Histone 2-Hydroxyisobutyryltransferase Encoded by Afngg1 Is Involved in Pathogenicity and Aflatoxin Biosynthesis in Aspergillus flavus

Aflatoxin, a carcinogenic secondary metabolite produced by Aspergillus flavus, is a significant threat to human health and agricultural production. Histone 2-hydroxyisobutyrylation is a novel post-translational modification that regulates various biological processes, including secondary metabolism. In this study, we identified the novel histone 2-hydroxyisobutyryltransferase Afngg1 in A. flavus, and explored its role in cell growth, development and aflatoxin biosynthesis. Afngg1 gene deletion markedly decreased lysine 2-hydroxyisobutyrylation modification of histones H4K5 and H4K8 compared with the control strain. Additionally, Afngg1 deletion inhibited mycelial growth of A. flavus, and the number of conidia and hydrophobicity were significantly decreased. Notably, aflatoxin B1 biosynthesis and sclerotia production were completely inhibited in the ΔAfngg1 strain. Furthermore, the pathogenicity of the ΔAfngg1 strain infecting peanut and corn grains was also diminished, including reduced spore production and aflatoxin biosynthesis compared with A. flavus control and Afngg1 complementation strains. Transcriptome analysis showed that, compared with control strains, differentially expressed genes in ΔAfngg1 were mainly involved in chromatin remodelling, cell development, secondary metabolism and oxidative stress. These results suggest that Afngg1 is involved in histone 2-hydroxyisobutyrylation and chromatin modification, and thus affects cell development and aflatoxin biosynthesis in A. flavus. Our results lay a foundation for in-depth research on the 2-hydroxyisobutyrylation modification in A. flavus, and may provide a novel target for aflatoxin contamination prevention.


Introduction
Aspergillus flavus, one of the most abundant and widely distributed fungi on earth, is often found in pre-or post-harvest crops, and it causes serious economic losses [1,2]. It can produce teratogenic and carcinogenic aflatoxins that pose a serious safety hazard to both humans and animals [3]. It is estimated that aflatoxin contamination costs the US corn industry nearly USD 1.68 billion annually [4]. Aflatoxin causes 28% of hepatocellular carcinoma worldwide [5]. Therefore, an in-depth understanding of aflatoxin biosynthetic pathways and their regulatory mechanisms is urgently required to develop effective strategies to control aflatoxin contamination.
Histone post-translational modifications (HPTMs), including acetylation, phosphorylation, ubiquitination and 2-hydroxyisobutyrylation, play a crucial role in many cellular processes in eukaryotes, including secondary metabolite biosynthesis and chro-matin regulation [6]. Among them, 2-hydroxyisobutyrylation is a recently discovered post-translational modification that plays an important regulatory role in gene transcription, metabolism and enzyme activity [7]. Previous studies have shown that lysine de-2hydroxyisobutyrylase CobB can regulate the catalytic activity of enolase by decreasing modification at the K343 site, thereby regulating glycolysis and cell growth in bacteria [8].
In Ustilaginoidea virens, mutation of the 2-hydroxyisobutyrylation modification site of mitogen-activated protein kinase UvSlt2 decreased UvSlt2 enzyme activity and significantly reduced the development and virulence of plant pathogenic fungi [9]. Significantly, 2-hydroxyisobutyrylation modification is involved in the pathogenicity of Botrytis cinerea, Candida albicans, Fusarium oxysporum and Toxoplasma gondii [7,[10][11][12]. Our previous studies have shown that 2-hydroxyisobutyrylation modification regulates mycelial growth and development of A. flavus, as well as aflatoxin biosynthesis [13]. These results indicate that lysine 2-hydroxyisobutyrylation modification plays a key role in A. flavus, but the enzymes involved in histone 2-hydroxyisobutyrylation modification have not been reported in A. flavus. Herein, we identified histone 2-hydroxyisobutyryltransferase Afngg1 in A. flavus, and explored its underlying gene regulatory mechanisms using RNA sequencing (RNA-seq)-based transcriptional analysis.
In this study, Afngg1, a homolog of Candida albicans histone acetyltransferase Ngg1, was found to be associated with histone 2-hydroxyisobutyrylation modification. Additionally, the functions of Afngg1 in cell development and aflatoxin B 1 (AFB 1 ) production were profiled, and the underlying regulatory mechanisms were investigated through transcriptome analysis. The findings suggest that Afngg1 is a potential target for the effective control of aflatoxin contamination.

Characterisation of A. Flavus Histone 2-Hydroxyisobutyryltransferase Afngg1
In order to characterise homologs of Afngg1 in A. flavus, we obtained homologous sequences by BLAST searching the NCBI database, and constructed phylogenetic trees to analyse their evolutionary relationships. The results revealed that Afngg1 in A. flavus is closely related to homologs in other fungi, and most similar to homolog in A. oryzae RIB40 ( Figure 1A). The analysis also indicated that it was homologous to Candida albicans histone acetyltransferase Ngg1. The domains in Afngg1 were further analysed via the NCBI database and visualised using DOG2.0. A conserved Ada3 (histone acetyltransferase subunit 3) domain was found in homologs of Afngg1 proteins in nine species ( Figure 1B). Additionally, Afngg1 was localised in the nucleus of A. flavus ( Figure 1C).
Given the similarity to Ngg1 in C. albicans, we speculated that Afngg1 might also be involved in histone modification. To test this hypothesis, the effects of Afngg1 on the level of acetylation of histone H3 and 2-hydroxyisobutyrylation modification of histone H4 were investigated by collecting mycelia of A. flavus control and ∆Afngg1 strains, and proteins were extracted for Western blotting detection using corresponding antibodies. The results indicated that the acetylation modification level of H3K14 in strain ∆Afngg1 was decreased compared with the A. flavus control strain, and notably, acetylation modification of H3K9 was completely absent ( Figure S1). Unexpectedly, our results also showed that 2-hydroxyisobutyrylation modification levels of H4K8 and H4K5 were significantly lower than in the A. flavus control strain ( Figure 1D). These results suggest that Afngg1 contributed to lysine acetylation and 2-hydroxyisobutyrylation as a histone acetyltransferase.

Afngg1 Is Involved in A. flavus Growth, Conidia Formation and Aflatoxin Biosynthesis
To further explore the potential role of Afngg1 in A. flavus, we constructed ∆Afngg1 and ∆Afngg1-Com strains ( Figure S2), and the effects of Afngg1 on A. flavus growth, conidia formation and biosynthesis of aflatoxin were evaluated. Unlike the A. flavus control strain, mycelia of the ∆Afngg1 strain were white and fluffy (Figure 2Aa). Stereoscopic microscope observation showed that, compared with the A. flavus control strain, conidia heads of the ∆Afngg1 strain were smaller and fewer in number (Figure 2Ab). Microscopy observation revealed that deletion of Afngg1 reduced the length of conidiophores and conidial number (Figure 2Ac). SEM results showed that conidial heads of the A. flavus control strain were fully formed, while conidial heads of the Afngg1 strain were significantly smaller and deformed ( Figure 2B). Additionally, compared with the A. flavus control strain, the mycelia growth rate was decreased after Afngg1 deletion, and conidia production was decreased by 99.88%, consistent with microscopy observation (Figure 2C,D). Qualitative thin-layer chromatography (TLC) analysis showed that AFB 1 was no longer produced by the A. flavus strain lacking Afngg1 ( Figure 2E). Given the similarity to Ngg1 in C. albicans, we speculated that Afngg1 might also be involved in histone modification. To test this hypothesis, the effects of Afngg1 on the level of acetylation of histone H3 and 2-hydroxyisobutyrylation modification of histone H4 were investigated by collecting mycelia of A. flavus control and ΔAfngg1 strains, and proteins were extracted for Western blotting detection using corresponding antibodies. The results indicated that the acetylation modification level of H3K14 in strain ∆Afngg1 was decreased compared with the A. flavus control strain, and notably, acetylation modification of H3K9 was completely absent ( Figure S1). Unexpectedly, our results also showed that 2-hydroxyisobutyrylation modification levels of H4K8 and H4K5 were significantly lower than in the A. flavus control strain ( Figure 1D). These results suggest that Afngg1 contributed to lysine acetylation and 2-hydroxyisobutyrylation as a histone acetyltransferase.

Effects of Afngg1 on Hydrophobicity and Sclerotia Formation of A. flavus Colonies
We studied the effects of Afngg1 on the surface hydrophobicity of A. flavus cells. The results revealed spherical water droplets on A. flavus control strain colonies, but these were dispersed and absorbed on ∆Afngg1 strain colonies (Figure 3(Aa,Ab)). In order to further probe the hydrophobicity differences, the droplet state of mycelia around droplets on colonies and Bromophenol Blue droplets were observed by stereoscopic microscope. The results showed that mycelia and spores around the A. flavus control strain were attached to the outer side of the droplet, while Bromophenol Blue solution of the ∆Afngg1 strain was diffused throughout mycelia ( Figure 3(Ac)). Additionally, the results showed that compared with the A. flavus control strain, the ∆Afngg1 mutant did not produce sclerotia strain, conidia heads of the ∆Afngg1 strain were smaller and fewer in number ( Figure  2Ab). Microscopy observation revealed that deletion of Afngg1 reduced the length of conidiophores and conidial number (Figure 2Ac). SEM results showed that conidial heads of the A. flavus control strain were fully formed, while conidial heads of the Afngg1 strain were significantly smaller and deformed ( Figure 2B). Additionally, compared with the A. flavus control strain, the mycelia growth rate was decreased after Afngg1 deletion, and conidia production was decreased by 99.88%, consistent with microscopy observation ( Figure 2C,D). Qualitative thin-layer chromatography (TLC) analysis showed that AFB1 was no longer produced by the A. flavus strain lacking Afngg1 ( Figure 2E).

Effects of Afngg1 on Hydrophobicity and Sclerotia Formation of A. flavus Colonies
We studied the effects of Afngg1 on the surface hydrophobicity of A. flavus cells. The results revealed spherical water droplets on A. flavus control strain colonies, but these were dispersed and absorbed on ΔAfngg1 strain colonies (Figure 3(Aa,Ab)). In order to further probe the hydrophobicity differences, the droplet state of mycelia around droplets on colonies and Bromophenol Blue droplets were observed by stereoscopic microscope. The results showed that mycelia and spores around the A. flavus control strain were attached to the outer side of the droplet, while Bromophenol Blue solution of the ∆Afngg1 strain was diffused throughout mycelia ( Figure 3(Ac)). Additionally, the results showed that compared with the A. flavus control strain, the ∆Afngg1 mutant did not produce

Afngg1 Influences the Infection Ability of A. flavus on Peanut and Corn
In order to study the pathogenicity of Afngg1 deletion in A. flavus, maize and corn grains were inoculated with spore suspensions. The results showed that the ∆Afngg1 mutant had a decreased ability to colonise peanut and corn grains, and conidia production

Afngg1 Influences the Infection Ability of A. flavus on Peanut and Corn
In order to study the pathogenicity of Afngg1 deletion in A. flavus, maize and corn grains were inoculated with spore suspensions. The results showed that the ∆Afngg1 mutant had a decreased ability to colonise peanut and corn grains, and conidia production was decreased by 97.39% and 91.94%, respectively ( Figure 4A,B). We also explored the yield of AFB 1 on peanut and corn, and AFB 1 was not detected on either peanut or corn grains infected with the ∆Afngg1 mutant ( Figure 4C). These results indicate that Afngg1 plays a crucial role in conidia formation of A. flavus, and in the biosynthesis of aflatoxin during A. flavus colonisation on peanut and corn grains. and stereoscopic microscopic (c) analysis after Bromophenol blue treatment. (B) Colony morphology of different strains after 7 days of culture on PDA medium. The plate was sprayed with 70% alcohol to expose the sclerotium. (C) Determination of the number of sclerotium in (B). *** represent p < 0.001.

Afngg1 Influences the Infection Ability of A. flavus on Peanut and Corn
In order to study the pathogenicity of Afngg1 deletion in A. flavus, maize and corn grains were inoculated with spore suspensions. The results showed that the ∆Afngg1 mutant had a decreased ability to colonise peanut and corn grains, and conidia production was decreased by 97.39% and 91.94%, respectively ( Figure 4A,B). We also explored the yield of AFB1 on peanut and corn, and AFB1 was not detected on either peanut or corn grains infected with the ∆Afngg1 mutant ( Figure 4C). These results indicate that Afngg1 plays a crucial role in conidia formation of A. flavus, and in the biosynthesis of aflatoxin during A. flavus colonisation on peanut and corn grains.  The number of conidia number was determined from the infected peanut and corn grains; (C) TLC analysis of AFB 1 production from infected peanut and corn grains. *** represents p < 0.001.

Transcriptome Overview
The Pearson correlation coefficient between all pairs of samples was >0.97, suggesting that samples were adequate for subsequent analysis ( Figure 5A). Based on the FPKM (Fragments Per Kilobase per Million) value of each gene, gene expression levels were normalised and differentially expressed genes (DEGs) were compared ( Figure 5B). The volcano map directly shows the expression ratio and significance of genes ( Figure 5C). A total of 4008 significant DEGs were identified, among which 2186 (54.54%) were upregulated and 1822 (45.45%) were downregulated ( Figure 5D).
The biological functions of DEGs were characterised by GO functional enrichment analysis. The results showed that most DEGs were linked to oxidation-reduction metabolic processes ( Figure 6A). In terms of cellular components, DEGs were mainly associated with the plasma membrane, component of membrane and cell periphery subcategories ( Figure 6B). The main molecular functions of DEGs were oxidoreductase activity, binding and catalytic activity ( Figure 6C). KEGG pathway enrichment analysis results showed that DEGs were mainly involved in metabolic pathways, biosynthesis of secondary metabolites and aflatoxin biosynthesis ( Figure 6D).
suggesting that samples were adequate for subsequent analysis ( Figure 5A). Based on the FPKM (Fragments Per Kilobase per Million) value of each gene, gene expression levels were normalised and differentially expressed genes (DEGs) were compared ( Figure 5B). The volcano map directly shows the expression ratio and significance of genes ( Figure  5C). A total of 4008 significant DEGs were identified, among which 2186 (54.54%) were upregulated and 1822 (45.45%) were downregulated ( Figure 5D). The biological functions of DEGs were characterised by GO functional enrichment analysis. The results showed that most DEGs were linked to oxidation-reduction metabolic processes ( Figure 6A). In terms of cellular components, DEGs were mainly associated with the plasma membrane, component of membrane and cell periphery subcategories ( Figure 6B). The main molecular functions of DEGs were oxidoreductase activity, binding and catalytic activity ( Figure 6C). KEGG pathway enrichment analysis results showed that DEGs were mainly involved in metabolic pathways, biosynthesis of secondary metabolites and aflatoxin biosynthesis ( Figure 6D).

Categorisation of DEGs
To further elucidate the regulatory effects of Afngg1 on A. flavus cell development and aflatoxin biosynthesis, representative DEGs were divided into four groups: chromatin remodelling, cell development, secondary metabolism and oxidative stress (Table 1).

Discussion
As an important post-translational modification, 2-hydroxyisobutyrylation plays a vital role in protein synthesis, transcriptional regulation and metabolism [14]. In this study, we performed a functional analysis of histone 2-hydroxyisobutyryltransferase Afngg1 in A. flavus. Deletion of Afngg1 reduced the level of histone 2-hydroxyisobutyrylation modification, indicating that Afngg1 functions as a histone 2-hydroxyisobutyrylyltransferase. Additionally, Afngg1 plays a crucial role in the growth and development of A. flavus and the biosynthesis of aflatoxin, and its potential regulatory mechanisms were explored through transcriptome analysis.

Afnng1 Correlates with the Level of Histone 2-Hydroxyisobutyrylation Modification
Previous reports have shown that Ngg1 is required for histone H3 acetylation of Candida albicans histone H3, and loss of Ngg1 greatly reduced the level of acetylation modification of histone H3 in Ngg1-A and Ngg1-B strains compared with wild-type strains [15]. However, its specific lysine acetylation sites were not characterised. In the present study, we identified homologs of C. albicans acetyltransferase NGG1 in A. flavus, and identified K9ac and K14ac as the key sites for alteration of histone H3 acetylation. Notably, we also found that Afngg1 is associated with 2-hydroxyisobutyrylation of histone H4, and deletion of Afngg1 significantly reduced lysine 2-hydroxyisobutyrylation levels at the K5 and K8 sites of histone H4.

Afnng1 Is Involved in Altering Chromatin Remodelling
Histone modification plays a crucial role in altering chromatin structure, among which histone acetylation and 2-hydroxyisobutyrylation modification are important contributors to chromatin remodelling [16,17]. RPD3 is a conserved histone deacetylase that regulates various cellular processes. In Fusarium graminearum, deletion of FgRpd3 led to increased acetylation of histone H4 [18]. Additionally, the crl3 gene encoding histone deacetylase affects the growth of Penicillium brasilianum, and the production of many secondary metabolites [16]. As a chromatin remodelling factor, Mit1 is mainly involved in regulating transcription of heterochromatin regions, and it promotes the deacetylation of histone H3 by clr3 [19]. We found that following deletion of Afngg1, expression levels of clr3, RPD3, epl1 and mit1 genes related to chromatin remodelling were altered. Additionally, previous studies have shown that cobB, de-2-hydroxyisobutyrylation enzyme, is involved in regulating the growth and metabolism of prokaryotes [8]. In human cells, deletion of the histone 2-hydroxyisobutyltransferase EP300 decreases the level of 2-hydroxyisobutyrylation modification at specific regulatory sites, and thus reduces the activity of cell metabolismrelated enzymes [17]. Our current work showed that histone acetylation modification and 2-hydroxyisobutyrylation levels of the ∆Afngg1 strain were reduced. These results suggest that Afngg1 may alter chromatin conformation, altering the accessibility of genes, thus affecting cell development and secondary metabolism of A. flavus.

Afngg1 Affects Conidial Development and Sclerotium Formation
Recent studies have shown that fungal growth, development and production of secondary metabolites are regulated by post-translational modifications such as 2-hydroxyisobutyrylation [13]. Expression of A. flavus conidiogenesis genes is regulated by a cascade involving brlA, abaA and wetA [20]. BrlA and abaA are key transcription factors, of which brlA initiates conidia formation, and abaA contributes to conidia development and also affects sclerotia formation and AFB 1 production [21,22]. StuA affects the expression of downstream genes by influencing brlA and abaA, and regulates sporulation [23]. VosA, a member of the velvet protein family, co-regulates spore-specific genes with wetA, and is critical for spore maturation and dormancy [24]. Loss of VosA resulted in increased sensitivity to oxidative stress and decreased trehalose content in A. flavus conidia [25]. FluG plays a balancing role in asexual and sexual development, and deletion of fluG can delay and reduce the formation of A. flavus conidia [26]; fluG can also interact with velB or laeA to control the production of conidia and sclerotia [27]. In the present study, the genes wetA, abaA, brlA, stuA, vosA, fluG and con-6 related to conidia were downregulated following deletion of Afngg1, mutant strain colonies were white, conidia production was decreased significantly and the strain no longer produced sclerotia. These results suggest that Afngg1 may regulate developmental balance, oxidative stress and conidia trehalose synthesis, and thereby affect the formation of conidia. Additionally, our experiment found that the ∆Afngg1 strain no longer produced sclerotia, consistent with the results of fluG deletion [27]. Transcriptomic analysis also showed downregulation of genes encoding hydrophobic proteins, namely rodA and dewA, which was confirmed by hydrophobic assay, and is reminiscent of the deletion of laeA in A. flavus [28].

Afngg1 Is Required in the Formation of the Cell Wall
The fungal cell wall is a complex structure composed of chitin, dextran and other polymers that play an important role in fungal growth and survival [29]. Ags1 and Ags2 are involved in glucan synthesis, and growth was inhibited after Ags1 deletion [30]. DCW1 is necessary for incorporation of cell wall glycoproteins into the cell wall, and deletion of the dcw1 gene in yeast led to cell wall abnormalities [31]. MP65 encodes β-glucan mannoprotein, which is associated with cell wall integrity and biofilm formation [32]. The GPI-anchored protein Ecm33 is required for cell wall composition, and is involved in fungal cell wall integrity and multiple stress tolerance [33]. One study found that the absence of ecm33 and MP65 increased sensitivity to cell wall degraders and altered morphology [32,33]. Herein, transcriptomic results showed that expression levels of the cell wall-related genes ags1, ags2, MP65, chiB1, dcw1 and ecm33 in the ∆Afngg1 strain were downregulated. Additionally, ergosterol is an important component of cell membranes, and we found that expression levels of genes associated with ergosterol biosynthesis such as erg4, erg7 and erg5 were significantly downregulated after deletion of Afngg1. These results suggest that Afngg1 may impede the synthesis of cell wall components such as glucan, reduce cell wall tolerance and cause cell wall defects.

Afngg1 Plays a Key Role in Secondary Metabolism
It has been reported that chromatin remodelling is closely related to the biosynthesis of aflatoxin [34]. Aflatoxin biosynthesis includes complex enzymatic reactions that require multiple enzymes, such as fatty acid synthases, polyketone synthases, oxidoreductases, cytochrome P450 monooxygenases, esterases, and others [35]. AflC (pksA) encodes a polyketone synthase that is involved in the conversion of acetate to norsolorinic acid, and AFB 1 production can be significantly increased by upregulating aflC, aflR and aflK [36,37]. Both AflX and AflQ encode oxidoreductases, and AflQ is involved in the formation of AFB 1 precursor hydroxyl-methylsterigmatocystin, and it plays a role in the later stages of the biosynthetic pathway [38,39]. Previous studies revealed a strong linear relationship between aflQ expression and aflatoxin biosynthesis [40]. AflW encodes a single monooxygenase that catalyses the conversion of hydroxyversicolorone to versiconal hemiacetal acetate [41]. Additionally, aflS encodes an esterase (AflJ) that interacts with nearby aflR genes to regulate the biosynthesis of aflatoxin [42]. Following aflS deletion, expression levels of aflatoxin biosynthetic pathway genes including aflC, aflD, aflM and aflP were reduced 5-20-fold, and biosynthesis of aflatoxin intermediates was blocked [43]. In the present study, deletion of Afngg1 downregulated genes encoding enzymes related to the aflatoxin biosynthetic pathway, such as aflC, aflX, aflJ, aflQ, aflT and norA, indicating that Afngg1 can inhibit the biosynthesis of aflatoxin by regulating the expression of multiple genes in the gene cluster via chromatin modification. Our study also found that the ∆Afngg1 strain no longer produced AFB 1 , consistent with knockout of histone acetyltransferase AflGcnE [6].
Other secondary metabolic pathways are also affected. Ustiloxin B is a secondary metabolite of Ustilaginoidea virens, and 18 genes form a gene cluster for the synthesis of ustiloxin B. Deletion of a relevant gene led to a complete or substantial loss of ustiloxin B production in A. flavus [44]. Biosynthesis of gliotoxin is carried out by a synthetic gene cluster consisting of 12 genes in A. fumigatus [45]. Loss of gliA greatly reduced the biosynthesis of glioxin [46], and deletion of gliK significantly hindered efflux of glioxins, and increased sensitivity to exogenous glioxins [47]. Herein, we found that expression levels of ustYa, ustO, ustC, ustP, ustQ, ustD, ustF2, gliK, gliA and gliC, which are responsible for the biosynthesis of ustiloxin B and gliatoxin, were downregulated. In addition, A. flavus produces a variety of polyketone-derived secondary metabolites, which are the most abundant of fungal secondary metabolites [48]. PksP was found to be associated with the biosynthesis and virulence of A. fumigatus conidia pigments [49]. The Pks1 gene encodes a polyketone synthase involved in melanin biosynthesis [50]. Deletion of polyketone synthase gene Flup resulted in decreased filament growth rate and conidial production in A. parasitica [51]. The present study found that expression levels of polyketone synthaserelated genes fluP, pksP, aurA and Pks1 were downregulated. These results suggest that Afngg1 may play an important regulatory role in mycotoxin efflux and self-protection.

Afngg1 Alters Oxidative Stress Homeostasis
Oxidative stress is often associated with the synthesis of fungal secondary metabolites. Biosynthesis of aflatoxin has been reported to be inhibited by increased reactive oxygen species (ROS) levels and decreased activity of antioxidant enzymes [52]. Cells produce a variety of antioxidant enzymes to remove excessive ROS to protect cells from oxidative stress [53]. Previous reports have shown that deletion of cat1 significantly increased intracellular ROS levels, which in turn increased oxidative stress levels and reduced aflatoxin biosynthesis and virulence [52]. In our current work, expression levels of catalase-related genes cat1, ctl-2 and cat were downregulated, suggesting that deletion of Afngg1 might prevent cells from clearing ROS in a timely manner, thereby diminishing their defences against oxidative stress, and inhibiting the biosynthesis of aflatoxin.
In conclusion, we identified histone 2-hydroxyisobutyryltransferase Afngg1 in A. flavus, and explored its potential roles in cell development and aflatoxin biosynthesis. The results indicated that deletion of Afngg1 reduced the degree of lysine 2-hydroxyisobutyrylation modification of histone H4, inhibited cell growth, conidia formation, sclerotia production and AFB 1 production in A. flavus, and decreased hydrophobicity. RNA-seq was subsequently used to investigate the underlying regulatory mechanisms, and analysis of DEGs showed that genes associated with chromatin remodelling, cell development, oxygen stress and biosynthesis of aflatoxin, ustiloxin B and gliatoxin, were downregulated. Therefore, these results suggest that histone 2-hydroxyisobutyryltransferase Afngg1 affects chromosomal modification, alters gene expression and regulates the growth, development and aflatoxin biosynthesis of A. flavus. The findings enhance our understanding of the role of histone 2-hydroxyisobutyrylation in aflatoxin biosynthesis and fungal pathogenicity regulation, and provide potential targets for exploring new control strategies for A. flavus.

Strains and Culture Conditions
A. flavus CA14 (kusA − , pyrG + ) served as a control strain, and the Afngg1 knockout strain (∆Afngg1), Afngg1 complementation strain (∆Afngg1-Com) and Afngg1-enhanced green fluorescence protein (eGFP) strain were derived in this study. Potato dextrose agar (PDA) was used to evaluate growth, conidia formation, cell hydrophobicity and aflatoxin B 1 production, with appropriate amounts of uridine (10 mmol/L) added to the medium as needed. For sclerotia production, spore suspensions were maintained on PDA medium at 37 • C for 7 days in the dark. Each experiment was repeated three times.

Sequence Resources and Phylogenetic Tree Analysis
We used the protein sequence of Afngg1 to perform a BLAST search to identify homologous sequences in the National Center for Biotechnology Information (NCBI) database. Afngg1 domains were predicted by DOG 1.0 software (Ren and Wen, Hefei, China). Afngg1 protein sequences were aligned by ClustalW using MEGA software (7.0 version, iGEM, Temple University, Philadelphia, PA, USA), and a neighbour-joining phylogenetic tree was constructed [54].

Construction of ∆Afngg1, ∆Afngg1-Com and Afngg1-eGFP Strains
The ∆Afngg1 strains were constructed using homologous recombination with PCR primers listed in Table S1. PCR amplification was performed using P505 DNA polymerase (Vazyme, Nanjing, China). For construction of the ∆Afngg1 strain, primers ∆Afngg1-1F, ∆Afngg1-1R, ∆Afngg1-2F and ∆Afngg1-2R were used to amplify the 1.1 kb upstream and 1.5 kb downstream regions of Afngg1 fragments in the A. flavus genome, respectively. Meanwhile, primers pyrG-F and pyrG-R were used to amplify the pyrG selection marker. Fusion PCR was performed using primers ∆Afngg1-1F and ∆Afngg1-2R to generate a fragment containing the upstream and downstream segments, and the pyrG selection marker [55]. The fusion fragment was purified and transformed into the A. flavus control strain according to previous methods [34], and transformants were selected for verification.
Construction of the Afngg1-Com strain was performed according to previous methods [56]. Primers ∆Afngg1-Com-F and ∆Afngg1-Com-R were used to amplify a fragment including the promoter, coding sequence and terminator from A. flavus control strain genomic DNA, and this was ligated to the pPTRI plasmid following HindIII and SmaI restriction enzyme digestion. The resulting plasmid was transferred into A. flavus ∆Afngg1 protoplasts, and the ∆Afngg1-Com strain was confirmed by PCR.
For construction of the Afngg1-eGFP strain, linker, eGFP and TglaA were sequentially connected next to the Afngg1 coding sequence without a stop codon, using primers Afngg1-eGFP-2F and Afngg1-eGFP-2R. The fragment was ligated to the pPTRI plasmid according to previous methods [52]. The confirmed plasmid was transferred into ∆Afngg1 protoplasts, and transformants were selected for verification.

Protein Extraction and Western Blotting Analysis
To explore the effect of Afngg1 deletion on the acetylation level of histone H3 and the 2hydroxyisobutyrylation level of histone H4 in A. flavus, A. flavus control and ∆Afngg1 strains were cultured on PDA medium at 30 • C for 72 h. Mycelia were frozen with liquid nitrogen and ground to a powder. An appropriate amount of each protein sample was mixed with sample buffer and protein lysis solution to make a final concentration of 1-2 mg/mL, and heated at 95 • C for 10 min prior to sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were transferred to a PVDF membrane, and this was blocked with 5% skimmed milk powder after transfer. For analysis of lysine acetylation of histone H3, anti-H3K9ac, anti-H3K14ac and anti-H3 were used as primary antibodies. For analysis of lysine 2-hydroxyisobutyrylation, anti-H4K5 hib (PTM-854; 1:10,000 dilution; PTM-Bio, Hangzhoou, China) and anti-H4K8 hib (PTM-805; 1:10,000 dilution; PTM-Bio, Hangzhoou, China) were used as primary antibodies. Goat anti-mouse IgGI (1:10,000 dilution; Thermo, Shanghai, China) was used as secondary antibody, and chemiluminescent horseradish peroxidase (HRP) substrate (Millipore) was added and incubated for 2 min for signal detection. Meanwhile, protein gray scale analysis was performed using Image-Pro Plus software (4.5 version, Media Cybernetics, Silver Spring, USA) according to a previous method [6].

Morphological and Physiological Analyses
To investigate the effect of Afngg1 on the growth of A. flavus, 2 µL spore suspensions (10 6 spores/mL) of A. flavus control, ∆Afngg1 and ∆Afngg1-Com strains were separately inoculated in the centre of PDA plates and cultured at 30 • C for 5 days; then, the colony diameter was measured and conidial heads were counted under a stereoscopic microscope (Nikon, Shanghai, China). Next, 6 mL sterile water was added to each Petri dish to harvest the mycelium, and the number of conidia was calculated using a hemocytometer. To assess the hydrophobicity of A. flavus mycelia, 20 µL sterile water and 20 µL 3% Bromophenol Blue solution were added to the edges of colonies of A. flavus control, ∆Afngg1 and ∆Afngg1-Com strains, cells were cultured at 30 • C for 2 days, and analysed by a stereoscopic microscope. Quantitative analysis of sclerotia production was performed at 37 • C for 7 days. After 7 days, 75% ethanol was sprayed to wash the medium and expose the sclerotia, and five holes were made along the diameter to count the number of sclerotia. Additionally, scanning electron microscopy (SEM) was used to photograph the conidia microstructure of A. flavus control, ∆Afngg1 and ∆Afngg1-Com strains as described previously [58].

Determination of AFB 1 Production
AFB 1 production analysis was carried out according to a previously described method [59]. Spore suspensions (10 6 spores/mL) of A. flavus control, ∆Afngg1 and ∆Afngg1-Com strains were cultured in 50 mL Yeast Extract Sucrose (YES) liquid medium at 30 • C in the dark for 5 days. An equal volume of chloroform was added to extract aflatoxin from the 50 mL medium. After the chloroform was completely volatilised, the residue was dissolved in 1 mL methanol and then analysed by TLC. The AFB 1 yield of A. flavus strains was evaluated using an AFB 1 standard sample (0.1 mg/mL).

Peanut and Corn Infection Assay
A peanut and corn infection experiment was performed as described previously [60]. Peanut and corn seeds were inoculated with 10 6 /mL spore suspensions of A. flavus control, ∆Afngg1 strain and ∆Afngg1-Com strain and cultured at 28 • C in the dark for 6 days, with water replenished each day. Sterile water was added to collect the spore suspension and the number of conidia was calculated using a hemocytometer according to a previous method [34]. Finally, chloroform was added to extract aflatoxin as described above.

Transcriptome Analysis
Total RNA was extracted using Trizol reagent (Thermo Fisher, Shanghai, China) following the manufacturer's instructions. RNA quality and integrity analyses, cDNA library construction, and RNA-Seq were performed by Guangzhou Gene Denovo Biotechnology (Guangzhou, China). The Pearson correlation coefficient was used to analyse the accuracy and reliability of the three biological duplicate samples. DESeq was used to analyse differences in FPKM value, and differences were considered statistically when log2(fold change) >1 and p-value < 0.05 criteria were met. Genome Ontology (GO) functional analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were carried out using FungiFun (https://sbi.hki-jena.de/FungiFun/FungiFun.cgi) (accessed on 15 August 2022) and the KEGG Automatic Annotation Sever (http://www.kegg.jp/ or http://www.genome.jp/kegg/) (accessed on 15 August 2022).