflatoxin biosynthesis, while the above phenotypes of both development and mycotoxin bio-synthesis were fully recovered in the Com-sntB strain (Figure 1). To reveal the signaling pathways of sntB in conidiation, sclerotium formation, and aflatoxin biosynthesis, qRT-PCR analysis was performed to assess the expression levels of sporulation related transcriptional factor genes, steA, WetA, fluG, and veA, sclerotia formation related transcriptional factor genes, nsdC, nsdD, and sclR [46, 47], and the AFs synthesis gene cluster structural genes aflC, aflR, and aflP, and the main regulatory gene aflR and aflS. As shown in Figure S2, the relative expression levels of these genes were significantly lower in the ΔsntB strain compared to that of the WT strain, and recovered in the Com-sntB strain. These results indicated that SntB regulates the conidiation, sclerotium formation, and aflatoxin biosynthesis by the canonical signaling pathways mediated by these regulators.

Figure 1.

SntB plays important role in virulence of A. flavus to both plant and animal hosts

In order to explore the effect of SntB on the fungal colonization ability, peanut and maize kernels were infected with spore solution of each fungal strain. Compared with WT, the conidiation yield of ΔsntB on the infected host was significantly reduced (P<0.001) and no AFB1 could be detected on the hosts infected by ΔsntB, while in the Com-sntB strain, the capacity to produce conidia and AFB1 on both crop kernels recovered (Figure 2A-2C and Figure S3A). The role of sntB in fungal virulence to animals was also investigated. As Figure 2D-2E shown, the survival rate of silkworms injected by spores of ΔsntB strain was significantly higher than that from WT infected larvae. There were less fungal mycelium, conidia, and AFB1 production on the dead silkworms injected by ΔsntB compared to the silkworms from the WT injection groups, but when the gene was reintroduced (i.e. the Com-sntB group), similar to what found in the WT group, the survival rate of silkworms obviously dropped and more fungal mycelium, conidia, and AFB1 produced on the dead silkworms (Figure 2F-2H). All the above results revealed that SntB plays an essential role in virulence of A. flavus.

Figure 2.

The capacity of fungal infection is closely related to secreted hydrolases, such as amylase, lipase, protease, etc. In order to explore the effect of SntB on the activity of hydrolases, the activities of amylase in the above each fungal strain were further determined. The results showed that the colonies of ΔsntB produced almost no degradation transparent circle after adding iodine solution compared with that of WT and Com-sntB, which indicated that the activity of α-amylase in ΔsntB strain was significantly reduced (P<0.001) (Figure S3B and S3C). This suggests that SntB plays an important role in the fungal pathogenicity by changing the hydrolases activity of A. flavus.

SntB chords global gene expression

To explore the downstream signaling pathways regulated by SntB, samples with three biological replicates of WT and ΔsntB strains were submitted for RNA-Seq. In the assay, the bases score Q30 was more than 93.19% (Table S1) and mapping ratio was from 95.17% to 95.80% (Table S2). To further confirm the quality of RNA-Seq, PCA (Principal Components Analysis) and Pearson correlation analysis were performed. Correlation analysis revealed that the samples were clustered by groups (Figure 3A). A plot of PC1 (47.70%) and PC2 (23.20%) scores showed a clear separation between the groups (Figure 3B). A total of 1,446 and 1,034 genes were significantly up– and down-regulated, respectively, in the ΔsntB compared to the WT strain (Figure 3C and Supplementary Table S3). GO enrichment analysis identified 93 enriched GO terms (P<0.05) (Figure 3D and Supplementary Tables S4). In the biological process category, the most enriched terms were “oxidation-reduction process (GO: 0055114)”. In the molecular function category, “catalytic activity (GO: 0003824),” “oxidoreductase activity (GO: 0016491),” and “cofactor binding (GO: 0048037)” were the most significantly enriched terms. Whereas terms associated with “Set3 complex (GO: 0034967)”, “mitochondrial crista junction (GO: 0044284)” and “extracellular region (GO: 0005576)” were significantly enriched in the cellular component category. Additionally, all the DEGs were mapped according to the KEGG database, and 42 significantly enriched pathways were identified (P<0.05) (Supplementary Table S5). Among them, “metabolic pathways (ko01100),” “aflatoxin biosynthesis (ko00254),” and “microbial metabolism in diverse environments (ko01120)” were the most significantly enriched (Figure 3E and Supplementary Table S5).

Figure 3.

Characterization the binding regions of SntB by chromatin immunoprecipitation sequencing (ChIP-seq)

To characterize the chromatin regions targeted by SntB, ChIP-seq studies were carried out with both HA tag fused sntB strain (sntB-HA) and WT strain. The sntB-HA strain was constructed by homologous recombination through fused HA to the 3’ end of sntB (Figure 4A). In the ChIP-seq assay, more than 94.66% of bases score Q30 and above in each sample (Table S6), and reaching 52.50% to 94.48% of mapping ratio (Table S7). The principal component analysis (PCA) (Figure S4A) and heat map (Figure S4B) reflected that the quality of samples was competent for subsequent ChIP-seq analysis. There were 1,510 up-enriched differently accumulated peaks (DAPs) in sntB-HA fungal strain compared to the WT strain, which were distributed on the whole A. flavus genome (Figure 4B, and Table S8). Most of the up-enriched peaks were located in the promoter (82.85%) region (Figure 4C). The genes of the DAPs were further subjected to GO and KEGG analysis. The most strikingly enriched GO terms in the biological process category were “cell communication (GO:0007154)”, “response to stimulus (GO:0050896)”, and “response to external stimulus (GO:0009605)”. Whereas terms associated with “DNA-binding transcription factor activity (GO:0003700)”, “DNA-binding transcription factor activity, RNA polymerase II-specific (GO:0000981)”, and “sequence-specific DNA binding (GO:0043565)” were the most significantly enriched molecular function category (Figure 4D and Table S9). And these genes were mostly enriched in “Methane metabolism” and “MAPK signaling pathway-yeast” pathways (P value <0.05) (Figure 4E and Table S10).

Figure 4.

Integration of the results of ChIP-seq and RNA-seq assays

After the overlapping the results from both different sequence methods (ChIP-seq and RNA-seq), 238 DEGs were found (Figure 5A). According to the GO annotation, these DEGs were significantly enriched in 8 GO terms, including “cellular response to reactive oxygen species (GO:0034614)”, “reactive oxygen species metabolic process (GO:0072593)”, and “cellular response to oxygen-containing compound (GO:1901701)” (Figure 5B). It was further noted that the DEGs were significantly assigned to “Carbon metabolism (afv01200)”, “Peroxisome (afv04146)”, and “Glyoxylate and dicarboxylate metabolism (afv00630)” KEGG pathways (Figure 5C and Table S11-S12). These results revealed that SNT2 is essential for A. flavus to maintain the homeostasis of intracellular reactive oxygen species. Studies had shown that SNT2 could response to oxidative stress in yeast [27] and Magnaporthe oryzae [48]. As Figure 5D and Figure 5E showed, due to the deletion of the sntB gene, ΔsntB exhibited a severe MSB sensitivity phenotype compared to that of the wild-type (WT) strain, and the phenotype recovered in the complementary strain (Com-ΔsntB). The results showed that the inhibition rate of oxidant MSB to ΔsntB would be significantly enhanced with the increase of MSB concentration. These results showed that SntB deeply participates in the regulation of oxidative stress pathway. As the most abundant peroxisomal enzyme, catalases catalyze decomposition of hydrogen peroxide [49]. As the most abundant peroxisomal enzyme, catalases catalyze decomposition of hydrogen peroxide [49]. To further study the mechanism of SntB mediated oxidative response of A. flavus, the catC (encode a catalase) gene was selected based on the above integration results. The relative expression levels of catC in wild-type and ΔsntB strains under MSB treatment were measured. As Figure 5F shown, the expression level of catC was significantly higher in ΔsntB strain than in WT strain, it suggested that to compensate the absence of sntB, catC is up-regulated to respond the higher intracellular oxidative level. However, under the stress of oxidant MSB, the deletion of sntB obvious suppressed the expression level of catC compared to that of WT strain, which reflected that the absence of sntB significantly impaired the capacity of catC to further respond to extra external oxidative stress. These results revealed that SntB is deeply involved in CatC mediated oxidative response in A. flavus.

Figure 5.

CatC is important for A. flavus response to oxidative stress

The functions of catC gene in A. flavus were further explored by knockout of the catC (Figure S5). As shown in Figure 6A-6C, the diameter of ΔcatC strain was significantly smaller than that of WT, and the conidia number in the ΔcatC strain decreased significantly compared to that of WT. The sclerotia production of ΔcatC strain was also significantly less than that of WT strain (Figure 6D and 6E). In view of Catc is involved in the oxidative response pathways of A. flavus (Figure 4), both ΔcatC and WT strains were treated by a serial concentration of MSB, and the results showed that the inhibition rates of MSB to ΔcatC strain was significantly lower than that of WT group (Figure 6F and 6G). Catalase is a major peroxisome protein and plays a critical role in removing peroxisome-generated reactive oxygen species. This result echoed that deletion of sntB increased intracellular oxidative level and the inhibition rate of MSB, and up-regulated the expression of catC (Figure 5D-5F). The role of CatC in the bio-synthesis of AFB1 was also assessed (Figure 6H). The results showed that a relatively large amount of AFB1 was produced by the ΔcatC strain compared to the WT. But when under the stress of MSB, AFB1 yield of the WT strain was significantly more than that of ΔcatC strain. All the above results revealed that the catC plays an important role in SntB mediated regulation pathway on fungal morphogenesis, oxidative stress responding and AFB1 production.

Figure 6.

SntB regulates fungal virulence through peroxidase mediated lipolysis

Biogenesis of peroxisomes was reported to promote lipid hydrolysis, produce the production of glycerol, and further change fungal pathogenicity [50]. Due to deeply involved in oxidative response of A. flavus, we wondered if SntB also takes part in the regulation of the production of lipid and glycerol. As shown in Table S13, one gene (G4B84_008359) in lipase activity GO term was significantly down-regulated in ΔsntB strain, which encoded a secretory lipase belonged to the virulence factors reported in Pseudomonas aeruginosa [51]. The lipase activity was assayed by examining the ability to cleave glycerol tributyrate substrate [52]. The results showed that the colony diameter of ΔsntB strain on PDA medium with tributyrin were significantly smaller than that of the control, and the colony diameters of WT and Com-sntB strains on PDA medium were obviously bigger than those on 0.3% tributyrin PDA medium. The relative inhibition rate of tributyrin to the colony growth of ΔsntB strains was significantly higher than that of WT and Com-sntB strain (Figure S3D and S3E). Our previous study revealed that H3 lysine 36 trimethylation (H3K36me3) modification on the chromatin region of the sntB is regulated by AshA and SetB [32]. H3K36me3 usually promote gene transcription [53, 54]. Our study revealed the potential machinery associated with SntB mediated regulation on fungal morphogenesis, mycotoxin anabolism and fungal virulence, which lurks the axle of from SntB to fungal virulence and mycotoxin bio-synthesis through lipid catabolism (i.e. H3K36me3 modification-SntB-Peroxisomes-Lipid hydrolysis-fungal virulence and mycotoxin bio-synthesis).

Discussion

SntB is a conserved regulator in many species, including A. nidulans [25], Saccharomyces cerevisiae [27], Schizzosaccharomyces pombe [55], Fusarium oxysporum [30], Neurospora crassa [31], and A. flavus [20, 25]. SntB can regulate the production of uncharacterized secondary metabolites, including aspergillicin A1 and aspergillicin F2 [56]. MoSnt2 protein was required for regulation of infection-associated autophagy in Magnaporthe oryzae [48]. In A. flavus, the functions of sntB gene were analyzed by both ΔsntB and overexpression sntB genetic mutants[20]. SntB deletion impaired several developmental processes, such as sclerotia formation and heterokaryon compatibility, secondary metabolite synthesis, and ability to colonize host seeds, which were consistent with our results (Figure 1 and 2). Unlike, a complementation strain was constructed in this study which further clarified and confirmed the function of sntB gene. In this study, the potential mechanism under these effects were further analyzed by detection the related transcriptional factor genes of sporulation (steA, WetA, fluG, and veA), sclerotia formation (nsdC, nsdD, and sclR), and the AFs synthesis related genes aflC, aflD, aflO, aflP, and aflR (Figure S2). In the RNA-seq data, we also found some DEGs related to AFs synthesis (aflB, aflE, aflH, aflK, aflN, aflO, aflP, aflQ, aflR, aflS, aflV, and aflW) (Figure S6). And all these genes were down-regulated, which was consistent with that the AFs production in ΔsntB was significantly decreased compared to WT and Com-sntB (Figure 1F). These results inferred that SntB regulated the morphogenesis and the production of A. flavus through the above canonical signal pathways.

For the process of A. flavus invading hosts, in view of it is a notorious pathogen for plant and animal, we established both crop and insect models, especially silkworm represented animal mode profoundly revealed the critical role of SntB in fungal virulence (Figure 2). The results of crop kernel models showed that the number of spores of ΔsntB on kernels of both peanut and maize was dramatically lower than that of strains WT and Com-sntB (Figure 2A and 2B) and almost no AFB1 was detected on maize and peanut kernels inoculated with ΔsntB, while plenty of AFB1 were detected from the kernels infected by WT fungal strain, and AFB1 biosynthesis capacity of Com-sntB strains recovered compared to the ΔsntB and WT fungal strains (Figure 2C). These results were corroborated by previous study [20]. It was also found in this study that the survival rate of silkworms injected by spores of ΔsntB strain was significantly higher than the silkworms injected with spores from WT and Com-sntB fungal strains (Figure 2D and 2E). What’s more, we also assayed the effect of SntB on the activity amylase, which was closely related to the capacity of fungal infection [57]. As shown in Figure S3C, after adding iodine solution, the ΔsntB strain almost did not produce a degradation transparent ring compared with wild-type and complementary strains, indicating the amylase activity of the sntB gene knockout strain was significantly decreased (P<0.001). Our results comprehensively reveal the important function of SntB in the growth, development, secondary metabolite synthesis, and virulence of A. flavus.

SntB was reported as an important epigenetic reader. In A. flavus, SntB was reported to regulate global histone modifications (acetylation and methylation) and interact with EcoA and RpdA to form a conserved chromatin regulatory complex [20, 29]. Loss of sntB in Magnaporthe orzyae also led to an increase in H3 acetylation [48]. In our RNA-seq data, we also found a set domain containing histone-lysine N-methyltransferase (Ash1, G4B84_009862) was down-regulated in ΔsntB strain compared to WT (Table S3), which was reported to regulate mycotoxin metabolism and virulence via H3K36 methylation in A. flavus [32]. Besides, SntB is reported to be a transcriptional regulator in A. nidulans [25] and Fusarium oxysporum [30]. So, we use RNA-seq and ChIP-seq to study the transcriptional response of sntB in A. flavus. By integration analysis the results of ChIP-seq and RNA-seq assays, we found that the enriched GO terms and KEGG pathways of the DEGs were related to oxidative response (Figure 5A-5C). These results reflected that Snt2 plays an important role in fungal responding to oxidative stress, which is consistent with previous reports that Snt2 could response to oxidative stress in yeast [27], Fusarium oxysporum [30], and Magnaporthe oryzae [48].

As a harmful by-product of oxidative metabolism, reactive oxygen species (ROS) is unavoidable and essential for fungus development [58, 59]. ROS has also been shown to be required for aflatoxin production [60–62]. Several oxidative stress-responsive transcription factors have been identified as regulating aflatoxin production, including AtfB, AP-1, and VeA [61, 63–65]. Previous studies shown that Snt2 protein coordinates the transcriptional response to hydrogen peroxide-mediated oxidative stress in the yeast [26, 27] and is involved in fungal respiration and reactive oxidative stress in Fusarium oxysporum and Neurospora crassa [30, 31]. Several GO terms (“cellular response to reactive oxygen species”, “reactive oxygen species metabolic process”, and “cellular response to oxygen-containing compound”) and KEGG pathways (Peroxisome) were enriched by the DEGs screened out by integration of ChIP-seq and RNA-seq data in this study (Figure 5). This is the first time to show that the SntB in A. flavus is important in oxidative stress response, through which SntB participates the regulating of aflatoxin bio-synthesis and fungal development.

Fungal defense against ROS is mediated by superoxide dismutases (SOD), catalases (CAT), and glutathione peroxidases (GPX). The effect of MSB on cellular growth and antioxidant enzyme induction in A. flavus was previously explored [66–68]. Once in the cell, menadione may release superoxide anion [69], which was scavenged by superoxide dismutases (SOD) and transformed into hydrogen peroxide, or react with nitric oxide to form peroxynitrite [70]. This study found that after knock out sntB gene, the strain growth was significantly inhibited by MSB (Figure 5F and 5G). Some genes encoded superoxide dismutases and catalases were reported to associated with AF/ST synthesis [71], including mnSOD, sod1, sod2, catA, catB, and hyr1. In our RNA-seq data, 7 related genes were screened out (Table S14), among which, bZIP transcription factor Atf21 (G4B84_008675), catlase C (G4B84_009242), catlase A (G4B84_010740), superoxide dismutases sod2 (G4B84_003204), peroxisomal membrane protein PmpP24 (G4B84_001452) were up-regulated, while catlase B (G4B84_008381) and superoxide dismutase sod1 (G4B84_009129) were down-regulated in ΔsntB strain, respectively.

Some studies reported the correlation between ROS formation, aflatoxin production, and antioxidant enzyme activation. Aflatoxin B1 biosynthesis and the activity of total superoxide dismutase were effectively inhibited by cinnamaldehyde, whereas the activities of catalase and glutathione peroxidase were opposite [72]. The expression of catA, cat2 and sod1, and CAT enzymatic activity were opposite correlated to AFB1 biosynthesis under AFs inhibitor piperine treatment [73]. Deletion the gene of sod (GenBank accession no: CA747446) reduced AFs production [74], which was most similar to sod2 (G4B84_003204). The mitochondria-specific sod and the genes aflA, aflM, and aflP belonging to the AFs gene cluster were reported to co-regulated [75]. Ethanol can inhibit fungal growth and AFB₁ production in A. flavus and enhanced levels of anti-oxidant enzymatic genes including Cat, Cat1, Cat2, CatA, and Cu, Zn superoxide dismutase gene sod1. All these reports indicated that the expression of anti-oxidant enzymatic genes was opposite correlated to AFB1 biosynthesis.

In our study, 7 genes related to oxidative response were obviously differentially expressed in transcriptome data (Table S14). Among these DEGs, 5 out of 7 genes were up-regulated in ΔsntB strain. Based on the AFs production in ΔsntB was significantly decreased compared to WT and Com-sntB (Figure 1F), the most up-regulated gene in ΔsntB strain, catC (G4B84_000242), was selected for further analysis. We found that the deletion of sntB significantly up-regulated the catC gene, however, the expression of the catC gene was almost undetectable under MSB treatment (Figure 5F). Results also showed that the inhibition rates of MSB to ΔcatC strain was significantly lower than that of WT group and AFB1 yield of the ΔcatC strain was significantly decreased than that of WT strain under the stress of MSB (Figure 6F-6H). These results indicated that SntB is profoundly involved in the catC mediated oxidative response.

Peroxisomes are intimately associated with the metabolism of lipid droplets [76] and the histone lysine methyltransferase ASH1 promotes peroxisome biogenesis, inhibits lipolysis, and further affects pathogenesis of Metarhizium robertsii [50]. Set2 histone methyltransferase family in A. flavus, AshA and SetB, were found to regulate mycotoxin metabolism and virulence via H3K36me3, including the chromatin region of the sntB [32]. By ChIP-seq and RNA-seq, SntB was found to essential for A. flavus to maintain the homeostasis of intracellular reactive oxygen species (Figure 5A) and several anti-oxidant enzymes were up-regulated in ΔsntB strain (Table S14). In addition, we also found only one down-regulated DEG (G4B84_008359) in lipase activity GO term in our RNA-seq data (Table S13), which encodes a secretory lipase and belongs to the virulence factors reported in Pseudomonas aeruginosa [51]. These results suggested that SntB plays a pivotal role in regulating peroxisome biogenesis to promote lipolysis involving in fungal pathogenesis.

Overall, we explored and clarified the bio-function of the SntB and found that SntB responses to oxidative stress through related oxidoreductase represented by CatC in A. flavus (Figure 7). Our study revealed the potential machinery associated with SntB mediated regulation on fungal morphogenesis, mycotoxin anabolism and fungal virulence, which lurks the axle of from SntB to fungal virulence and mycotoxin bio-synthesis (i.e. SntB-Peroxisomes-Lipid hydrolysis-fungal virulence and mycotoxin bio-synthesis). The work of this study provided a novel perspective for developing new prevention and control strategies against pathogenic fungi.

Figure 7.

Compliance with ethical requirements

This article does not contain any studies with human or animal subjects.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability

All data needed to evaluate the conclusions are present in the paper and/or the Supporting Information. Raw data of the ChIP and RNA-seq were submitted to GSE247683.

Funding

This work was funded by the grants of the National Natural Science Foundation of China (No. 32070140), the Nature Science Foundation of Fujian Province (No. 2021J02026), and the State Key Laboratory of Pathogen and Biosecurity (Academy of Military Medical Science) (SKLPBS2125).

Acknowledgements

We especially thank Professor Jun Yuan, Xiuna Wang, Yu Wang and Xinyi Nie for their support in instrument maintenance and reagent ordering.

Supplementary figure

Figure S1.

Figure S2.

Figure S3.

Figure S4.

Figure S5.

Figure S6.

Supplementary table captions

Table S1. Sequencing data statistics in RNA-seq.

Table S2. Alignment results of each sample in RNA-seq.

Table S3. The information of DEGs in transcriptome data.

Table S4. GO enriched in the DEGs in transcriptome data.

Table S5. KEGG pathways enriched of the DEGs in transcriptome data.

Table S6. Sequencing data statistics in ChIP-seq.

Table S7. Alignment results of each sample in ChIP-seq.

Table S8. The information of up regulated peak in ChIP-seq data.

Table S9. GO enriched in the up regulated genes in ChIP-seq data.

Table S10. KEGG enriched in the up regulated genes in ChIP-seq data.

Table S11. GO enriched in the 238 common DEGs in RNA-seq and ChIP-seq data.

Table S12. KEGG enriched in the 238 common DEGs in RNA-seq and ChIP-seq data.

Table S13. The information of DEGs related to lipase activity in transcriptome data.

Table S14. The information of DEGs related to oxidative response in transcriptome data.