FgSnt1 of the Set3 HDAC complex plays a key role in mediating the regulation of histone acetylation by the cAMP-PKA pathway in Fusarium graminearum

Abstract

The cAMP-PKA pathway is critical for regulating growth, differentiation, and pathogenesis in fungal pathogens. In Fusarium graminearum, mutants deleted of PKR regulatory-subunit of PKA had severe defects but often produced spontaneous suppressors. In this study eleven pkr suppressors were found to have mutations in FgSNT1, a component of the Set3C histone deacetylase (HDAC) complex, that result in the truncation of its C-terminal region. Targeted deletion of the C-terminal 98 aa (CT98) in FgSNT1 suppressed the defects of pkr in growth and H4 acetylation. CT98 truncation also increased the interaction of FgSnt1 with Hdf1, a major HDAC in the Set3 complex. The pkr mutant had no detectable expression of the Cpk1 catalytic subunit and PKA activities, which was not suppressed by mutations in FgSNT1. Cpk1 directly interacted with the N-terminal region of FgSnt1 and phosphorylated it at S443, a conserved PKA-phosphorylation site. CT98 of FgSnt1 carrying the S443D mutation interacted with its own N-terminal region. Expression of FgSNT1^S443D rescued the defects of pkr in growth and H4 acetylation. Therefore, phosphorylation at S443 and suppressor mutations may relieve self-inhibitory binding of FgSnt1 and increase its interaction with Hdf1 and H4 acetylation, indicating a key role of FgSnt1 in crosstalk between cAMP signaling and Set3 complex.

Author summary

The cAMP-protein kinase A (PKA) pathway is one of the well-conserved signal transduction pathways in F. graminearum that regulates various infection and developmental processes. In this study we showed that deletion of the C-terminal region of FgSnt1 suppressed the defects of the pkr mutants deleted of the regulator subunit of PKA in growth and H4 histone acetylation. Snt1 is a component of the well-conserved Set3 histone deacetylase (HDAC) complex that is important for regulating histone acetylation and gene expression. Truncation of FgSnt1 increased its interaction with Hdf1, a major HDAC in the Set3 complex. Although Pkr did not, the Cpk1 catalytic subunit of PKA directly interacted with the N-terminal region of FgSnt1 and phosphorylated it at the conserved S443 site. Nevertheless, Pkr is important for protecting the Cpk1 proteins from degradation by the 26S proteosome. The S443D mutation mimicking phosphorylation of FgSnt1 by Cpk1 enabled the likely self-inhibitory interaction between its N-terminal and C-terminal regions and partially rescued the defects of pkr and H4 acetylation. Overall, this work described the comprehensive genetic interaction of the Set3 complex with the cAMP-PKA pathway. Our work extends and deepens our understanding of cAMP-PKA pathway in histone acetylation and virulence of plant pathogenic fungi.

Introduction

The filamentous ascomycete Fusarium graminearum is the major causal agent of Fusarium head blight (FHB), which is one of the most important diseases of wheat and barley worldwide [1]. It overwinters on plant debris and releases ascospores from perithecia in the spring. As the primary inoculum, ascospores initiate the disease cycle after landing on flowering wheat or barley heads and form infection cushions or compound appressoria for plant penetration. Morphologically distinct infectious hyphae can grow inter- and intra-cellularly in infected plant tissues and spread via the rachis from the initial infection site to neighboring spikelets [2,3]. Under favorable environmental conditions, outbreaks of FHB cause severe yield losses and often contaminate infested grains with the harmful mycotoxins deoxynivalenol (DON) and zearalenone [4]. DON is also an important virulence factor that facilitates the spread of F. graminearum infection from the initial infection site via the rachis to the rest of spikelets [4,5].

Unlike many other plant pathogenic ascomycetes, the genome of F. graminearum has no active transposable elements and contains less than 0.05% repetitive sequences. In addition, it has chromosomal regions that have higher genetic variations and are enriched with fungal-plant interaction-related genes [6]. Further studies showed that these chromosomal regions are correlated to facultative heterochromatin regions with high levels of H3K27me3 in vegetative hyphae [7,8]. Therefore, chromosomal organization and chromatin modification appear to play a critical role in regulating genes that are specifically expressed during different developmental and infection stages in F. graminearum. Consistent with the chromosomal organizations, the TBL1 gene encoding a transducing-like protein homologous to yeast Sif2 was found to be important for conidiation and infectious growth in the rachis tissues [9]. Sif2 is a component of the well-conserved Set3 histone deacetylase (HDAC) complex that also consists of Set3, Snt1, Hos2, Cpr1, Hst1, and Hos4 in yeast [10]. In the rice blast fungus Magnaporthe oryzae, genes orthologous to the key components of the yeast Set3 complex are also important for conidiation, appressorium penetration, and invasive growth [11]. In F. graminearum, HDF1, an ortholog of yeast HOS2, is the major class II HDAC gene that plays a critical role in regulating hyphal growth, sexual and asexual reproduction, DON biosynthesis, and plant infection [12]. Recently, another HDAC gene, FgRPD3, and the FgESA1 histone deacetylase (HAT) gene were found to play opposing roles in balancing histone acetylation and regulating transcriptional accessibility, growth, development, and infection processes in this fungal pathogen [13].

F. graminearum is a homothallic fungus that is amenable to classic and molecular genetic studies. Since the publication of its genome sequence, many genes with various biological functions, including those encoding protein kinases, transcription factors, and key components of intracellular signaling pathways, have been characterized for their functions in regulating infection, sexual and asexual reproduction, and DON production [14–19]. Although accelerated by the release of the F. graminearum genome, a substantial amount of functional work had already been accomplished. Like in other fungal pathogens [20,21], the cAMP-protein kinase A (PKA) pathway is one of the well-conserved signal transduction pathways in F. graminearum that regulates various infection processes [22,23]. The Fac1 adenylate cyclase and Pde1 cAMP phosphodiesterase responsible for regulating the intracellular cAMP level are the first two components of this important signaling pathway characterized for their functions in vegetative growth, conidiation, ascosporogenesis, and DON production in this pathogen [24,25]. The CPK1 and CPK2 genes encode the major and minor catalytic subunits of PKA, respectively, as well as the PKR gene encoding the regulatory subunit also have been functionally characterized in F. graminearum [22,23]. Whereas deletion of CPK1 results in pleiotropic defects, the cpk2 mutant has no detectable phenotypes. However, the cpk1 cpk2 double mutant has more severe defects than the cpk1 mutant [23]. The pkr mutant shares similar phenotypes with the cpk1 cpk2 double mutant, including severe defects in growth and pathogenesis as well as loss of fertility, but the latter produced normal conidia. In contrast, conidia of pkr mutant have morphological defects and autophagy-related cell death [22].

Interestingly, both the cpk1 cpk2 double and pkr deletion mutants of F. graminearum were unstable and produced spontaneous suppressor strains that had faster growth rate. All but one of the 30 spontaneous suppressor strains of the cpk1 cpk2 mutant had null mutations in FgSFL1 [26,27]. In the budding yeast Saccharomyces cerevisiae, Sfl1 is a transcription factor that functions downstream from cAMP signaling and interacts with the Cyc8-Tup1 co-repressor complex for regulating gene expression [28,29]. In the rice blast fungus M. oryzae, suppressor mutations in MoSFL1 rescue the growth defect of the cpkA cpk2 mutant by relieving the suppression of subsets of genes by the MoCyc8-Tup1 complex that normally requires the phosphorylation of MoSfl1 by PKA [26]. For the pkr mutant in F. graminearum, a total of 67 suppressor strains were isolated. Like suppressors of the cpk1 cpk2 mutant, suppressors of the pkr mutant were rescued in growth defects but still blocked in sexual reproduction and spreading infectious growth via the rachis [22,23]. Whereas none of the suppressor strains of the cpk1 cpk2 mutant had mutations in PKR, 12 of the pkr suppressors had mutations in CPK1 that affects its PKA activities [22]. However, suppressor mutations in the remaining 55 suppressor strains of the pkr mutant remain to be identified.

To further characterize the cAMP-PKA pathway and suppressors of pkr in F. graminearum, in this study we identified mutations in FgSNT1, a core component of the Set3C HDAC complex, in 11 suppressor strains of the pkr mutant. All of these suppressor mutations resulted in the truncation of the C-terminal 275 or 98 (CT98) amino acids of FgSNT1. CT98 of FgSnt1 had self-inhibitory binding with its N-terminal region, which was relieved by the S443D or suppressor mutations. Targeted deletion of CT98 suppressed the defects of pkr in growth and H4 acetylation and increased the interaction of FgSnt1 with Hdf1, a HDAC of the Set3 complex. Taken together, these results indicated that FgSnt1 plays a key role in the functional relationship between the cAMP-PKA pathway and Set3 HDAC complex.

Results

Identification of suppressor mutations of pkr

In a previous study, 67 spontaneous suppressors of the pkr mutant were isolated from different culture plates and categorized into three groups based on their growth rate, including 12 of them with suppressor mutations in the CPK1 gene [22]. To identify suppressor mutations in the remaining suppressor strains that had no mutations in CPK1, 10 of them with various growth rates and colony morphology were selected for whole genome sequencing (WGS) analysis (S1 and S2 Figs) [22,30]. Mutations were identified by comparative analysis with the original pkr mutant used for suppressor isolation and updated PH-1 genome sequence [31] in 13 predicted genes in F. graminearum (Table 1). Two frameshift mutations that were caused by an insertion of T after C⁶⁰⁶⁵ and deletion of 7 base pairs (C⁶⁵⁹⁵GCTACC⁶⁶⁰¹) in FGRRES_00324 [17,32], an ortholog of yeast SNT1, were identified in two suppressor strains H11 and H17 (Table 1). Frameshift mutations in FGRRES_16648, an ortholog of yeast BLM10 also were identified in two suppressor strains. For the other genes, including orthologs of yeast PRE5, PRE6, CYC8, MIG1, NHP6, and SGF73, suppressor mutations were only identified in one suppressor strain each (Table 1). Interestingly, suppressor H11 had frameshift mutations in both FgSNT1 and FgBLM10.

Table 1. Mutations identified in the ORFs of predicted genes in suppressors of pkr.

Suppressor Strain	Predicted gene	Yeast ortholog	Mutations*
			DNA	Protein
Mutations identified by whole genome sequencing analysis
H4	FGRRES_07282	PRE6	G410 to A	D82N
H6	FGRRES_05222	PRE5	A325 to G	K62E
H10	FGRRES_16648	BLM10	ΔT1220-C3362	S392fs
H11	FGRRES_16648	BLM10	ΔG4185	G1380fs
	FGRRES_00324	SNT1	Insertion of a T after C6065	A1958fs
H17	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs
H22	FGRRES_12901	CYC8	G314 to A	G88D
	FGRRES_10588	No homolog	ORF deletion	Del
	FGRRES_10589	No homolog	ORF deletion	Del
	FGRRES_00887	No homolog	ORF deletion	Del
H28	FGRRES_03942	No homolog	G467 to T	R140M
H30	FGRRES_00385	NHP6A/6B	T281 to C	L57P
	FGRRES_04083	No homolog	Insertion of a T at G1209	M368fs
H34	FGRRES_05396	SGF73	G928 to A	R275K
H57	FGRRES_09715	MIG1	A211 to G	K71E
Mutations identified by amplifying and sequencing FgSNT1
H5	FGRRES_00324	SNT1	Insertion of a T after C6065	A1958fs
H7	FGRRES_00324	SNT1	Insertion of a T after C6065	A1958fs
H16	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs
H26	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs
H29	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs
H44	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs
H51	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs
H52	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs
H53	FGRRES_00324	SNT1	ΔC6595GCTACC6601	Y2135fs

* Mutations identified in the DNA sequence of predicted genes resulting in changes in amino acid sequences.

fs, frameshift mutant.

Del, deletion of the whole open reading frame

Snt1 is a core component of the Set3 HDAC complex that is well conserved from yeast to human and important for H3 and H4 acetylation [10,33,34]. Because the relationship between PKA signaling and Set3 HDAC is not well characterized, we selected FgSNT1 for further characterization in this study. First, we amplified and sequenced the FgSNT1 gene in the remaining suppressor strains and found nine of them had frameshift mutations (Table 1). All 11 spontaneous suppressor strains with mutations in FgSNT1 grew faster than the pkr mutant but they varied in growth rate and colony morphology (Fig 1A), which may be caused by additional mutations in the genome, such as the frameshift mutation in FgBLM10 in suppressor H11. When orthologs of yeast BLM10, PRE5, PRE6, CYC8, MIG1, and NHP6 were amplified and sequenced, suppressor H5 was found to have a non-sense mutation in FgBLM10 (S1 Table). For example, suppressors H5, H7, and H11 had the same insertion of a T after C⁶⁰⁶⁵ that will result in a frameshift and truncation of the C-terminal 275 amino acids (CT275, residues 1958–2233) but they differed in mutations in FgBLM10. Whereas H7 had no and H11 had a suppressor mutation, suppressor H5 had a nonsense mutation in FgBLM10 when it was amplified and sequenced (S1 Table). All other 8 suppressors had the same deletion of 7 base pairs (C⁶⁵⁹⁵GCTACC⁶⁶⁰¹) that will result a frameshift and truncation of the C-terminal 98 amino acids (CT98, residues 2135–2233). FgSnt1 has two conserved SANT DNA-binding domains in the middle region (Fig 1B) that are not affected by these suppressor mutations. In comparison with its orthologs from other fungi, the C-terminal 98 residues of FgSnt1 were conserved in Sordariomycetes, including Neurospora crassa and M. oryzae and this region is proline-rich and glycine-rich (Fig 1B).

Fig 1. Suppressors of pkr with mutations in FgSNT1 and structural features of FgSnt1.

(A). Three-day-old PDA cultures of the pkr mutant and marked suppressors with mutations in FgSnt1. (B). Schematic drawing of the FgSnt1 protein and alignment of its C-terminal 98-aa with orthologs from Magnaporthe oryzae (Mo), Neurospora crassa (Nc), F. oxysporum (Fo), and F. verticillioides (Fv). Stars mark the sites of suppressor mutations. SANT, SANT domain; CT98, the C-terminal 98-aa of FgSnt1.

Deletion of the C-terminal region of FgSnt1 rescues the defects of the pkr mutant

To verify the suppressive effects of truncations in FgSNT1 on the pkr mutant, we used the split marker approach to generate the FgSNT1^ΔCT98 gene replacement construct in which the C-terminal 98 amino acids were replaced with the geneticin-resistance cassette (S3 Fig). Three FgSNT1^ΔCT98 pkr mutant strains (Table 2) that were resistant to both geneticin and hygromycin were identified. They had identical phenotypes although only data for one of them, mutant SCP2, were presented.

Table 2. The wild-type and mutant strains of F. graminearum used in this study.

Strains	Brief description	References
PH-1	Wild type	[6]
P1	pkr deletion mutant of PH-1	[22]
C1M-1	cpk1 deletion mutant of PH-1	[23]
H1 to H67	Spontaneous suppressors of pkr mutant	[22]
S1	Fgsnt1 deletion mutant of PH-1	This study
S7	Fgsnt1 deletion mutant of PH-1	This study
S10	Fgsnt1 deletion mutant of PH-1	This study
SC3	FgSNT1ΔCT98 mutant of PH-1	This study
SC7	FgSNT1ΔCT98 mutant of PH-1	This study
SC11	FgSNT1ΔCT98 mutant of PH-1	This study
SP4	Fgsnt1 pkr mutant	This study
SP7	Fgsnt1 pkr mutant	This study
SP10	Fgsnt1 pkr mutant	This study
SP13	Fgsnt1 pkr mutant	This study
SCP2	FgSNT1ΔCT98 pkr mutant	This study
SCP5	FgSNT1ΔCT98 pkr mutant	This study
SCP12	FgSNT1ΔCT98 pkr mutant	This study
SDP1	FgSNT1S443D transformant of Fgsnt1 pkr	This study
SDP2	FgSNT1S443D transformant of Fgsnt1 pkr	This study
SD1	FgSNT1S443D transformant of PH-1	This study
SD3	FgSNT1S443D transformant of PH-1	This study

In comparison with the pkr mutant, the FgSNT1^ΔCT98 pkr mutant grew faster (Fig 2A) and produced fertile perithecia (Fig 2B). It also produced conidia with normal morphology (Fig 2C) and was normal in conidiation (Table 3). In infection assays with flowering wheat heads, the FgSNT1^ΔCT98 pkr mutant caused typical head blight symptoms (Fig 2D) and had a disease index of 7.7 (Table 3). The FgSNT1^ΔCT98 pkr mutant was able to form infection cushions on wheat lemma (Fig 2E) and spread to nearby spikelets (Fig 2F). These results indicate that truncation of the C-terminal region of FgSnt1 partially rescues the defects of the pkr mutant in growth, conidiation, and pathogenesis.

Fig 2. Deletion of the C-terminal region of FgSnt1 partially rescues the defects of pkr.

(A). Three-day-old PDA cultures of the wild-type strain PH-1 and the pkr, FgSNT1^ΔCT98 pkr and Fgsnt1 pkr mutants. (B). Perithecia from mating cultures of the same set of strains were examined at 12 days post-fertilization (dpf). Arrows point to ascospore cirrhi oozed out from black perithecia (C). Conidia of the same set of strains from 5-day-old CMC culture. Bar = 10 μm. (D). Wheat heads inoculated with the marked strains were examined for head blight symptoms at 14 days post-inoculation (dpi). Black dots mark the inoculated spikelets. (E). Infection cushions formed on wheat lemma were examined by SEM under ×850 at 2 dpi. Scale bars = 20 μm. (F). Thick sections of wheat rachises inoculated with the marked strains were examined for invasive hyphae at 5 dpi.

Table 3. Phenotypes of the Fgsnt1 mutant and its transformant strains in growth, conidiation, and plant infection.

Strain	Growth rate(mm/day)a,b	Conidiation (×104 conidia/ml)a,c	Disease Indexa,d
PH-1	10.5 ± 0.1a	235.8 ± 28.9a	10.9 ± 3.0a
pkr	2.3 ± 0.2c	7.5 ± 2.1e	0.0 ± 0.0d
Fgsnt1	10.0 ± 0.2a	76.3 ± 12.3cd	6.2 ± 1.1c
FgSNT1 ΔCT98	10.1 ± 0.1a	195.8 ± 26.5b	8.4 ± 2.8b
Fgsnt1 pkr	2.0 ± 0.4c	65.0 ± 16.0d	1.0 ± 0.0d
FgSNT1ΔCT98 pkr	10.0 ± 0.2a	192.5 ± 24.0b	7.7 ± 3.6bc
FgSNT1S443D pkr	4.6 ± 0.2b	90.0 ± 14.3c	6.3 ± 1.0c

^a Standard deviation (mean ± standard deviation) were calculated from at least three independent measurements. Different letters indicate significant differences based on ANOVA analysis followed by Duncan’s multiple range test (P = 0.05) in a, b, c, d, e.

^b Average daily extension of colony radium.

^c Conidiation was measured with 5-day-old Carboxymethylcellulose(CMC) culture.

^d Diseased spikelets per wheat head examined 14 dpi.

To determine whether deletion of the entire FgSNT1 can rescue the pkr mutant, we also used the gene replacement approach to generate the Fgsnt1 pkr double mutant. All four Fgsnt1 pkr mutant strains (Table 2) were similar to the pkr mutant in growth rate, conidiation, sexual reproduction, and virulence although only data for strain SP4 were presented (Fig 2). Therefore, truncation of its C-terminal region, but not deletion of the entire FgSNT1, is suppressive to the PKR deletion.

The C-terminal region of FgSnt1 is not essential for its functions in conidiation and plant infection

To determine the functions of FgSNT1 and its C-terminal region, we then transformed the FgSNT1 and FgSNT1^ΔCT98 gene replacement constructs into the wild-type strain PH-1, respectively. The resulting Fgsnt1 and FgSNT1^ΔCT98 gene replacement mutants had no significant changes in vegetative growth (Fig 3A, Table 3), conidium morphology (Fig 3B), and sexual reproduction (Fig 3C). However, conidiation was significantly reduced in the Fgsnt1 deletion mutant in comparison with the wild type (Table 3). For the FgSNT1^ΔCT98 mutant, it was only slightly reduced (less than 20% reduction) in conidiation compared to PH-1 but it produced over two-fold more conidia than the Fgsnt1 mutant (Table 3).

Fig 3. Assays for the phenotypes of the Fgsnt1 and FgSNT1^ΔCT98 mutants.

(A). Three-day-old PDA cultures of the wild type (PH-1) and the Fgsnt1 and FgSNT1^ΔCT98 mutants. (B). Conidia from 5-day-old CMC cultures were examined for morphological defects. Bar = 10 μm. (C). Perithecia from mating cultures of the marked strains were examined at 12 dpf. (D). Wheat heads inoculated with the marked strains were examined for head blight symptoms at 14 dpi. Black dots mark the inoculated spikelets.

In infection assays with wheat heads, the Fgsnt1 and FgSNT1^ΔCT98 deletion mutants were still pathogenic and caused typical head blight symptoms on inoculated spikelets (Fig 3D). However, both of them were reduced in virulence in comparison with PH-1. On average, the disease index was 10.9, 6.2, and 8.4, respectively, for the wild type, Fgsnt1, and FgSNT1^ΔCT98 strains (Table 3). Therefore, the Fgsnt1 and FgSNT1^ΔCT98 deletion mutants had approximately 43% and 23% reduction in virulence. These results indicate that FgSNT1 is important for conidiation and plant infection. Because deletion of CT98 had much less effects on conidiation and virulence than deletion of the entire gene, the C-terminal region of FgSnt1 likely has a minor role in conidiation and pathogenicity but is not essential for its normal functions in F. graminearum.

CT98 of FgSnt1 negatively regulates histone H4 acetylation in the pkr mutant

Because yeast Snt1 is a conserved component of the Set3 HDAC complex, we then assayed global histone acetylation levels with selected anti-H3ac and anti-H4ac antibodies by western blot analysis (Fig 4A). In comparison with the wild type, the pkr mutants were significantly reduced in H4 acetylation, but had no significant differences in H3 acetylation compare to the wild type (Fig 4B) (S4A and S4B Figs). The FgSNT1^ΔCT98 and Fgsnt1 mutants had similar H3 acetylation levels but were slightly reduced in H4 acetylation compared to the wild-type strain (Fig 4A). The FgSNT1^ΔCT98 pkr double mutant had significantly higher H4 acetylation level than the pkr mutant although it was still slightly reduced in comparison with PH-1 (Fig 4). However, the Fgsnt1 pkr double mutant had a similar H4 acetylation level with the pkr mutant (S4C Fig). These results indicate that deletion of CT98, but not the entire FgSNT1, partially recovered the reduction of the pkr mutant in H4 acetylation.

Fig 4. Deletion of the C-terminal region of FgSnt1 increases H4 acetylation in the pkr mutant.

(A). Western blots of total proteins isolated from PH-1 and the pkr, FgSNT1^ΔCT98 pkr, Fgsnt1, FgSNT1^ΔCT98, and hdf1 mutants were detected with antibodies specific for H3ac, H4ac, and H4K16ac. Detection with anti-H3 and anti-H4 antibodies was used as a loading control. (B). Quantitative analysis of the histone acetylation level by calculating the ratio of the quantified acetylated and non-acetylated H3 and H4 proteins.

In yeast, Hos2, a HDAC in the Set3 complex and an ortholog of F. graminearum Hdf1, is important for H4K16 deacetylation [35]. In comparison with the wild type, the levels of H4 and H4K16 acetylation were increased 38% and 39%, respectively, in the hdf1 mutant (Fig 4). In contrast, the pkr deletion mutant was reduced over 74% in H4K16 acetylation compared to PH-1. The FgSNT1^ΔCT98 pkr mutants had a higher level of H4K16 acetylation than PH-1 or pkr but lower than the hdf1 mutant (Fig 4). These results showed that the defect of the pkr mutant in H4 acetylation is suppressed by deletion of the C-terminal region of FgSnt1. Because of the conserved role of Pkr regulatory subunit of PKA in cAMP signaling, it is likely that the Set3 HDAC complex is regulated by the cAMP-PKA pathway via FgSnt1 in F. graminearum. In the absence of PKR, the C-terminal region of FgSnt1 may be suppressive to H4 acetylation by interfering with the Hdf1 HDAC activity. Deletion of the C-terminal 98 residues may bypass the requirement of normal cAMP-PKA signaling for regulating activities of the Set3 HDAC complex and the expression of subsets of genes affected by H4 acetylation.

The C-terminal region of FgSnt1 plays a negative role in its interaction with Hdf1

To determine the interaction of FgSnt1 with Hdf1 by yeast two-hybrid assays, we generated the bait constructs with FgSnt1 and FgSnt1^ΔCT98 and co-transformed them with the prey constructs of Hdf1 into yeast strains AH109. Yeast transformants expressing the Hdf1 prey and FgSnt1^ΔCT98 bait, but not FgSnt1 bait, constructs were able to grow on SD-Trp-Leu-His (synthetic defined (SD) medium without Trp, Leu, His) medium and had LacZ activity (Fig 5A). These yeast two-hybrid assay results suggested that Hdf1 directly interacts with the FgSnt1^ΔCT98 but not with full-length FgSnt1, at least in yeast. It is likely that CT98 functions as a negative regulator of the FgSnt1-Hdf1 interaction, which may be stimulatory to Hdf1 HDAC activity.

Fig 5. Assays for the interaction of Hdf1 with FgSnt1 and FgSnt1^ΔCT98.

(A). Yeast two-hybrid assays for the interaction of Hdf1 (prey) with FgSnt1 or FgSnt1^ΔCT98 (bait). The positive and negative controls were from the Matchmaker kit. (B). Immunoprecipitation (IP) assays for the interaction between Hdf1 with FgSnt1^ΔCT98 and FgSnt1. Western blots of proteins isolated from E. coli cells expressing Hdf1-HIS together with FgSnt1-GST or FgSnt1^ΔCT98-GST (Input) or proteins eluted from anti-GST beads (GST IP) were detected with the marked anti-GST and anti-HIS antibodies. Proteins from E. coli cells expressing the empty GST vector pGEX4T-1 were used as the control.

To confirm the importance of FgSnt1^CT98 in its interaction with Hdf1, we isolated the FgSnt1-GST, FgSnt1^ΔCT98-GST, and Hdf1-HIS recombinant proteins expressed in Escherichia coli. For GST pull down assays, Hdf1-HIS proteins were mixed with FgSnt1-GST or FgSnt1^ΔCT98-GST proteins and anti-GST agarose beads. On western blots with proteins eluted from anti-GST agarose beads, the Hdf1-HIS band of expected size was detected by an anti-HIS antibody only when Hdf1-HIS proteins were co-incubated with FgSnt1^ΔCT98-GST but not with FgSnt1-GST proteins (Fig 5B). These results from in vitro assays further suggest an important role of CT98 in the interaction of FgSnt1 with Hdf1.

Pkr is important for protecting Cpk1 from degradation by 26S proteosome

To determine the effects of truncation mutations in FgSNT1 on PKA activities in the pkr mutant, total proteins were isolated from PH-1, cpk1, pkr, and FgSNT1^ΔCT98 pkr and assayed for PKA activities [36]. In comparison with the wild type, the cpk1 mutant had no detectable PKA activities (Fig 6A). To our surprise, the pkr and FgSNT1^ΔCT98 pkr mutants also lacked detectable PKA activities (Fig 6A). Because CPK1 expression was found to be significantly upregulated in the pkr mutant in a previous study [22], it is likely that the catalytic subunits of PKA are unstable or degraded in the absence of Pkr regulatory subunits. To test this hypothesis, we generated an anti-Cpk1 antibody with the synthetic oligopeptide VKAGAGDASQFDRYPE. In proteins isolated from vegetative hyphae, the resulting antibody detected a 67-kDa band of expected Cpk1 size and two minor bands in PH-1 but not in the cpk1 mutant (Fig 6B), indicating that this anti-Cpk1 antibody is suitable for Cpk1 detection. The two minor bands may be related to post-translational modifications of Cpk1 proteins such as ubiquitination. In the pkr mutant and FgSNT1^ΔCT98 pkr mutant, the 67-kDa Cpk1 band as well as the two minor bands were not detected with the anti-Cpk1 antibody (Fig 6A, 6B), indicating that Pkr is necessary for the stability of Cpk1.

Fig 6. Assays for PKA activities and Cpk1 expression.

(A). PKA activity was assayed with proteins isolated from hyphae of PH-1, cpk1 and pkr mutant, and FgSNT1^ΔCT98 pkr mutant without or with 50 μM MG132 treatment. (B). Western blots of total proteins isolated from the marked strains and MG132 treatment were detected with an anti-Cpk1 antibody. The expected size of Cpk1 is 67 kDa. Detection with an anti-tubulin antibody was used as the loading control. (C). Western blots of total proteins isolated from PH-1 and the pkr, PH-1 treatment with MG132, and pkr treatment with MG132 mutants were detected with antibodies for H4ac. Detection with anti-H4 antibodies was used as a loading control. (D). Quantitative analysis of the histone acetylation level by calculating the ratio of the quantified acetylated and non-acetylated H4 proteins.

Because suppressor mutations were identified in orthologs of yeast BLM10, PRE5, and PRE6 that encode key components of the 26S proteasome (Table 1), we then assayed the effects of MG132, an 26S proteasome inhibitor, on Cpk1 expression. When treated with 50 μM MG132, the Cpk1 band and PKA activities were detected in the pkr and FgSNT1^ΔCT98 pkr mutants (Fig 6A and 6B), indicating that Cpk1 proteins may be degraded by the 26S proteasome in the pkr mutant. In addition, the suppressor H10 with frameshift mutation on FgBlm10 also rescued the Cpk1 stability and PKA activities (S5 Fig). These results indicate that, in the absence of Pkr that is necessary to form stable PKA holoenzymes, Cpk1 proteins are degraded in the pkr mutant, which can be suppressed by inhibition of 26S proteasome and genetic mutations in its key components. H4 acetylation of the pkr mutant was significantly reduced. When treated with 50 μM MG132, the H4 acetylation of pkr partially rescued, and showed reduced significant differences compare to the wild type (Fig 6C and 6D), indicating that the inhibition of Cpk1 degradation rescues the reduction in H4 acetylation of pkr mutant.

FgSnt1 interacts with Cpk1 but not Pkr

To explore the interaction between PKA and FgSnt1, we generated the bait constructs of FgSnt1 and FgSnt1^ΔCT98 and co-transformed them with the prey constructs of Pkr or Cpk1. In yeast two-hybrid assays, Pkr had no direct interaction with FgSnt1^ΔCT98 or full-length FgSnt1 (Figs 7A and S6). However, both FgSnt1 and FgSnt1^ΔCT98 interacted with Cpk1 in yeast two-hybrid assays (Figs 7B and S6).

Fig 7. Assays for the interaction of Cpk1 with FgSnt1.

(A). Yeast two-hybrid assays for the interaction of Pkr (prey) with FgSnt1 or FgSnt1^ΔCT98 (bait). The positive and negative controls were from the Matchmaker kit. (B). Yeast two-hybrid assays for the interaction of Cpk1 (prey) with FgSnt1 or FgSnt1^ΔCT98 (bait). (C). Yeast two-hybrid assays for the interaction of Cpk1with FgSnt1^1-744, FgSnt1^745-1490, or FgSnt1^1491-2233. (D). Immunoprecipitation (IP) assays for the interaction between Cpk1 with FgSnt1^1-744. Western blots of proteins isolated from E. coli cells expressing Cpk1-HIS together with FgSnt1^1-744-GST (Input) or proteins eluted from anti-GST beads (GST IP) were detected with the marked anti-GST and anti-HIS antibodies. Proteins from E. coli cells expressing the empty GST vector pGEX4T-1 were used as the control. (E). The ITC thermogram of binding interactions between Cpk1 and FgSnt1^1-744. The upper panel show raw data obtained from 10 μl injections of Cpk1 to FgSnt1^1-744. The lower panel display plots of integrated total energy exchanged (as kcal/mol of injected compound) as a function of molar ratio of Cpk1 to FgSnt1^1-744(dots). Solid lines indicate the fit to a single-site saturation model.

To determine which region of FgSnt1 interacts with Cpk1, we generated bait constructs of FgSnt1 with the N-terminal 1–744 aa, middle 745–1490 aa, and C-terminal 1491–2233 aa. In yeast two-hybrid assays, FgSnt1^1-744 and FgSnt1^745-1490 but not FgSnt1^1491-2233 interacted with Cpk1 (Fig 7C). FgSnt1^1-744 appeared to have a stronger interaction with Cpk1 than FgSnt1^745-1490 (Figs 7C and S6).

To confirm their interactions, we detected the interaction of FgSnt1^1-744-GST and Cpk1-HIS with GST pull down assays. The in vitro result showed that the Cpk1-HIS band of expected size was detected by an anti-HIS antibody in samples mixing with FgSnt1^1-744-GST (Fig 7D). The isothermal titration calorimetry (ITC) assay [37] was also used to measures the amount of heat absorbed or released as Cpk1-HIS is titrated into FgSnt1^1-744-GST (Figs 7E and S7). The dissociation constant (Kd), stoichiometry of the ligand-to-protein binding (n), enthalpy change (ΔH), and entropy change (ΔS) were determined (Fig 7E). Theoretical fits to FgSnt1^1-744-Cpk1 association data were obtained using a single binding site model, confirming the interaction of FgSnt1^1-744-GST and Cpk1-HIS.

FgSnt1 is a direct phosphorylation target of Cpk1

To test whether FgSnt1 is phosphorylated by Cpk1, we conducted in vitro phosphorylation assays with purified recombinant FgSnt1^1-744-GST and Cpk1-HIS proteins [38,39]. On Phos-tag SDS gels, FgSnt1^1-744-GST proteins incubated with Cpk1 and ATP shifted to a slowly migrating band, suggesting that FgSnt1^1-744 was phosphorylated by Cpk1 (Fig 8A). In the control without Cpk1, gel mobility shift of the FgSnt1^1-744-GST band was not observed (Fig 8A).

Fig 8. Assays for the phosphorylation of FgSnt1^1-744 by Cpk1.

(A). In vitro kinase assays with FgSnt1^1-744-GST and Cpk1-HIS recombinant proteins. Blots of SDS-PAGE or 20μM Mn²⁺-Phos-tag SDS-PAGE gels were detected with an anti-GST antibody for the FgSnt1^1-744-GST bands. (B). In vitro kinase assays with FgSnt1^S443A-GST and Cpk1-HIS fusion proteins. The S443 to A mutation in FgSnt1^1-744-GST eliminated its phosphorylation by Cpk1.

Primary sequence analysis predicts the presence of five PKA phosphorylation sites in the N-terminal region of FgSnt1 (S18, S38, S322, S443, and S511) (S3 Table), which fit the PKA phosphorylation motif [R][R][X][S/T] (S8 Fig). To test which sites are phosphorylated by Cpk1, we generated the FgSnt1^S18A-, FgSnt1^S38A-, FgSnt1^S322A-, FgSnt1^S443A-, and FgSnt1^S511A-GST constructs by replacing the respective serine with alanine [40] in the FgSnt1^1-744-GST construct, and isolated these recombinant proteins expressed in E. coli. In the in vitro kinase assays, Cpk1 failed to phosphorylate the FgSnt1^S443A-GST protein (Fig 8B), indicating that S443 is the critical phosphorylate site (Fig 8). However, unlike the FgSnt1^S443A -GST protein, the remaining recombinant proteins FgSnt1^S18A-, FgSnt1^S38A-, FgSnt1^S322A-, and FgSnt1^S511A-GST were still phosphorylated by Cpk1, indicating that phosphorylation of FgSnt1^1-744 by Cpk1 is not affected by the S18A, S38A, S322A, and S511A mutations (Figs 8B and S9). Therefore, Cpk1 specifically phosphorylated FgSnt1^1-744 at S443. Sequence alignments showed that only S443 of these phosphorylation sites is conserved among FgSnt1 orthologs from filamentous ascomycetes (S8 Fig), suggesting its phosphorylation by PKA may be also conserved with important function in other ascomycetes.

S443D mutation in FgSnt1 partially rescues defects of pkr mutant

To determine the role of S443 phosphorylation, we introduced an S443D mutation by overlapping PCR and transformed the resulting FgSNT1^S443D gene replacement construct into the Fgsnt1 pkr mutant. Transformants resistant to geneticin and 5-fluorodeoxyuridine were screened by PCR with primers specific for the mutant alleles. The FgSNT1^S443D pkr transformants (Table 2) were further confirmed by sequencing analysis for the S443D mutation. The FgSNT1^S443D pkr transformants grew faster than the pkr mutant although it still grew slower than the wild type and FgSNT1^ΔCT98 pkr (Fig 9A, Table 3). These results indicate that the S443D mutation in FgSnt1 partially rescued the growth defect of pkr mutant. Because the FgSNT1^S443D pkr transformants still grew slower than FgSNT1^ΔCT98 pkr, it is likely that other putative phosphorylation sites (Table 3, S9 Fig) may also contribute to the functional relationship between FgSnt1 and Pkr.

Fig 9. The S443D mutation in FgSNT1 partially rescues defects of the pkr mutant.

(A). Three-day-old PDA cultures of PH-1 and the FgSNT1^S443D pkr, FgSNT1^ΔCT98 pkr and pkr mutants. (B). Conidia of the same set of strains from 5-day-old CMC culture. Bar = 10 μm. (C). Perithecia from mating cultures of the marked strains were examined at 12 dpf. (D). Wheat heads inoculated were examined for head blight symptoms at 14 dpi. Black dots mark the inoculated spikelets.

In comparison with the pkr mutant, conidium morphology was normal (Fig 9B) and conidiation was increased in the FgSNT1^S443D pkr transformant although it still produced fewer conidia than the wild type. On self-mating carrot agar plates, FgSNT1^S443D pkr produced abundant perithecia and ascospore cirrhi (Fig 9C). In infection assays, the FgSNT1^S443D pkr mutant caused typical head blight symptoms on inoculated wheat heads and had a disease index of 6.3, which is similar to that of FgSNT1^ΔCT98 pkr and approximately 50% of that of the wild type (Fig 9D). These data indicate that the S443D mutation in FgSnt1 also partially rescued the defect of the pkr mutant in plant infection but fully rescued its defect in conidium morphology and sexual reproduction.

Expressing the FgSNT1^S443D allele and exogenous cAMP treatment increase H4 acetylation

Because FgSNT1^S443D partially rescued the defects of the pkr mutant, it is possible that FgSnt1 is phosphorylated by PKA to regulate HDAC activities of Hdf1 or the Set3 complex as a whole. When assayed with the anti-H4ac antibody, the FgSNT1^S443D transformant had increased H4 acetylation in comparison with the wild type (Fig 10A). These results suggest that expression of the dominant active FgSNT1^S443D allele may reduce the Set3 or Hdf1 HDAC activities in F. graminearum.

Fig 10. Phosphorylation of FgSnt1 ^S443 is important for H4 acetylation.

(A). Western blots of total proteins isolated from PH-1 (WT) and its transformant expressing FgSNT1^S443D (S443D) were detected with anti-H4 and anti-H4ac antibodies (left) and quantified for the band intensities with Image J to estimate the relative H4 acetylation level after normalization with the non-acetylated H4 band. Data from three replicates were used to estimate the error bar. (B). Western blots of proteins isolated from regular PH-1 cultures (CK) and cultures treated with 50 μM cAMP (cAMP) or 50 μM H89 (H89) were detected with anti-H4 and anti-H4ac antibodies and quantified for the band intensities with Image J to estimate the relative H4 acetylation level after normalization with the non-acetylated H4 band. Data from three replicates were used to estimate the error bar.

To further characterize the relationship between PKA activity and H4 acetylation, we treated PH-1 hyphae with 50 μM cAMP and 50 μM H89, a PKA inhibitor [41]. H4 acetylation was significantly increased by exogenous cAMP (Fig 10B). In contrast, treatments with H89 reduced H4 acetylation (Fig 10B). Taken together, overstimulating and inhibiting the cAMP-PKA pathway appear to have opposite effects on H4 acetylation, likely by affecting the phosphorylation of FgSnt1 and Set3/Hdf1 HDAC activities in F. graminearum.

The S443D, but not S443A, mutation affects the interaction between the N-terminal and C-terminal regions of FgSnt1

Because deletion of CT98 region and expression of the FgSNT1^S443D allele both partially rescued the defects of the pkr mutant, it is likely that phosphorylation of S443 by PKA has similar effects as deletion of CT98 on the function of FgSnt1. One possible explanation is that the CT98 region may physically interact with its upstream part of FgSnt1 and phosphorylation at S443 releases this self-inhibitory binding and enables its interaction with the Hdf1 HDAC. To test this hypothesis, we generated prey constructs of FgSnt1^N447 (residues 1–447), FgSnt1^SANT (SANT domain, residues 448–1502), and FgSnt1^M633 (residues 1503–2136) and transformed them in pairs with the bait construct of FgSnt1^CT98 in to yeast strain AH109. The resulting yeast transformants expressing the FgSnt1^CT98 bait and FgSnt1^N447, FgSnt1^SANT, or FgSnt1^M633 prey constructs failed to grow on SD-Trp-Leu-His plates and lacked LacZ activities (Figs 11A and S10).

Fig 11. Assays for intramolecular interactions in FgSnt1 and the effect of S443D mutation.

(A). Yeast two-hybrid assays for the intramolecular interactions in FgSnt1. Yeast cells expressing the marked bait (BD) and prey (AD) constructs were assayed for growth on SD-Trp-Leu (left), SD-Trp-Leu-His plates (middle), and LacZ activities (right). N447, SANT, M633, and CT98 are fragments containing residues 1–447, 448–1502 (SANT domains), 1503–2136, and 2135–2233 of FgSnt1, respectively. FgSnt1^S443D and FgSnt1^S443A have the S443D and S443A mutations in N447. The positive and negative controls are from the Matchmaker kit. (B). IP assays for the interaction of FgSnt1^CT98 with FgSnt1^N447 and FgSnt1^S443D. Blots of the mixtures of marked recombinant proteins (input) and proteins eluted from eluted from anti-GST beads (GST IP) were detected with the anti-GST and anti-HIS antibodies. Proteins from E. coli cells expressing the empty GST vector pGEX4T-1 were used as the control.

To determine the effect of S443 phosphorylation, we then introduced the S443D and S443A mutations into the prey construct of FgSNT1^N447. The resulting FgSnt1^S443D and FgSnt1^S443A constructs were co-transformed with FgSnt1^CT98 into yeast strain AH109. Yeast transformants expressing the FgSnt1^CT98 bait and FgSnt1^S443D prey constructs grew on SD-Trp-Leu-His plates and had LacZ activities. However, yeast transformants expressing the FgSnt1^CT98 bait and FgSnt1^S443A prey constructs failed to grow on SD-Trp-Leu-His plates (Fig 11A), indicating the importance of S443D mutation on the interaction between FgSnt1^N447 and FgSnt1^CT98. To confirm this observation, we generated the FgSnt1^N447-HIS, and FgSnt1^S443D-HIS and FgSnt1^CT98-GST fusion proteins. In GST pull down assays, interactions also were only observed between FgSnt1^S443D-HIS and FgSnt1^CT98-GST (Fig 11B). These results suggest that the C-terminal tail region of FgSnt1 specifically interacts with its N-terminal region after phosphorylation of S443 by PKA, which may affect the Set3 complex and Hdf1 HDAC activities.

C-terminal region deletion and S443D mutation of FgSNT1 rescues the genes expression in the pkr mutant though H4 acetylation

Because histone acetylation is associated with gene expression, we used the RNA-seq approach to identify genes affected by deletion of PKR and mutation in FgSNT1. RNA samples were isolated from hyphae of the wild type, pkr, FgSNT1^S443D pkr, and FgSNT1^ΔCT98 pkr mutants harvested from YEPD cultures at 12 h. In comparison with the wild type, 638 differentially expressed genes (DEGs) were down-regulated over 2-fold in the pkr mutant (S4 Table). Among these DEGs, 270 (27.9%) had their expression increased to the wild-type levels in both FgSNT1^S443D pkr, and FgSNT1^ΔCT98 pkr mutants (S4 Table, Fig 12A). GO enrichment analysis showed that those 270 genes were enriched for genes involved in integral component of membrane, carbohydrate transport, metabolic process, and cellular component disassembly (Fig 12B) and 204 of them are in the H3K27me3-enriched chromosomal regions [8]. Interestingly, CPK1 and the PTH11 homolog (FGRRES_03897) are among these DEGs. In M. oryzae, PTH11 encodes a G-protein coupled receptor (GPCR) that functions upstream the cAMP-PKA pathway for regulating appressorium formation [42] FGRRES_03897, FGRRES_05294, and FGRRES_08448 were selected for verification by qRT-PCR assays and confirmed to be down-regulated in the pkr mutant and up-regulated in FgSNT1^S443D pkr, and FgSNT1^ΔCT98 pkr transformants (Fig 12C). FGRRES_05294 is predicted to encode an arrestin, which is involved in regulation signal transduction related to GPCRs [43] and FGRRES_08448 encodes an arrestin-domain containing protein.

Fig 12. RNA-seq and ChIP-seq analyses with the pkr1 mutant and its transformants expressing FgSNT1^ΔCT98 and FgSNT1^S443D.

(A). A Venn diagram showing the numbers of genes down-regulated (left panel) in the pkr, FgSNT1^ΔCT98 pkr, and FgSNT1^S443D pkr mutants. (B). GO enrichment analysis of genes down-regulated in the pkr mutant but recovered in expression in the FgSNT1^ΔCT98 pkr and FgSNT1^S443D pkr mutants. (C). The expression levels of FGRRES_08448, FGRRES_03897, and FGRRES_05294 in PH-1 (arbitrarily set to 1) and the pkr mutant were assayed by qRT-PCR. (D). The abundance of the promoters of marked genes in the input samples (no ChIP) and chromatin fragments immunoprecipitated with the anti-H4ac antibody in ChIP assays with PH-1 and the pkr mutant was assayed by qPCR. Relative abundance (y-axis) is expressed as the percentage of abundance in input samples in immunoprecipitated samples. Different letters (a, b) indicate significant differences based on ANOVA analysis followed by Duncan’s multiple range test (P = 0.05).

To determine the regulatory role of H4 acetylation on the expression of FGRRES_08448, FGRRES_03897, and FGRRES_05294, we isolated chromatin fragments of PH-1 and the pkr mutant for chromatin precipitation (ChIP) with an anti-H4ac antibody as described [44,45]. Genomic DNAs extracted from chromatin fragments isolated by ChIP or without ChIP (input control) were then used for qPCR assays with primers amplifying the promoter sequences of these three target genes. In comparison with the input control, ChIP with the anti-H4ac antibody enriched the promoter of FGRRES_08448 by 0.91% in PH-1 but only 0.43% in the pkr mutant (Fig 12D). Similarly, the level of enrichment for the promoters of FGRRES_03897 and FGRRES_05294 by ChIP with the anti-H4ac antibody was lower in the pkr mutant than in PH-1 (Fig 12D). These results suggested that changes in the expression level of FGRRES_08448, FGRRES_03897, and FGRRES_05294 are associated with H4 acetylation, which is affected by deletion of PKR in F. graminearum.

Discussion

Like in other fungal pathogens, the cAMP-PKA pathway is involved in regulating hyphal growth, conidiation, ascosporogenesis, and plant infection in F. graminearum and mutants deleted of its key components have severe defects in growth, pathogenesis, and reproduction. [22,23]. An earlier study showed that 11 of the 67 spontaneous suppressors of pkr had mutations in CPK1 [23]. In this study, we found that Cpk1 proteins were not detectable in the pkr mutant. However, treatments with MG132 inhibited Cpk1 degradation. Furthermore, suppressor mutations in FgBLM10 were identified in 11 suppressor strains, and all of them were nonsense or frameshift mutations. Missense suppressor mutations in FgPRE5 and FgPRE6 also were identified in two suppressor strains. In yeast, PRE5, PRE6, and BLM10 encode important components of the 26S proteasome. Therefore, it is likely that binding with Pkr regulatory subunits to form inactive holoenzymes is important for protecting Cpk1 proteins from degradation by 26S proteasome in the wild type under normal growth conditions. In the pkr mutant, Cpk1 proteins are degraded by 26S proteasome, which may explain similar phenotypes between the pkr and cpk1 cpk2 mutants [22,23]. Suppressor mutations in FgPRE5, FgPRE6, and FgBLM10 may negatively impact the activity of 26S proteasome, preventing the degradation of Cpk1 proteins in the absence of Pkr.

Like the pkr mutant, the cpk1 cpk2 mutant also often produces spontaneous suppressors with fast growth rate [26,27]. However, unlike 11 suppressors of pkr with mutations in CPK1, none of the 30 suppressors of cpk1 cpk2 sequenced have mutations in PKR. Instead, 29 of them have mutations in the transcription factor gene orthologous to Sfl1, which is one of the downstream transcription factors of the cAMP-PKA pathway in S. cerevisiae [29]. In M. oryzae, the association of MoSfl1 with the Cyc8-Tup1 transcriptional co-repressor is relieved by phosphorylation in the wild type but disrupted by suppressor mutations in MoSFL1 the cpk1 cpk2 mutant [26]. Interestingly, one of the suppressor strains sequenced in this study had a G88D mutation in FgCYC8 (Table 1), which may impact the expression of genes important for hyphal growth regulated by the Cyc8-Tup1 co-repressor. In addition, we identified one suppressor of pkr with a missense mutation in FGRRES_09715, an ortholog of yeast Mig1 (Table 1). In yeast, Mig1 is another one of the downstream transcription factors of PKA that is functionally related to the Cyc8-Tup1 co-repressor for glucose repression [46]. Therefore, suppressor mutations in the direct targets of the cAMP-PKA pathway or Cyc8-Tub1 transcriptional co-repressor may also rescue the growth defect of the pkr mutant in F. graminearum.

Among the 11 pkr suppressors with truncation mutations in FgSNT1, eight of them had the deletion of C⁶⁵⁹⁵GCTACC⁶⁶⁰¹ and the other three had an insertion of T after C⁶⁰⁶⁵. The independent recovery of multiple suppressor strains with the same mutations indicates that these are likely hot spots for deletion and insertion mutations. Unfortunately, the flanking DNA sequences of the C⁶⁵⁹⁵GCTACC⁶⁶⁰¹ and C6065 sites have no distinct sequence features that may cause the deletion or insertion except that both of them are in GC-rich regions. Frameshift mutations in these suppressors result in the truncation of C-terminal region of FgSnt1 but do not affect its SANT domains. In this study, we verified that the truncation of C-terminal 98 residues (CT98) caused by the insertion of 7 bp is suppressive to the pkr mutant. For the insertion of T that results in the truncation of C-terminal 275 residues (CT275), we did not experimentally verify its suppressive effects on pkr because CT98 is a part of CT275.

Snt1 is a key component of the Set3 HDAC complex that is conserved from yeast to human [10,47]. In F. graminearum, truncation of the C-terminal region of FgSnt1 rescued the defect of pkr mutant in H4 acetylation. Furthermore, the Hdf1 HDAC interacted with FgSNT1^ΔCT98 but not with FgSnt1 in yeast two-hybrid and GST pull down assays, indicating that FgSnt1 is also a component of the Set3 complex in F. graminearum. However, the snt1 deletion mutant has similar defects with mutants deleted of orthologs of yeast SIF2 and HOS2 in M. oryzae. In F. graminearum, deletion of FgSNT1 had no significant effect on vegetative growth and only a minor effect on virulence, which is different from the mutants deleted of the SIF2 and HOS2 ortholog [9,12]. Therefore, the SNT1 orthologs appear to vary in their importance in growth, differentiation, and pathogenesis between M. oryzae and F. graminearum, which may be also true to other filamentous fungi or fungal pathogens. Unfortunately, there are only limited studies with SNT1 orthologs although they are well conserved in fungi. In the fission yeast Schizosaccharomyces pombe, Snt1 plays a role in promoting the successful completion of cytokinesis [48]. In the human pathogen Candida albicans, Snt1 is important for biofilm formation [33].

Whereas yeast Snt1 is shorter and lacks this tail region, FgSnt1 orthologs from Sordariomycetes, N. crassa and M. oryzae, and other Fusarium spp. were conserved in the C-terminal 98 amino acids. We showed that the C-terminal 98 aa of FgSnt1 (FgSnt1^CT98), although not essential for its function, interacted with its N-terminal region (N-terminal 447 aa, FgSnt1^N447). This C-terminal tail region lacks any known protein motifs but contains intrinsically disordered regions (IDRs) based on analysis with the PONDR (Predictor of Natural Disordered Regions) program [49]. The C-terminal region of FgSnt1 is proline-rich and glycine-rich (Fig 1B). In comparison with FgSnt1, its orthologs in N. crassa and M. oryzae have a longer C-terminal region due to the insertion of multiple glutamine residues. The glutamine-rich region may affect the folding of intrinsically disordered proteins [50] but its role in FgSnt1 is unknown.

The cAMP-PKA signal transduction pathway and Set3 HDAC complex both are well conserved in eukaryotic organisms. Although the underlying mechanisms are not clear, cross-interactions between PKA and Set3 complex have been reported in fungal pathogens. In Ustilago maydis, Hos2 act as a downstream component of the cAMP-PKA pathway and a transcriptional regulator of the mating-type genes required for the dimorphic switch and pathogenesis [51]. The Set3 complex functions as a key negative regulator of PKA signaling for morphogenesis and virulence in C. albicans [52]. In this study, we found that Cpk1 interacted with the N-terminal 744 aa of FgSnt1 in F. graminearum, which shows the physical interaction between a catalytic subunit of PKA with a key component of the Set3 HDAC complex. Furthermore, S443, a conserved PKA phosphorylation site in this region was found be phosphorylated by Cpk1, showing the phosphorylation of a Set3 complex component by PKA. To our knowledge, phosphorylation of Snt1 by PKA and their direct interaction have not been reported in other organisms. It is likely that the physical interaction and functional relationship between PKA and FgSnt1 are conserved in other filamentous fungi.

In F. graminearum, expression of the phosphorylation-mimic FgSNT1^S443D allele, like truncation of the C-terminal region of FgSnt1, were suppressive to the pkr mutant. In yeast two-hybrid assays and GST pull down assays, FgSnt1^CT98 only interacted with FgSnt1^N447 with the S443D mutation. The S443D mutation in FgSNT1 also affected its interaction with Hdf1. Therefore, PKA may play a direct role in regulating the interaction between the N-terminal and C-terminal regions of FgSnt1 by phosphorylation at S443. The phosphorylation of S443 by PKA may result in the self-inhibitory binding of FgSnt1 via its two terminal regions, which may affect its interaction with Hdf1, Set3 HDAC activities, and H4 acetylation. Interestingly, Set3 was identified as potential PKA phosphorylation targets in C. albicans by phosphoproteomic and bioinformatic analyses [53]. It is possible that FgSet3 and other components of the FgSet3 complex may also be phosphorylated by PKA in F. graminearum.

Although the S443D mutation in FgSNT1 was suppressive to the pkr mutant, the FgSNT1^S443D pkr transformant grew slower and produced fewer conidia in comparison with the FgSNT1^ΔCT98 pkr mutant, suggesting that the S443D mutation has a weaker suppressive effect than truncation of CT98. One possible explanation is that FgSnt1 may have other phosphorylation sites that may be phosphorylated by PKA [54]. Because Cpk1 also interacted with FgSnt1^745-1490, additional PKA phosphorylate site, for example S1325 which also fit the PKA phosphorylation motif, may be present between residues 745 and 1490. In yeast, treatments with nocodazole induce the phosphorylation of Snt1 by CDK1, which affects its conformational shifts and roles in establishing activated chromatin states [51]. FgSnt1 may be also subjected to phosphorylation by other protein kinases in regulating genes specific for different developmental and infection stages in F. graminearum.

A total of 638 genes were downregulated over two-fold in the pkr mutant. For 79% and 35% of these genes, their expression was increased to the wild-type level in the FgSNT1^ΔCT98 pkr and FgSNT1^S443D pkr transformants, respectively. The number of genes with recovered expression levels by the ΔCT98 and S443D mutations in FgSNT1 in the pkr mutant background are consistent with the degree of recovery in growth rate in these two transformants. Nevertheless, the expression of 79 DEGs downregulated in the pkr mutant was not rescued by the ΔCT98 and S443D mutations in FgSNT1, suggesting that Pkr and FgSnt1 play distinct roles in regulating their expression. Furthermore, the level of H4 acetylation known to be correlated with gene expression [55,56] was significantly reduced in the pkr mutant but recovered in the FgSNT1^ΔCT98 pkr transformant. For all three genes selected for ChIP-qPCR analysis, the degree of enrichment of their promoters by ChIP with an anti-H4ac was significantly reduced in the pkr mutant compared to PH-1. These results indicate that a subset of genes regulated by Pkr are controlled via FgSnt1 and H4 acetylation in F. graminearum.

Overall, results from this study suggest that PKA regulates Hdf1 or all Set3 HDAC activities via phosphorylating FgSnt1. In the wild type, unphosphorylated FgSnt1 has no direct interaction with Hdf1. Phosphorylation of S443 by PKA results in the interaction of FgSnt1 between its N-terminal and C-terminal regions and possibly conformation all changes, which leads to the interaction of FgSnt1-S443^P with Hdf1 (similar to FgSnt1^ΔCT98). The interaction of FgSnt1 with Hdf1 may reduce HDAC activities of Hdf1 or FgSet3 complex. Proper regulation of the H4 acetylation level is important for hyphal growth and other developmental or infection processes. In the pkr mutant, the absence of Pkr regulatory subunits causes the degradation of Cpk1 by 26S proteasome. Therefore, the pkr and cpk1 mutants had similar defects. When Hdf1 or FgSet3 HDAC activities are not properly regulated by PKA via FgSnt1 in these mutants, a significant reduction in H4 acetylation and reduced expression of genes regulated by H4 acetylation result in growth defects in F. graminearum. Based on this hypothesis, mutations in other FgSet3 complex components may also be suppressive to the pkr mutant. Therefore, it will be important to sequence other suppressor strains for mutations in other key components of the FgSet3 complex, including FgSET3, FTL1 (SIF2 ortholog), and HDF1.

Materials and methods

Strains and culture conditions

The wild-type strain PH-1 and all the mutants of F. graminearum generated in this study were routinely cultured on potato dextrose agar (PDA) at 25°C and assayed for growth rate as described [57]. Conidiation and conidium morphology were assayed with conidia harvested from 5-day-old liquid carboxymethyl cellulose (CMC) cultures [58]. To induce self-fertilization, aerial hyphae of 7-d-old carrot agar (CA) cultures were pressed down with sterile 0.1% Tween 20 and incubated at 25°C. Perithecium formation were examined at seven days after self-fertilization as described [59]. For transformation of F. graminearum, protoplast preparation and PEG-mediated transformation were performed as described [60]. For transformant selection, hygromycin B and geneticin (Coolaber, Beijing, China) were added to the final concentration of 300 and 200 μg ml⁻¹ to both top and bottom agar for selection.

Identification of suppressor mutations

To identify mutations in pkr suppressor strains, genomic DNA was isolated from hyphae harvested from 24 h YEPD cultures and sequenced by Illumina HiSeq 2500 with the Illumina platform at Novogene (Beijing, China) to 50×coverage with pair-end libraries. The resulting genomic sequence data were deposited in the NCBI SRA repository (bioproject accession No. PRJNA855364). The sequence reads were mapped onto the reference genome of strain PH-1 [31]. The variants in all the genome were firstly found by the DNAseq pipeline in hits package (https://github.com/xulab-nwafu/hits), which integrated bowtie2, Picard, SAMtools, and BCFtools [61,62]. The specific mutations in the suppressor genomes were then identified by vcfcmp (https://github.com/xulab-nwafu/filter_vcf), which wrapped the intersection function of BCFtools and variant annotation function of SnpEff [63]. To identify mutations in FgSNT1 in the other suppressors of pkr, its coding region was amplified and sequenced.

Generation of the Fgsnt1 and FgSNT1^ΔCT98 deletion mutants

To generate Fgsnt1 deletion mutant with the split marker approach, the flanking sequences of FgSNT1 were amplified from PH-1 and connected to fragments of the neomycin resistance marker amplified from pFL2 [64] by overlapping PCR. The resulting FgSNT1 gene replacement construct was transformed into protoplasts of PH-1. Geneticin-resistant transformants were screened by PCR for the deletion of FgSNT1. The same approach was used to generate the FgSNT1^ΔCT98 mutant in which only the C-terminal 98 aa residues of FgSNT1 (instead of the entire gene) were deleted.

To generate the Fgsnt1 pkr and FgSNT1^ΔCT98 pkr double mutants, the flanking sequences of PKR were amplified and connected to the hygromycin phosphotransferase (hph) cassette [22] and transformed into the Fgsnt1 and FgSNT1^ΔCT98 mutants. Transformants resistant to both hygromycin and geneticin were screened for the deletion of PKR and verified for the deletion of FgSNT1 or its CT98 region. All the primers used to generate and identify these mutants were listed in S2 Table.

Generation the FgSNT1^S443D allele and transformants

The S443D mutation in FgSNT1 was introduced by overlapping PCR as described [26] with primers list in the S2 Table. The resulting PCR products were cloned into vector pKNTG carrying the geneticin-resistance marker [65]. The FgSNT1^S443D construct was then transformed into protoplasts of Fgsnt1 mutant that was generated by replacing the ORF of FgSNT1 with the hygromycin phosphotransferase (hph) and thymidine kinase cassettes (Human Simplex Virus, HSV-tk). Transformants resistant to geneticin and floxuridine (5-Fluorouracil 2’-deoxyriboside) were screened by PCR to identify FgSNT1^S443D transformants and verified by sequencing analysis for the S443D mutation. The PKR gene replacement construct carrying hph was transformed into protoplasts of the FgSNT1^S443D mutant to generate the FgSNT1^S443D pkr mutants.

Plant infection assays

Flowering wheat heads of 6-week-old wheat cultivar XiaoYan 22 were inoculated with 10 μl of conidia suspensions (2×10⁴ conidia ml⁻¹) at the fifth spikelet from the base as described [57]. Spikelets with typical wheat scab symptoms were examined at 14 dpi to estimate the disease index [24]. For assaying infectious growth, infected rachis tissues were embedded in Spurr resin after fixation and dehydration as described previously [23]. Thick sections were then prepared and stained with 0.5% (wt/vol) toluidine blue as described [66]. For assaying infection cushion formation, infected lemmas were fixed with 4% (vol/vol) glutaraldehyde and dehydrated in a series of acetone. The samples were coated with gold–palladium and examined with a JEOL 6360 scanning electron microscope (Jeol Ltd., Tokyo, Japan) as described [66].

Detection of histone acetylation levels

Total proteins were isolated from vegetative hyphae harvested from 12 h YEPD cultures as described [24]. Acetylation of histone H3 and H4 was detected with anti-H3ac (ab47915), anti-H4ac (ab177790), and anti-H4K16ac (ab194352) antibodies from Abcam (Cambridge, UK). Detection with the anti-H3 (ab209023, Abcam), and anti-H4 (ab10158, Abcam) antibodies were used as the loading controls. The band intensity was quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Relative intensity of the histone acetylation was normalized to that of non-acetylated histone, and compared with that of the control group [67]. Each experiment was performed for three independent biological replicates, and error bars represent standard deviation estimated with data from three independent replicates.

Assays for PKA activity and Cpk1 expression

Vegetative hyphae were harvested from 24 h YEPD (Yeast extract peptone dextrose) cultures by filtration through two layers of Miracloth (Sigma, USA) and washed with sterile water [25]. PKA activities were assayed with the PKA kinase assay kits (Type I) (IMMUNECHEM, China) [68]. The total protein isolated from F. graminearum and ATP solution were added to the 96 wells plate, which was pre-immobilized with 50 μL of PKA-substrate (KRREILSRRPSYR). After the kinase reaction, the PKA-substrate in wells were mixed with anti-pSubstrate antibodies and then incubated with anti-rabbit IgG HRP solution with TMB (3,3’-5,5’-Tetramethylbenzidine). The PKA phosphorylation activity is proportional to the color intensity. The primary antibody to F. graminearum Cpk1 was generated in rabbits using the synthetic oligopeptide VKAGAGDASQFDRYPE (ABclonal, Wuhan, China). The specificity of the resulting anti-Cpk1 antibody was verified by western blot analysis with total proteins isolated from the cpk1 mutant [23] and PH-1 as described [36,69]. Detection with an anti-Tub2 β-tubulin antibody [70] was used as the loading control.

Yeast two-hybrid assays

The Matchmaker yeast two-hybrid system (Clontech, Mountain View, CA, USA) was used to assay protein-protein interactions. The FgSNT1 and FgSNT1^ΔCT98 ORFs were amplified from the 1^st-strand cDNA synthesized with the HiScript II Q RT SuperMix (Vazyme Biotech, Nanjing, China) as described [71] and cloned into pGBK7 as the bait vector. The prey construct of HDF1 and FgSET3 were generated with pGADT7 (Clontech). The resulting bait and prey vectors were co-transformed in pairs into yeast strain AH109 (Clontech). The Leu⁺ Trp⁺ transformants were isolated and assayed for growth on SD-Trp-Leu-His medium and galactosidase activities with filter lift assays [72]. The positive and negative controls were provided in the Matchmaker library construction kit (Clontech).

GST pull-down assays

To generate the FgSNT1-GST and FgSNT1^ΔCT98-GST fusion constructs, their ORFs were amplified from 1^st strand cDNA synthesized with RNA isolated from PH-1 and FgSNT1^ΔCT98 mutant as described [71] and cloned into pGEX4T1 [73]. To generate HDF1-HIS constructs, its ORFs were amplified from PH-1 by RT-PCR and cloned into pCOLDI [74]. Recombinant Hdf1-HIS, FgSnt1-GST, and FgSnt1^ΔCT98-GST fusion proteins were isolated from Escherichia coli cells as described with anti-GST beads [75]. Equal amounts of Hdf1-HIS and FgSnt1-GST or FgSnt1^ΔCT98-GST fusion proteins were mixed and incubated at 4°C for 4 h before adding glutathione resins (Smart Lifesciences, Changzhou, China). After incubation for 12 h and washing for three times, proteins bound to glutathione resins were eluted as described [76,77]. Western blots of total proteins and proteins eluted from glutathione resin were detected with the anti-GST (#CW0084) and anti-HIS (#CW0286) antibodies from CWBIO (Beijing, China). The FgSNT1^1-774-GST and Cpk1-HIS fusion constructs were generated with the same approach. Equal amounts of Cpk1-HIS and FgSNT1^1-774-GST fusion proteins were mixed and incubated with glutathione resins (Smart Lifesciences, Changzhou, China) [76,77]. Western blots of total proteins and proteins eluted from glutathione resin were detected with the anti-GST (#CW0084) and anti-HIS (#CW0286) antibodies from CWBIO (Beijing, China).

Phos-tag gel assays

The FgSNT1^1-774-GST, and Cpk1-HIS fusion constructs were generated from the cDNA of PH-1 as described [71] and cloned into pGEX4T1 and pCOLDI [73,74]. Recombinant FgSNT1^1-774-GST and Cpk1-HIS fusion proteins were isolated and incubated with 75 μM ATP solution (TakaRa) for 2 h as described [38]. Samples were boiled for 5 min prior to loading onto a 10% SDS-PAGE gel and Phos-tag SDS-PAGE, respectively. SDS-PAGE was carried out with polyacrylamide gels. Phos-tag SDS-PAGE was performed with 7.5% polyacrylamide gels containing 20 μM Phos-tag acrylamide and 40 μM MnCl₂ [39]. After electrophoresis, proteins were transferred to PVDF membranes (Sigma) and probed with an anti-GST antibody. The phosphorylation status of FgSNT1^1-774-GST was analyzed based on the changes in the mobility shifts of phosphorylated proteins on Phos-tag SDS-PAGE. To determine the critical phosphorylated sites, FgSnt1^S18A-, FgSnt1^S38A-, FgSnt1^S322A-, FgSnt1^S443A-, and FgSnt1^S511A-GST were constructed by overlapping PCR with primers list in the S2 Table. Similar approaches were used to determine the phosphorylation of these potential phosphorylated residues.

Quantitative interaction studies by isothermal titration calorimetry (ITC)

Protein-protein interactions were analyzed according to the enthalpy changes produced in titration of Cpk1-HIS binding with FgSnt1^1-744-GST with the isothermal titration calorimetry instrument (NanoITC, New Castle, USA). Specifically, 50 μL of 0.1 mM Cpk1 was loaded into an ITC syringe and titrated into an ITC sample cell containing 300 μL of 0.01 mM FgSnt1^1-744-GST dissolved solution at a sequence of 25 injections. The control experiments including the titration of Cpk1-HIS to PBS, titration of PBS to PBS and titration of PBS to FgSnt1^1-744-GST were performed, in which the enthalpy change produced in the titration of Cpk1 to PBS was subtracted from that produced in the titration of Cpk1-HIS to the FgSnt1^1-744-GST (S7 Fig) [78].

RNA-seq analysis

Vegetative hyphae of PH-1 and the pkr mutant, FgSNT1^ΔCT98 pkr, and FgSnt1^S443D pkr mutants were harvested from 12 h liquid YEPD cultures. Total RNA was extracted with the TRIzol Reagent (Life technologies, US) following the manufacturer’s instructions. Library construction and sequencing with an Illumina Nova-PE150 sequencer were performed at Novogene (Beijing, China). Over 25 Mb high-quality RNA-seq reads (deposited at the NCBI Sequence Read Archive database under the bioproject accession No. PRJNA858617) were obtained for each sample and mapped onto the reference genome of F. graminearum [31] with HISAT2[79]. The number of reads mapped to each gene was analyzed with the featureCounts software [80]. DEGs were identified using the edgeRun package [81] with the log2 FC (log₂ fold change) greater than 1 and false discovery rate (FDR) less than 0.05 as the cut off values and analyzed with Blast2GO for GO enrichment [82] with the p-values calculated by the Benjamini-Hochberg procedure [83]. All the Perl, R, and Shell scripts used in this study for sequencing and other analysis were available at GitHub as described [7].

ChIP-qPCR assays

Vegetative hyphae of PH-1 and the pkr mutant were harvested from 12 h liquid YEPD cultures. Chromatins were cross-linked with 0.75% formaldehyde for 15 min at room temperature [84]. Samples were ground with liquid nitrogen and re-suspended in lysis buffer (50 mM, HEPES-KOH pH 7.5, 140 mM NaCl, 1 mM EDTA PH 8.0, 1% Triton, 0.1% Sodium Deoxycholate, 1% SDS) with protease inhibitors (Sigma, USA). The cross-linked chromatins were sonicated by Vibra-Cell (Sonics, USA) to an average DNA fragment size of 100–500 bp. An aliquot of the sheared lysate was saved as the input control for each sample. The rest was incubated with 10 μg anti H4ac antibody (ab177790, Abcam) and protein A agarose beads (Cell Signaling Technology, USA) as described [44]. Genomic DNA in the immunocomplexes formed on agarose beads was recovered with the HiPure Gel Pure DNA Mini Kit (Megan, China). The presence of the promoter sequences of targeted genes in the input and recovered ChIP samples was assayed by quantitative PCR with the ChamQ SYBR qPCR Master Mix (Vazyme, China) with primers listed in S2 Table. The resulting ChIP data were analyzed with the Percent Input Method [85].

Supporting information

S1 Table. Mutations identified by sequencing PCR products in the pkr suppressor strains.

(DOC)

S2 Table. Primers used in this study.

(DOCX)

S3 Table. Putative PKA Phosphorylation sites in FgSnt1.

(DOCX)

S4 Table. RNA-seq data analysis of of wild type PH-1, pkr mutant, FgSNT1^S443D pkr, and FgSNT1^ΔCT98 pkr mutants.

(XLSX)

S1 Fig. Assay of growth rate of all spontaneous pkr suppressors.

60 subcultures of spontaneous sectors were collected and categorized into three types based on their growth rate. 14 type I suppressor strains grew more than two-fold faster than pkr mutant. 34 type II suppressors grew 1.5-fold faster than pkr mutant. 12 type III suppressors grew less than 1.5-fold faster than pkr mutant.

(TIF)

S2 Fig. Phenotypes of the spontaneous suppressor strains selected for whole genome sequencing.

(A). Three-day-old PDA cultures of the wild-type strain PH-1, pkr mutant and pkr suppressors (B). Perithecia from mating cultures of the marked strains were examined at 12 dpf. (C). Flowering wheat heads were drop-inoculated with conidia of the marked strains and photographed 14 days post-inoculation (dpi). Black dots mark the inoculated spikelets.

(TIF)

S3 Fig. Schematic diagram of the primers used to generate mutants in Fusarium graminearum.

(A). Schematic diagram of the primers used to generate Fgsnt1 mutants (left panel), and screening of Fgsnt1 transformants by PCR amplification (right panel). (B). Schematic diagram of the primers used to generate FgSNT1^ΔCT98 transformants (left panel), and screening of FgSNT1^ΔCT98 transformants by PCR amplification (right panel). (C). Schematic diagram of the primers used to generate pkr mutant in Fgsnt1 and FgSNT1^ΔCT98 mutants (left panel), and screening of pkr transformants by PCR amplification (right panel). (D). Schematic diagram of the primers used to introduce S443D mutation. (E). Schematic diagram of the primers used to generate FgSNT1^S443D gene replacement constructs.

(JPG)

S4 Fig. Deletion of the of FgSNT1 have no H4 acetylation affection in the pkr mutant.

(A). Western blots of total proteins isolated from PH-1 and 5 independent pkr mutants were detected with antibodies specific for H4ac. Detection with anti-H4 antibodies was used as a loading control. (B). Western blots of total proteins isolated from PH-1, FgSNT1^ΔCT98 pkr and pkr mutants were detected with antibodies specific for H4ac. Detection with anti-H4 antibodies was used as a loading control. (C). Western blots of total proteins isolated from PH-1 and the pkr, Fgsnt1 pkr, and Fgsnt1mutants were detected with antibodies specific for H4ac. Detection with anti-H4 antibodies was used as a loading control.

(TIF)

S5 Fig. Assays for PKA activities and Cpk1 expression in suppressor H10.

(A). Western blots of total proteins isolated from hyphae of PH-1, pkr, cpk1, and suppressor strain H10 were detected with an anti-Cpk1 antibody. (B). PKA activities were assayed with proteins isolated from the marked strains.

(TIF)

S6 Fig. Yeast two-hybrid assays for the interaction of FgSnt1 with Pkr and Cpk1.

The negative controls with empty prey or bait constructs were presented to support the results of Yeast two-hybrid assays in Fig 7.

(TIF)

S7 Fig. The ITC thermogram of binding interactions between Cpk1 and FgSnt1^1-744.

The control experiments were performed including the titration of Cpk1-HIS to PBS (A), titration of PBS to PBS (B), and titration of PBS to FgSnt1^1-744-GST (C). The enthalpy changes produced in titration of Cpk1-HIS binding with FgSnt1^1-744-GST were analyzed accordingly (D).

(TIF)

S8 Fig. Predicted phosphorylation sites in the FgSnt1.

Schematic drawing of the FgSnt1 protein and alignment of its predicted phosphorylation sites with orthologs from F. graminearum (Fg), F. oxysporum (Fo), M. oryzae (Mo), and N. crassa (Nc). The predicted phosphorylation sites were labeled with stars.

(TIF)

S9 Fig. In vitro kinase assays for the phosphorylation of FgSnt1^1-744 by Cpk1.

In vitro kinase assays with Cpk1-HIS and FgSnt1^S18A-GST, FgSnt1^S38A-GST, FgSnt1^S322A-GST, or FgSnt1^S511A-GST fusion proteins. The S18, S38, S322, or S511 to A mutation in FgSnt1^1-744-GST did not change its phosphorylation by Cpk1.

(TIF)

S10 Fig. Yeast two-hybrid Assays for intramolecular interactions in FgSnt1.

Yeast two-hybrid assays with cells expressing the empty prey or bait constructs as the negative controls for Fig 11.

(TIF)

Data Availability

DNA-seq and RNA-seq data generated in this study were deposited in the NCBI Sequence Read Archive database under bioproject accession No. PRJNA855364 and PRJNA858617 respectively. All other relevant data are included in the main text paper and its Supporting Information files.

Funding Statement

This work was supported by grants to XZ from China Postdoctoral Science Foundation (No. 2020M673500) and National Natural Science Foundation of China (No. 3210170916), and grants to JRX from USWBSI and NSF. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.