The CfSnt2-Dependent Deacetylation of Histone H3 Mediates Autophagy and Pathogenicity of Colletotrichum fructicola

Camellia oleifera is one of the most valuable woody edible-oil crops, and anthracnose seriously afflicts its yield and quality. We recently showed that the CfSnt2 regulates the pathogenicity of Colletotrichum fructicola, the dominant causal agent of anthracnose on C. oleifera. However, the molecular mechanisms of CfSnt2-mediated pathogenesis remain largely unknown. Here, we found that CfSnt2 is localized to the nucleus to regulate the deacetylation of histone H3. The further transcriptomic analysis revealed that CfSnt2 mediates the expression of global genes, including most autophagy-related genes. Furthermore, we provided evidence showing that CfSnt2 negatively regulates autophagy and is involved in the responses to host-derived ROS and ER stresses. These combined functions contribute to the pivotal roles of CfSnt2 on pathogenicity. Taken together, our studies not only illustrate how CfSnt2 functions in the nucleus, but also link its roles on the autophagy and responses to host-derived stresses with pathogenicity in C. fructicola.


Introduction
Camellia oleifera is one of the most valuable woody edible-oil crop and has been grown in China for over 2300 years [1]. The oil extracted from its seed has been widely utilized in cooking, lubricant, and cosmetics [2,3]. Through the planting areas of C. oleifera was 4.39 million hectares in China, and anthracnose afflicted its high yield and quality [4][5][6]. We recently showed that the dominant causal agent of anthracnose on C. oleifera is Colletotrichum fructicola, and CfSnt2 regulates the pathogenicity of C. fructicola through mediating the formation of functional appressorium [6,7]. However, the molecular mechanisms of CfSnt2mediated pathogenesis is unclear.
The Snt2 protein was initially identified and named for the DNA binding SANT domain, and it was widely conserved in fungal species [8,9]. The ScSnt2 acts as an E3 ubiquitin ligase to regulate the degradation of excess histone proteins H3 and H4 [8]. The yeast ScSnt2 also recruits histone deacetylase ScRpd3 to the promoters of genes related to oxidative stress for the transcriptional response [10]. In phytopathogenic fungi, there are only limited studies on Snt2 proteins. In Fusarium oxysporum and Neurospora crassa, Snt2 proteins mediate fungal respiration and reactive oxidative stress by regulating the expression of oxidase genes [11]. Further study in F. oxysporum firstly revealed the roles of Snt2 in fungal autophagy and pathogenicity [12]. Furthermore, in Fusarium graminearum, Snt2 links histone deacetylation, autophagy, and pathogenicity [13].
Macroautophagy/autophagy is a highly conserved process, by which proteins, membranes, and organelles are degraded in the vacuole (lysosome) and recycled to satisfy energy needs [14][15][16][17]. A growing body of evidence points to the pivotal roles of autophagy in fungal pathogenicity [18][19][20]. In Magnaporthe oryzae, autophagy is important for the formation of functional appressoria, lipid metabolism, and pathogenicity [21,22]. In F. graminearum, targeted gene deletion of autophagy-related genes FgATG9 showed decreased pathogenicity in wheat, which was caused by the severe defects in autophagy and lipid metabolism [23]. We also previously revealed that the CfATG8 gene deletion mutant in C. fructicola was non-pathogenic to C. oleifera [24].
Recent studies in the phytopathogenic fungus have shown the critical regulatory roles of acetylation levels for autophagy-dependent pathogenicity. The histone acetyltransferase MoHat1 governs autophagy-dependent pathogenicity through acetylates MoAtg3 and MoAtg9 in M. oryzae [25]. The M. oryzae histone acetyltransferase MoGcn5 links the acetylation of MoAtg7 between the light-induced autophagy and pathogenicity [26]. The F. graminearum FgGcn5 acetylates histone H3 and FgAtg8, governing the function associated with autophagy and pathogenicity [27]. Our recent study also revealed that CfGcn5 regulates development and pathogenicity in C. fructicola [28]. Whether acetylation modifications are related with autophagy-dependent pathogenicity in C. fructicola is unknown.
In the present study, we revealed that CfSnt2 is localized to nucleus for the deacetylation of histone H3, resulting in its regulation of global genes expression. Meanwhile, we revealed that CfSnt2 mediates the expression of multiple CfATG genes and negatively regulates autophagy. In addition, we also demonstrated that CfSnt2 is involved in the tolerance of oxidative and endoplasmic reticulum (ER) stress.

CfSnt2 Is Localized to the Nucleus and Mediates H3 Deacetylation
To further dissect the function of CfSnt2, we fused a GFP tag to the C-terminus of CfSnt2 and found that CfSnt2-GFP showed a slight fluorescence affecting visualization. Therefore, we fused the GFP tag to the N-terminus of CfSnt2 and found that GFP-CfSnt2 was clearly visible as spot green fluorescence. To investigate whether the spot green fluorescence represents the nucleus, we fused the RFP tag to histone H1, a widely used nucleus marker. The co-localization of GFP-CfSnt2 and H1-RFP showed that CfSnt2 localized to the nucleus in the mid and tip regions of hyphae and conidia ( Figure 1A,B). We then wondered whether CfSnt2 mediates H3 deacetylation in the nucleus. The previous study in M. oryzae revealed that H3K18 acetylation (H3K18ac) mediated by the histone acetyltransferase MoTig1 is essential for pathogenicity [29]. The immunoblotting assays, using the anti-H3K18ac and anti-H3 antibodies, revealed the increased acetylation levels of histone H3K18 in ∆Cfsnt2 mutant, compared with the wide type (WT) and complemented strains ∆Cfsnt2/CfSNT2-1 and ∆Cfsnt2/CfSNT2-2 ( Figure 1C). Since the complemented strains ∆Cfsnt2/CfSNT2-1 and ∆Cfsnt2/CfSNT2-2 showed similar phenotypes, we selected ∆Cfsnt2/CfSNT2-1 for our further study.

Transcriptomic Analysis of the WT and ∆Cfsnt2 Mutant
To test the roles of CfSnt2-mediated H3 deacetylation in the nucleus, we carried out transcriptome analysis for the WT and ∆CfSnt2 mutant by RNA sequencing (RNA-seq). Three biological replicates were established for each strain, and six RNA-seq data sets were generated, among which, more than 90% reads were mapped to the genome of C. fructicola (GenBank accession number: GCA_000319635.2). All raw data were submitted to the NCBI SRA database, with accession numbers SRR19052615, SRR19052614, and SRR19052613 for WT 1-3 and SRR19052618, SRR19052617, and SRR19052616 for ∆Cfsnt2 1-3. Principal component analysis (PCA) of WT and ∆Cfsnt2 mutant revealed a clear separation of the two tested strains, as well as the proximity of the biological replicates ( Figure 2A). Gene expression analysis revealed that a total of 9572 predicted genes were expressed in both the WT and ∆Cfsnt2 mutant, and 990 and 924 predicted genes were specifically expressed in the WT and ∆Cfsnt2 mutant ( Figure 2B). Differentially expressed genes (DEGs) analysis showed that 2408 predicted genes were up-regulated and 2565 predicted genes were downregulated at least two-fold (p < 0.01) in the ∆Cfsnt2 mutant ( Figure 2C). (C) Total proteins extracted from mycelia of WT, ΔCfsnt2 mutant and complemented strains ΔCfsnt2/CfSNT2-1 and ΔCfsnt2/CfSNT2-2 were immunoblotted with the anti-H3K18ac and anti-H3 antibodies. The intensity of the protein bands were analyzed by Image J, and the intensity of WT was defined as a reference, with 1.00.

Transcriptomic Analysis of the WT and ΔCfsnt2 Mutant
To test the roles of CfSnt2-mediated H3 deacetylation in the nucleus, we carried out transcriptome analysis for the WT and ΔCfSnt2 mutant by RNA sequencing (RNA-seq). Three biological replicates were established for each strain, and six RNA-seq data sets were generated, among which, more than 90% reads were mapped to the genome of C. fructicola (GenBank accession number: GCA_000319635.2). All raw data were submitted to the NCBI SRA database, with accession numbers SRR19052615, SRR19052614, and SRR19052613 for WT 1-3 and SRR19052618, SRR19052617, and SRR19052616 for ΔCfsnt2 1-3. Principal component analysis (PCA) of WT and ΔCfsnt2 mutant revealed a clear separation of the two tested strains, as well as the proximity of the biological replicates (Figure 2A). Gene expression analysis revealed that a total of 9572 predicted genes were expressed in both the WT and ΔCfsnt2 mutant, and 990 and 924 predicted genes were specifically expressed in the WT and ΔCfsnt2 mutant ( Figure 2B). Differentially expressed genes (DEGs) analysis showed that 2408 predicted genes were up-regulated and 2565 predicted genes were down-regulated at least two-fold (p < 0.01) in the ΔCfsnt2 mutant (Figure 2C). (C) Total proteins extracted from mycelia of WT, ∆Cfsnt2 mutant and complemented strains ∆Cfsnt2/CfSNT2-1 and ∆Cfsnt2/CfSNT2-2 were immunoblotted with the anti-H3K18ac and anti-H3 antibodies. The intensity of the protein bands were analyzed by Image J, and the intensity of WT was defined as a reference, with 1.00.

Gene Ontology Enrichment of the DEGs
All the DEGs were further analyzed by gene ontology (GO) enrichment (p < 0.01), and a total of 25 GO enrichment terms were categorized into three main categories: biological process, cellular component, and molecular function ( Figure 3). The biological process contains nine terms, including translation, response to oxidative stress, amine metabolic process, cellular amide metabolic process, fatty acid biosynthetic process, replication fork protection, amide biosynthetic process, invasive growth in response to glucose, and limitation peptide metabolic process. The cellular component contains three terms, including ribosome, integral component of membrane, and small ribosomal subunit. The molecular function contains 13 terms, including structural constituent of ribosome, N-acetyltransferase activity, ATPase activity, coupled to transmembrane movement of substances, pyridoxal phosphate binding, aspartic-type endopeptidase activity, molybdenum ion binding, glutamate-ammonia ligase activity, glutaminase activity, ATPase activity, alpha-1,6-mannosyltransferase activity, unfolded protein binding, copper ion binding, and nucleotidyltransferase activity.

Gene Ontology Enrichment of the DEGs
All the DEGs were further analyzed by gene ontology (GO) enrichment (p < 0.01), and a total of 25 GO enrichment terms were categorized into three main categories: biological process, cellular component, and molecular function ( Figure 3). The biological process contains nine terms, including translation, response to oxidative stress, amine metabolic process, cellular amide metabolic process, fatty acid biosynthetic process, replication fork protection, amide biosynthetic process, invasive growth in response to glucose, and limitation peptide metabolic process. The cellular component contains three terms, including ribosome, integral component of membrane, and small ribosomal subunit. The molecular function contains 13 terms, including structural constituent of ribosome, Nacetyltransferase activity, ATPase activity, coupled to transmembrane movement of substances, pyridoxal phosphate binding, aspartic-type endopeptidase activity, molybdenum ion binding, glutamate-ammonia ligase activity, glutaminase activity, ATPase activity, alpha-1,6-mannosyltransferase activity, unfolded protein binding, copper ion binding, and nucleotidyltransferase activity.

KEGG Pathway Enrichment of the DEGs
Significantly enriched (p < 0.01) KEGG pathways of the DEGs were also analyzed. The results showed that 10 KEGG pathways were significantly enriched, including one carbon pool by folate, arginine biosynthesis, nitrogen metabolism, steroid biosynthesis, ribosome, nicotinate and nicotinamide metabolism, glyoxylate and dicarboxylate metab-

KEGG Pathway Enrichment of the DEGs
Significantly enriched (p < 0.01) KEGG pathways of the DEGs were also analyzed. The results showed that 10 KEGG pathways were significantly enriched, including one carbon pool by folate, arginine biosynthesis, nitrogen metabolism, steroid biosynthesis, ribosome, nicotinate and nicotinamide metabolism, glyoxylate and dicarboxylate metabolism, alanine, aspartate and glutamate metabolism, butanoate metabolism, and glycine, serine, and threonine metabolism (Figure 4).

CfSnt2 Mediates the Expression of Multiple CfATG Genes
To verify the gene expression profiles in the transcriptomic data, three up-regulated and three down-regulated genes were also analyzed by qRT-PCR. Though the magnitude of fold changes showed a slight change to the transcriptomic data, the qRT-PCR data were all consistent with that from the transcriptomic data ( Figure 5A). In the RNA-seq data, we also found most autophagy-related genes (Table S1), particularly for the homologs of pathogenicity-associated macro-autophagy in M. oryzae [21], were among the DEGs. The qRT-PCR analysis demonstrated the upregulation of 13 CfATGs among 15 genes, including CfATG1, CfATG2, CfATG3, CfATG4, CfATG6, CfATG7, CfATG8, CfATG9, CfATG10, CfATG12, CfATG13, CfATG15, and CfATG18 ( Figure 5B). These results revealed that CfSnt2 mediates the expression of most autophagy-related genes, and we hypothesized its roles in autophagy.

CfSnt2 Mediates the Expression of Multiple CfATG Genes
To verify the gene expression profiles in the transcriptomic data, three up-regulated and three down-regulated genes were also analyzed by qRT-PCR. Though the magnitude of fold changes showed a slight change to the transcriptomic data, the qRT-PCR data were all consistent with that from the transcriptomic data ( Figure 5A). In the RNA-seq data, we also found most autophagy-related genes (Table S1), particularly for the homologs of pathogenicity-associated macro-autophagy in M. oryzae [21], were among the DEGs. The qRT-PCR analysis demonstrated the upregulation of 13 CfATGs among 15 genes, including CfATG1, CfATG2, CfATG3, CfATG4, CfATG6, CfATG7, CfATG8, CfATG9, CfATG10, CfATG12, CfATG13, CfATG15, and CfATG18 ( Figure 5B). These results revealed that CfSnt2 mediates the expression of most autophagy-related genes, and we hypothesized its roles in autophagy.

CfSnt2 Is Involved in the Rapamycin Response
To demonstrate our hypothesis, we firstly test whether CfSnt2 is involved in the response of rapamycin, which induces autophagy via the Tor (target of rapamycin) signaling pathway [30,31]. When exposed to 25 nM rapamycin, the ΔCfsnt2 mutant exhibited a significantly higher inhibition rate than that of WT and ΔCfsnt2/CfSNT2 ( Figure 6A,B). The results indicated that CfSnt2 is involved in the rapamycin response. Error bars indicate SD of three replicates (** p < 0.01).

CfSnt2 Negatively Regulates Autophagy
To examine how CfSnt2 mediates autophagy, the fusion gene GFP-CfATG8, homologs of which were widely used as markers for macro-autophagy in other organisms [32,33], were introduced into the WT and ΔCfsnt2. When grown in PDA-rich medium, ΔCfsnt2 mutant proved to be more autophagosomes than WT ( Figure 7A,B). We further assessed autophagic flux by immunoblot in MM-N condition, another way to induce autophagy [34]. The free GFP (26 kDa) and full length GFP-CfAtg8 (46 kDa) could easily be

CfSnt2 Is Involved in the Rapamycin Response
To demonstrate our hypothesis, we firstly test whether CfSnt2 is involved in the response of rapamycin, which induces autophagy via the Tor (target of rapamycin) signaling pathway [30,31]. When exposed to 25 nM rapamycin, the ∆Cfsnt2 mutant exhibited a significantly higher inhibition rate than that of WT and ∆Cfsnt2/CfSNT2 ( Figure 6A,B). The results indicated that CfSnt2 is involved in the rapamycin response.

CfSnt2 Is Involved in the Rapamycin Response
To demonstrate our hypothesis, we firstly test whether CfSnt2 is involved in the response of rapamycin, which induces autophagy via the Tor (target of rapamycin) signaling pathway [30,31]. When exposed to 25 nM rapamycin, the ΔCfsnt2 mutant exhibited a significantly higher inhibition rate than that of WT and ΔCfsnt2/CfSNT2 ( Figure 6A,B). The results indicated that CfSnt2 is involved in the rapamycin response. Error bars indicate SD of three replicates (** p < 0.01).

CfSnt2 Negatively Regulates Autophagy
To examine how CfSnt2 mediates autophagy, the fusion gene GFP-CfATG8, homologs of which were widely used as markers for macro-autophagy in other organisms [32,33], were introduced into the WT and ΔCfsnt2. When grown in PDA-rich medium, ΔCfsnt2 mutant proved to be more autophagosomes than WT ( Figure 7A,B). We further assessed autophagic flux by immunoblot in MM-N condition, another way to induce autophagy [34]. The free GFP (26 kDa) and full length GFP-CfAtg8 (46 kDa) could easily be

CfSnt2 Negatively Regulates Autophagy
To examine how CfSnt2 mediates autophagy, the fusion gene GFP-CfATG8, homologs of which were widely used as markers for macro-autophagy in other organisms [32,33], were introduced into the WT and ∆Cfsnt2. When grown in PDA-rich medium, ∆Cfsnt2 mutant proved to be more autophagosomes than WT ( Figure 7A,B). We further assessed autophagic flux by immunoblot in MM-N condition, another way to induce autophagy [34]. The free GFP (26 kDa) and full length GFP-CfAtg8 (46 kDa) could easily be detected in both the WT-and ∆Cfsnt2 mutant-expressing GFP-CfATG8 strains. Before MM-N induction, WT showed lower amounts of free GFP than GFP-CfAtg8, but the ∆Cfsnt2 mutant showed comparable free GFP to GFP-CfAtg8. Upon MM-N induction for 4 h, ∆Cfsnt2 mutant contained a slight GFP-CfAtg8, which contrasted with the still large amounts of GFP-CfAtg8 in WT ( Figure 7C). The level of autophagy was also evaluated by measuring free GFP, compared with the total of free GFP and intact GFP-CfAtg8 together. The ratio of GFP in ∆Cfsnt2 mutant was significantly higher than that in WT, both in the non-induction and induction conditions ( Figure 7D), indicating the increased autophagy in ∆Cfsnt2 mutant. The results revealed that CfSnt2 negatively regulates autophagy. detected in both the WT-and ΔCfsnt2 mutant-expressing GFP-CfATG8 strains. Before MM-N induction, WT showed lower amounts of free GFP than GFP-CfAtg8, but the ΔCfsnt2 mutant showed comparable free GFP to GFP-CfAtg8. Upon MM-N induction for 4 h, ΔCfsnt2 mutant contained a slight GFP-CfAtg8, which contrasted with the still large amounts of GFP-CfAtg8 in WT ( Figure 7C). The level of autophagy was also evaluated by measuring free GFP, compared with the total of free GFP and intact GFP-CfAtg8 together.
The ratio of GFP in ΔCfsnt2 mutant was significantly higher than that in WT, both in the non-induction and induction conditions ( Figure 7D), indicating the increased autophagy in ΔCfsnt2 mutant. The results revealed that CfSnt2 negatively regulates autophagy.

CfSnt2 Regulates Pathogenicity and the Responses to Oxidative and ER Stresses
CfSnt2 is involved in pathogenicity in C. fructicola ( Figure 8A,B), partially due to its roles in the formation of functional appressoria [7]. To explore other underlying mechanism in the pathogenesis, we found that the response to oxidative stress and unfolded protein binding terms are among the GO enrichment of the DEGs. The reactive oxygen species (ROS) is a general defense response for the plants in plant-pathogen interactions, and it is necessary for the successful infection of pathogens to overcome host-derived ROS [35,36]. Additionally, the previous studies showed that pathogens must also activate the unfolded protein response to face host-derived ER stress during infection [37,38]. Thus, we used H2O2 and dithiothreitol (DTT) to mimic the host-derived ROS and ER stress, respectively. We found that ΔCfsnt2 mutant were more sensitive to both H2O2 and DTT stresses, compared with WT and ΔCfsnt2/CfSNT2 (Figure 8C,D). These results indicated that CfSnt2 mediates pathogenicity and the responses to oxidative and ER stresses.

CfSnt2 Regulates Pathogenicity and the Responses to Oxidative and ER Stresses
CfSnt2 is involved in pathogenicity in C. fructicola ( Figure 8A,B), partially due to its roles in the formation of functional appressoria [7]. To explore other underlying mechanism in the pathogenesis, we found that the response to oxidative stress and unfolded protein binding terms are among the GO enrichment of the DEGs. The reactive oxygen species (ROS) is a general defense response for the plants in plant-pathogen interactions, and it is necessary for the successful infection of pathogens to overcome host-derived ROS [35,36]. Additionally, the previous studies showed that pathogens must also activate the unfolded protein response to face host-derived ER stress during infection [37,38]. Thus, we used H 2 O 2 and dithiothreitol (DTT) to mimic the host-derived ROS and ER stress, respectively. We found that ∆Cfsnt2 mutant were more sensitive to both H 2 O 2 and DTT stresses, compared with WT and ∆Cfsnt2/CfSNT2 ( Figure 8C,D). These results indicated that CfSnt2 mediates pathogenicity and the responses to oxidative and ER stresses. Three independent experiments were performed with three biological replicates each time. Error bars indicate SD of three replicates (** p < 0.01).

Discussion
Colletotrichum spp. is ranked as the eighth most important fungal phytopathogens in plant pathology and causes anthracnose disease on almost every crop [39]. Despite the economical and genetic importance, its pathogenesis remains in a state that is largely unclear, particularly for the species of C. fructicola. We previously revealed that CfSnt2 regulates the pathogenicity by mediating the formation of functional appressorium in C. fructicola, but the underlying molecular mechanisms are unknown [3]. In this study, we not only provided evidence regarding how the CfSnt2 mechanism regulates global gene expressions, but also highlighted its roles in the pathogenicity-related autophagy and stress responses.
The yeast ScSnt2 protein was localized to the promoters of the stress response genes in the nucleus [10]. The M. oryzae MoSnt2 protein was also localized to the nucleus in different developmental stages [13]. Consistent with these studies, our research revealed that CfSnt2 was localized to the nucleus in the hyphae and conidia of C. fructicola, thus indicating its conserved regulatory roles in the nucleus. The further decreased acetylation level of H3K18 supports its functions on histone deacetylation in nucleus.
Acetylation and deacetylation of histones are pivotal epigenetic modifications for the transcription of genes [40,41]. Over the past decade, growing cases have indicated that the epigenetic acetylation and deacetylation of histones share important functions in fungal

Discussion
Colletotrichum spp. is ranked as the eighth most important fungal phytopathogens in plant pathology and causes anthracnose disease on almost every crop [39]. Despite the economical and genetic importance, its pathogenesis remains in a state that is largely unclear, particularly for the species of C. fructicola. We previously revealed that CfSnt2 regulates the pathogenicity by mediating the formation of functional appressorium in C. fructicola, but the underlying molecular mechanisms are unknown [3]. In this study, we not only provided evidence regarding how the CfSnt2 mechanism regulates global gene expressions, but also highlighted its roles in the pathogenicity-related autophagy and stress responses.
The yeast ScSnt2 protein was localized to the promoters of the stress response genes in the nucleus [10]. The M. oryzae MoSnt2 protein was also localized to the nucleus in different developmental stages [13]. Consistent with these studies, our research revealed that CfSnt2 was localized to the nucleus in the hyphae and conidia of C. fructicola, thus indicating its conserved regulatory roles in the nucleus. The further decreased acetylation level of H3K18 supports its functions on histone deacetylation in nucleus.
Acetylation and deacetylation of histones are pivotal epigenetic modifications for the transcription of genes [40,41]. Over the past decade, growing cases have indicated that the epigenetic acetylation and deacetylation of histones share important functions in fungal pathogenesis by regulating gene expression and other mechanisms [42,43]. The large amount of DEGs (2408 up-regulated and 2565 down-regulated) in the ∆Cfsnt2 mutant of the transcriptomic data revealed its roles in gene expression through histone deacetylation. However, histone deacetylation is generally associated with transcriptional inactivation [40,44]. The slightly more down-regulated genes, rather than up-regulated genes, indicated that Cfsnt2 might also mediate the acetylation of other histones. This was apparent in the recent study of F. graminearum, which demonstrated that FgFng3 not only regulates H3 acetylation, but also mediates H4 deacetylation [45]. This might be interesting for illustrating the link between CfSnt2-mediated acetylation and gene expression, and further studies are warranted.
The GO and KEGG enrichment analysis of the DEGs indicated that CfSnt2 regulates the genes that are mainly involved in translation, ribosomal biogenesis, amino acid, and carbohydrate biosynthesis/metabolism, thus revealing the basic roles of CfSnt2 in the physiological process of C. fructicola. Particularly, most of the pathogenicity-related CfATG genes were significantly increased in the ∆Cfsnt2 mutant, thus forecasting its roles in autophagy. The contributions of CfSnt2 on rapamycin stress response further confirmed this hypothesis. Indeed, the autophagosome abundance was increased in the ∆Cfsnt2 mutant, which showed similar phenotype to the F. graminearum and F.oxysporum ∆snt2 mutant [12,13]. Again, the furthermore immunoblot analysis also demonstrated the negative role of CfSnt2 in autophagy. Multiple evidences point to the essential roles of autophagy in fungal pathogenicity [21,38]. The disruption of autophagic homeostasis in the ∆Cfsnt2 mutant might be one reason for the pathogenicity defects. Additionally, the ∆Cfsnt2 mutant is sensitive to the mimicked hostderived ROS and ER stresses, which might be another reason for its pathogenicity defects.
In summary, we demonstrated that CfSnt2 is localized to the nucleus, whereby it reduces the acetylation abundance of histone H3, resulting in the changed expression of multiple genes associated with developmental processes, including pathogenicity-related autophagy and ROS and ER stress responses. The identification of epigenetic histone and non-histone modifications for CfSnt2 in pathogenesis remains interesting work, and further studies are highly warranted.

Strains and Culture Conditions
The C. fructicola CFLH16 strain was used as WT in this study. The CfSNT2 gene deletion mutant was generated by the standard one-step gene replacement strategy in our previous study [7]. Briefly, two~1.0 kb sequences flanking the CfSNT2 gene were PCR amplified and overlapped to the flanks of hygromycin resistance cassette (1.4 kb), respectively. After sequencing, the~3.4 kb fragment was transformed into the protoplasts of WT for gene deletion. The complement fragment, which contains the entire CfSNT2 gene and its native promoter region, was amplified by PCR and inserted into pYF11 (bleomycin resistance) to complement the ∆Cfsnt2 mutant. All strains were cultured on potato dextrose agar (PDA) agar plates in the darkness at 28 • C. Liquid potato dextrose broth (PDB) medium was used to culture the mycelia for DNA, RNA, and protein extraction.
To test the autophagic flux, the strains were incubated in MM-N medium (0.52 g KCl, 0.152 g MgSO 4 ·7H 2 O, 1.52 g KH 2 PO 4 , 0.01 g vitamin B1, 1 mL trace elements, and 10 g D-glucose in 1 L of distilled water).

Generation of the CfSnt2-GFP, H1-RFP, and GFP-CfSnt2 Constructs
For generating CfSnt2-GFP, PCR fragment of the~1.5 kb CfSNT2 native promoter and gene coding sequence was co-transformed with XhoI-digested pYF11 into yeast strain XK1-25 [46]. Then, the fused CfSnt2-GFP construct was transformed into the Escherichia coli Trelief TM 5α (TsingKe Biological Technology Co., Ltd, Beijing, China) and further sequenced. Similarly, the H1-RFP construct was generated by cloning the H1 gene into the pHZ126 vector (hygromycin resistance). For generating GFP-CfSnt2, the native promoter, 0.7 kb GFP fragment, and full-length of CfSNT2 were successively fused together and then inserted into the pYF11.

Localization Assays
The hyphae and conidia of the strain co-transformed with GFP-CfSnt2, and H1-RFP were visualized under a fluorescent microscope (ZEISS, Axio Observer. A1, Jena, Germany).

Transcriptome Sequencing and Analysis
The RNA of the collected sample was extracted from WT and ∆Cfsnt2 mutant with three biological replicates for each strain. Transcriptome sequencing was carried out by the Biomarker Technologies Company (Beijing, China). The libraries were sequenced on an Illumina HiSeq platform with 150 bp paired-end reads, followed by the alignment to the reference genome by Hisat2 [47]. DEG analysis were performed with an adjusted p < 0.01 and fold change ≥2 by DESeq2 [48].

QRT-PCR Assays
Total RNA were reverse transcribed into first-strand cDNA. The qRT-PCR using 2×TSINGKE ® Master qPCR Mix (SYBR Green I, TsingKe Biological Technology Co., Ltd, Beijing, China) with primers (Table S2) was run on the ABI QuantStudio 3. The ACTIN gene was used as an internal reference for relative expression analysis. The experiment was repeated in at least 3 biological experiments independently, with 3 replicates.

Stress Response Analysis
The strains were cultured on PDA, and PDA added with 25 nM rapamycin, 10 mM or 20 mM H 2 O 2 , and 5 mM DTT. After 3 days of growth, the colony diameters were examined, and the inhibited rates were analyzed statistically.

Pathogenicity Assays
The pathogenicity assays were performed as previously described [49]. The mycelial plugs of the strains were inoculated onto the margin area of the wounded C. oleifera leaves. After incubation in a humidity plate with 12 h light and 12 h dark cycles for 4 days, the inoculated leaves were photographed and analyzed by Image J.

Statistical Analysis
Each result was presented as the mean ± SD of three replicates. The significance of differences between samples were analyzed by ANOVA (analysis of variance) with Duncan's new multiple range test. The level of significance was set at p < 0.01 or p < 0.05.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jof8090974/s1, Table S1: The expression levels of CfATG genes in transcriptome data; Table S2: Primers used in this study. Data Availability Statement: All data supporting the findings of this study are available within the paper and its supplementary materials (published online).

Conflicts of Interest:
The authors declare no competing interests.