Histidine Kinase Sln1 and cAMP/PKA Signaling Pathways Antagonistically Regulate Sporisorium scitamineum Mating and Virulence via Transcription Factor Prf1

Many prokaryotes and eukaryotes utilize two-component signaling pathways to counter environmental stress and regulate virulence genes associated with infection. In this study, we identified and characterized a conserved histidine kinase (SsSln1), which is the sensor of the two-component system of Sln1–Ypd1–Ssk1 in Sporisorium scitamineum. SsSln1 null mutant exhibited enhanced mating and virulence capabilities in S. scitamineum, which is opposite to what has been reported in Candida albicans. Further investigations revealed that the deletion of SsSLN1 enhanced SsHog1 phosphorylation and nuclear localization and thus promoted S. scitamineum mating. Interestingly, SsSln1 and cAMP/PKA signaling pathways antagonistically regulated the transcription of pheromone-responsive transcription factor SsPrf1, for regulating S. scitamineum mating and virulence. In short, the study depicts a novel mechanism in which the cross-talk between SsSln1 and cAMP/PKA pathways antagonistically regulates mating and virulence by balancing the transcription of the SsPRF1 gene in S. scitamineum.


Introduction
The basidiomycetous fungus Sporisorium scitamineum is a global pathogen of sugarcane smut disease that causes substantial losses in cane yield and the sugar industry. S. scitamineum, as Ustilago maydis, is bipolar and undergoes three life stages, of haploid sporidium, yeast-like, non-pathogenic, dikaryotic hypha and diploid teliospore [1]. Haploid sporidia of two opposite mating types, MAT-1 and MAT-2, can recognize each other to undergo fusion, a process known as sexual mating [2]. After sexual mating, the b locus encodes an active heterodimeric transcription factor complex composed of bE and bW proteins derived from different alleles to control filamentation in S. scitamineum [2]. The fusion of two haploid cells of opposite mating types is necessary to form invasively dikaryotic hyphae in S. scitamineum to infect the host [1]. Thus, mating/filamentation plays a key role in S. scitamineum pathogenicity.
The two-component phosphorelay system is widely found, which regulates a variety of cellular processes such as response to environmental stimuli, cell differentiation, secondary metabolite production, antibiotic resistance and virulence, in plant and animal pathogens [3,4]. The two-component phosphorelay system in most eukaryotes mainly

Construction of Strains
The deletion, reintegration and overexpression of genes were performed by polyethylene glycol (PEG) mediate protoplast transformation using a split marker approach as described previously [38,41,42]. The Hygromycin resistance (HYG R ) cassette was used as a resistance screening gene. The deletion mutants were generated by individually disrupting SsSLN1 and SsATF1 genes in the MAT-1 and MAT-2 strains. The left and right borders of SsSLN1 and SsATF1 genes were PCR amplified from S. scitamineum wild-type genomic DNA with the primers SsSLN1-LB-F/SsSLN1-LB-R, SsSLN1-RB-F/SsSLN1-RB-R, SsATF1-LB-F/SsATF1-LB-R and SsATF1-RB-F/SsATF1-RB-R, respectively. Two truncated and partially overlapped fragments of HYG R gene were separately PCR amplified from pDAN plasmid with the primers pDAN-F/LB-226-R and pDAN-R/RB-225-F. These PCR products served as templates in fusion PCR to generate two PCR fragments with the primers SsSLN1-LB-F/LB-226-R, RB-225-F/SsSLN1-RB-R, SsATF1-LB-F/LB-226-R and RB-225-F/SsATF1-RB-R, individually. The mixture of two fusion homologous fragments for each targeted gene was transformed into MAT-1 and MAT-2 protoplasts via PEG-mediated protoplast transformation. The transformants were recovered in a regeneration medium impregnated with 200 mg/mL hygromycin B (Merck, Saint Louis, MO, USA). Putative deletion mutants were screened and identified by PCR amplification using the following primers: SsSLN1-inside-F/SsSLN1-inside-R, SsSLN1-outside-F/SsSLN1-outside-R, SsATF1-inside-F/SsATF1-inside-R and SsATF1-outside-F/SsATF1-outside-R. This operation was used to generate the double-deletion mutants ss1sln1∆gpa3∆, ss1sln1∆uac1∆ and ss1sln1∆adr1∆, with the Zeocin resistance (ZEO R ) cassette as selection marker in the ss1sln1∆ background.
SsSLN1 and SsATF1 reintegration was carried out, as previously described [38]. The HYG R gene in the mutant was replaced with the reintegrated gene, together with the ZEO R gene as a selection marker, by a split marker approach. A fragment containing the native promoter and SsSLN1 or SsATF1 gene was PCR amplified with wild-type genomic DNA as a template, using the primers SsSLN1-COM-F/SsSLN1-COM-R and SsATF1-COM-F/SsATF1-COM-R, respectively. These PCR products were ligated with vector pEASY-COM containing HYG R -LB (HYG R -left homologous arm), ZEO R gene and HYG R -RB (HYG R -right homologous arm). Two reintegrated homologous fragments, one fragment including HYG R -LB, a complete target gene and partially overlapped fragments of the ZEO R gene, and the other containing partially overlapped fragments of the ZEO R gene and HYG R -RB, were PCR amplified with pEASY-SsSLN1-COM or pEASY-SsATF1-COM as a template using the primers COM-LB-F/COM-LB-R and COM-RB-F/COM-RB-R, respectively. Two reintegrated homologous fragments were individually transformed into ss1sln1∆, ss2sln1∆, ss1atf1∆ and ss2atf1∆ protoplasts through PEG-mediated protoplast transformation. Putative complementation transformants were selected with 100 mg/mL zeocin and identified with the primers SsSLN1-inside-F/SsSLN1-inside-R and SsATF1inside-F/SsATF1-inside-R.
The IP locus, a DNA sequence without a function in S. scitamineum genomic DNA, was used as the target sequence for the overexpression of SsPTP1, SsPTP2 and SsPRF1.
It was replaced with the overexpressed gene together with the ZEO R gene as a selection marker by a split marker approach. The CD fragments of SsPTP1, SsPTP2 and SsPRF1 genes were PCR amplified with S. scitamineum cDNA as a template, using the primers SsPTP1-OE-F/SsPTP1-OE-R, SsPTP2-OE-F/SsPTP2-OE-R and SsPRF1-OE-F/SsPRF1-OE-R, individually. These PCR products were individually ligated with vector pEASY-OE containing IP-LB (IP-left homologous arm), a constitutive GPA promoter, ZEO R gene and IP-RB (IP-right homologous arm). Two overexpressed homologous fragments, one fragment including IP-LB, the GPA promoter fused with the target gene and partially overlapped fragments of the ZEO R gene, and the other containing partially overlapped fragments of the ZEO R gene and IP-RB, were PCR amplified with templates pEASY-SsPTP1-OE, pEASY-SsPTP2-OE or pEASY-SsPRF1-OE using the primers OE-LB-F/OE-LB-R and OE-RB-F/OE-RB-R, respectively. Two overexpressed homologous fragments were mixed and transformed into MAT-1, MAT-2 or ss1hog1∆ protoplasts by PEG-mediated protoplast transformation, respectively. Putative overexpression transformants were selected with 100 mg/mL zeocin and identified with RT-qPCR using the primers qRT-SsPTP1-F/qRT-SsPTP1-R, qRT-SsPTP2-F/qRT-SsPTP2-R and qRT-SsPRF1-F/qRT-SsPRF1-R.
The mixture of two fusion homologous fragments was transformed into MAT-1 or ss1sln1∆ protoplasts via PEG-mediated protoplast transformation for each targeted gene. The transformants were recovered in a regeneration medium impregnated with 200 mg/mL hygromycin B or 100 mg/mL zeocin. Putative fluorescent strains were screened and identified by fluorescence microscope and PCR amplification, respectively. The primers and sequences used in this study are listed in Table S1. Details of the strains generated and used in this study are listed in Table 1.

Nucleic Acid Related Manipulation
The S. scitamineum strains were grown on YePSA medium for 1-2 days at 28 • C. Then, the strains were collected and rapidly ground in liquid nitrogen to extract the genomic DNA of S. scitamineum using a modified SDS-based method [2]. The fresh haploid sporidia grown on YePSA medium at 28 • C for 30 h were used for the total RNA extraction of S. scitamineum with TRIzol reagent (ThermoFisher Scientific, Carlsbad, CA, USA) following established protocol [2]. HiScript ® II 1st Strand cDNA Synthesis Kit (Vazyme, Nanjing, China) was used to synthesize the cDNA. Real-Time Quantitative PCR was performed using Fast SYBR™ Green Master Mix (ThermoFisher Scientific, Carlsbad, CA, USA) on QuantStudio 6 Flex (ThermoFisher Scientific, Carlsbad, CA, USA). The relative gene expression level was calculated by adopting the −∆∆Ct method [43] and the cytoskeletal protein gene ACTIN was used as an internal control [38]. The experiment was carried out in triplicate for three independent biological replicates. To carry out Southern blot analysis, genomic DNA of negative control MAT-1 and MAT-2, positive control pDAN plasmid and genomic DNA of deletion mutant were digested with the restriction enzyme Hind III. The HPT sequence was amplified as the probe with PCR DIG Labeling Mix (Roche, Mannheim, BW, Germany). Probe hybridization was performed with a DIG Easy Hyb (Roche, Mannheim, BW, Germany) and detected by CSPD (Roche, Mannheim, BW, Germany). Probed bands of >3.0 kb size in deletion mutants confirmed the correct gene replacement.

Sporidia Staining and Epifluorescence Microscopy
The haploid sporidia of Ss1Hog1:RFP, Ss1Hog1:eGFP(sln1∆), Ss1Atf1:eGFP and Ss1Atf1: eGFP(hog1∆) strains were cultured overnight in YePS liquid medium at 28 • C, diluted to O.D.600 = 0.2 with the fresh YePS liquid medium and grown up to O.D.600 = 1.0. The fresh haploid sporidia were collected for nucleic acid staining by centrifugation and washed twice with 1 × PBS. Then, the cells were re-suspended in 20 µL of the Antifade Mounting Medium with DAPI (Beyotime, Shanghai, China). Finally, the samples (10 µL) were mounted on the slide and photographed under a Leica DMI8 Inverted Fluorescence Microscope using DAPI, GFP and RFP filters. The pictures were taken through Leica Application Suite (LAS) v.X software. Scale bar = 10 µm.

SsHog1 Phosphorylation Assays
The total protein was extracted from the fresh haploid sporidia grown on YePSA medium at 28 • C for 30 h [38]. Protein samples were separated by 10% SDS-PAGE. Phosphorylation of SsHog1 was determined by Western blot analysis with an antibody Phospho-p38 MAPK (Thr180/Tyr182) (D3F9) XP ® Rabbit (Cell Signaling Technology, Boston, MA, USA). Total levels of SsHog1 were detected by probing with an anti-Hog1 antibody (Genecreate Biological Engineering Company, Wuhan, China). Blot signals were displayed using enhanced chemiluminescence (BIO-RAD, 170-5061) after the binding of an Anti-Rabbit IgG-Peroxidase secondary antibody (Sigma, Louis, MO, USA), as described previously [38].

Assessment of Pathogenicity and Relative Fungal Biomass
A highly susceptible sugarcane cultivar ROC22 was used for the pathogenicity assay as described [2,39]. The haploid sporidia of S. scitamineum strains were grown on YePS medium in a shaking incubator at 28 • C for 1-2 days. The fresh haploid sporidia were collected, re-suspended in the sterilized ddH 2 O, adjusted to 1 × 10 6 cells/mL and mixed with haploid sporidia of opposite mating types in equal volume. Sugarcane seedlings of ROC22 grown to 5-6 leaf stage were inoculated by injection with approximately 0.2 mL of the mixture per plant. MAT-1 and MAT-2 mixture served as a positive control. Inoculated plants were kept in the greenhouse and a natural cycle of day and night was followed for 3-6 months. Three biological repeats were performed in the inoculation and each replicate involved the infection of at least 15 plants. The symptoms of 'black whip' were documented and photographed at about six months post inoculation. Percentage (%) of 'black whip'/total seedlings was estimated.
Fungal biomass assay of the inoculated sugarcane seedlings was carried out according to Sun [44]. The same amount of S. scitamineum sporidia (mixture of compatible mating types) was injected into sugarcane seedlings at 3 days post-inoculation (dpi), and total DNA of the inoculated sugarcane tissue was extracted. The relative fungal biomass was measured using the fungal ACTIN gene as a reference, whereas the sugarcane glyceraldehyde dehydrogenase (GAPDH) gene served as an internal control.

Statistical Analysis
Data were expressed as mean ± standard error (SE). Differences among different treatments were analyzed using GraphPad Prism v.5 software.

Identification of SsSln1 Protein in S. scitamineum
A BLASTp search of Sporisorium reilianum Sln1 (SJX65361.1) protein sequence revealed that the S. scitamineum proteome harbors a putative histidine kinase osmosensor protein (CDU22142.1), which we named SsSln1. SsSln1 was predicted as a peptide of 1302 amino acids. Putative domains of SsSln1 protein were further predicted through the SmartBLAST tool (https://blast.ncbi.nlm.nih.gov/smartblast/ (accessed on 8 June 2021)), which revealed that SsSln1 had a conserved histidine kinases (HisKA) domain, a histidine kinase-like ATPases (HATPase_c) domain and a cheY-homologous receiver (REC) domain ( Figure 1A). Moreover, the SsSln1 protein and its orthologs from other fungal species, including basidiomycetous and ascomycetous, were selected for phylogenetic analysis. The results indicate that SsSln1 was closely related to its orthologs in S. reilianum and Ustilago trichophora, the smut fungi phylogenetic clade, whereas it was distant to Saccharomyces cerevisiae (TPN14626.1) and C. albicans (KHC30928.1), which were present in another clade of the phylogenetic tree ( Figure 1B).
Overall, we identified the putative histidine kinase SsSln1 in S. scitamineum, and found that SsSln1 was highly conserved with its orthologs in basidiomycetous fungi.
Overall, we identified the putative histidine kinase SsSln1 in S. scitamineum, and found that SsSln1 was highly conserved with its orthologs in basidiomycetous fungi.

Deletion of SsSLN1 Gene Promotes Mating and Pathogenicity of S. scitamineum
To characterize the role of SsSln1 in S. scitamineum, the deletion mutants of ss1sln1Δ in MAT-1 and ss2sln1Δ in MAT-2 background were separately generated, whereas reintegrated mutants of Ss1SLN1-COM in ss1sln1Δ and Ss2SLN1-COM in ss2sln1Δ background were generated by following the homologous recombination approach, as described before [38,41]. Mutants were confirmed by PCR amplification (Figure S1A,B) and Southern blotting ( Figure S2A). Real-Time Quantitative PCR (RT-qPCR) analysis confirmed the complete deletion of SsSLN1 gene in ss1sln1Δ and ss2sln1Δ, and verified that SsSLN1 gene fully recovered in Ss1SLN1-COM and Ss2SLN1-COM strains, respectively ( Figure S2B). S. scitamineum wild-type (WT) strains and generated mutants generated and used in this study are listed in Table 1.

Deletion of SsSLN1 Gene Promotes Mating and Pathogenicity of S. scitamineum
To characterize the role of SsSln1 in S. scitamineum, the deletion mutants of ss1sln1∆ in MAT-1 and ss2sln1∆ in MAT-2 background were separately generated, whereas reintegrated mutants of Ss1SLN1-COM in ss1sln1∆ and Ss2SLN1-COM in ss2sln1∆ background were generated by following the homologous recombination approach, as described before [38,41]. Mutants were confirmed by PCR amplification ( Figure S1A,B) and Southern blotting ( Figure S2A). Real-Time Quantitative PCR (RT-qPCR) analysis confirmed the complete deletion of SsSLN1 gene in ss1sln1∆ and ss2sln1∆, and verified that SsSLN1 gene fully recovered in Ss1SLN1-COM and Ss2SLN1-COM strains, respectively ( Figure S2B). S. scitamineum wild-type (WT) strains and generated mutants generated and used in this study are listed in Table 1.
To evaluate the impact of SsSln1 on mating, the haploid cells of MAT-1, MAT-2, ss1sln1∆, ss2sln1∆, Ss1SLN1-COM and Ss2SLN1-COM were mixed with the sporidia of opposite mating type and inoculated on YePSA medium. Successful formation of dikaryotic hyphae was observed as the appearance of white, fuzzy colonies at 30 h post inoculation. However, the ss1sln1∆ × ss2sln1∆ combination exhibited a stronger white and fuzzy colony as compared to MAT-1 × MAT-2 ( Figure 2A). Meanwhile, the reintegrated strain of the Ss1SLN1-COM × Ss2SLN1-COM combination displayed a semblable mating of MAT-1 × MAT-2 on the same plate ( Figure 2A). To examine the role of SsSln1 in virulence, the sporidial suspensions of MAT-1 × MAT-2, ss1sln1∆ × ss2sln1∆ and Ss1SLN1-COM × Ss2SLN1-COM combination were inoculated on the susceptible sugarcane cultivar ROC22. The results show that the typical symptoms of 'black whip' disease were observed in all strain-infected seedlings. However, the disease symptoms of ss1sln1∆ × ss2sln1∆ were more conspicuous than the wild-type or reintegrated strains ( Figure 2B). Statistical results reveal that 68.15% of the total seedlings infected with ss1sln1∆ × ss2sln1∆ displayed 'black whip' symptoms, which was significantly higher (p < 0.05) than the incidence in other in treatments ( Figure 2C). ss1sln1Δ, ss2sln1Δ, Ss1SLN1-COM and Ss2SLN1-COM were mixed with the sporidia of opposite mating type and inoculated on YePSA medium. Successful formation of dikaryotic hyphae was observed as the appearance of white, fuzzy colonies at 30 h post inoculation. However, the ss1sln1Δ × ss2sln1Δ combination exhibited a stronger white and fuzzy colony as compared to MAT-1 × MAT-2 ( Figure 2A). Meanwhile, the reintegrated strain of the Ss1SLN1-COM × Ss2SLN1-COM combination displayed a semblable mating of MAT-1 × MAT-2 on the same plate ( Figure 2A). To examine the role of SsSln1 in virulence, the sporidial suspensions of MAT-1 × MAT-2, ss1sln1Δ × ss2sln1Δ and Ss1SLN1-COM × Ss2SLN1-COM combination were inoculated on the susceptible sugarcane cultivar ROC22. The results show that the typical symptoms of 'black whip' disease were observed in all strain-infected seedlings. However, the disease symptoms of ss1sln1Δ × ss2sln1Δ were more conspicuous than the wild-type or reintegrated strains ( Figure 2B). Statistical results reveal that 68.15% of the total seedlings infected with ss1sln1Δ × ss2sln1Δ displayed 'black whip' symptoms, which was significantly higher (p < 0.05) than the incidence in other in treatments ( Figure 2C).
In short, the SsSln1 negatively regulated the mating and virulence of S. scitamineum.  In short, the SsSln1 negatively regulated the mating and virulence of S. scitamineum.

SsSln1 Negatively Regulates Phosphorylation and Nuclear Localization of SsHog1
The Sln1-Ypd1-Ssk1 "two-component" system regulates the Hog1-MAP kinase cascade in the budding yeast [7]. To explore the effects of SsSln1 on SsHog1 activity, we constructed a Ss1Hog1:RFP fusion strain in MAT-1 and Ss1Hog1:eGFP(sln1∆) fusion strain in ss1sln1∆. The localization analysis was carried out by staining the cell of Ss1Hog1:RFP and Ss1Hog1:eGFP(sln1∆) fusion strains with 4 ,6-diamidino-2-phenylindole (DAPI), photographed under a fluorescence microscope. In the absence of osmotic stress conditions, the SsHog1:RFP fusion protein was mainly distributed in the cytoplasm ( Figure 3A), whereas the SsHog1 protein signal accumulated in the nucleus of about 23.3% of cells ( Figure 3B). However, 0.8 M sorbitol treatment enhanced the entry of SsHog1:RFP fusion protein signals, and the number of cells increased up to 48.4% of cells ( Figure 3A,B). Surprisingly, in ss1sln1∆ mutants, a large amount of SsHog1:eGFP fusion protein concentrated in the nucleus ( Figure 3A), and the proportion of SsHog1 protein nuclear localization was noted in about 73.3% of cells ( Figure 3B). These results suggest that SsHog1 was activated in the absence of SsSLN1 even without hyperosmosis treatment. To investigate the effect of SsSln1 on the phosphorylation level of SsHog1 in S. scitamineum, the level of SsHog1 phosphorylation was examined in MAT-1, ss1hog1∆, ss1sln1∆ and Ss1SLN1-COM sporidia by Western blotting. The SsHog1 was found to be highly activated and phosphorylated in ss1sln1∆ mutants as compared to wild-type and Ss1SLN1-COM, and the phosphorylation of SsHog1 was abolished in ss1hog1∆ ( Figure 3C).
Taken together, the SsSln1 negatively regulated the phosphorylation and nuclear localization of SsHog1 in S. scitamineum.

Phosphorylation of SsHog1 Is Necessary for Mating and Virulence of S. scitamineum
The role of protein tyrosine phosphatase Ptps in the dephosphorylation of Hog1 orthologs in C. neoformans was reported [30]. Therefore, we first identified the putative Ptps orthologs SsPtp1 (CDU22358.1) or SsPtp2 (CDU23562.1), and generated the overexpression mutants of SsPTP1-OE or SsPTP2-OE in MAT-1 and MAT-2 background, respectively ( Figures S1E and S2D,E). As expected, the overexpression of Ss1PTP2-OE exhibited a lower level of SsHog1 phosphorylation ( Figure 4A) in comparison to the wild type. To investigate the effects of the SsHog1 phosphorylation level on the mating and virulence of S. scitamineum, mating assays were performed by cospotting the compatible strains onto YePSA medium. The results reveal that during mating, the white hyphae was slightly reduced in the Ss1PTP1-OE × Ss2PTP1-OE combination but significantly reduced in the Ss1PTP2-OE × Ss2PTP2-OE combination, compared to the wild type ( Figure 4B). In addition, the ss1hog1∆ mutant was also noted to be significantly deficient in mating in comparison to wild type and Ss1HOG1-COM ( Figure 4B). Further pathogenicity analysis revealed that the typical symptoms of the disease 'black whip' were significantly (p < 0.05) reduced in ss1hog1∆ × MAT-2 combination, as compared to the wild-type and reintegrated strains ( Figure 4C). Moreover, the percentage (%) of 'black whip'/total seedlings was also markedly reduced in ss1hog1∆ × MAT-2 combination than wild-type or Ss1HOG1-COM × MAT-2 combination ( Figure 4D). The inoculation of Ss1PTP1-OE × Ss2PTP1-OE and Ss1PTP2-OE × Ss2PTP2-OE combination also produced the 'black whip' symptoms but their virulence was significantly more reduced than the wild-type ( Figure 4D).
Overall, the results demonstrate that SsPtp2 was involved in the dephosphorylation of SsHog1 and resulted in defective mating and virulence in S. scitamineum. phorylation was examined in MAT-1, ss1hog1Δ, ss1sln1Δ and Ss1SLN1-COM sporidia Western blotting. The SsHog1 was found to be highly activated and phosphorylated ss1sln1Δ mutants as compared to wild-type and Ss1SLN1-COM, and the phosphorylat of SsHog1 was abolished in ss1hog1Δ ( Figure 3C).
Taken together, the SsSln1 negatively regulated the phosphorylation and nuclear calization of SsHog1 in S. scitamineum. p < 0.01, p < 0.001). Mean ± S.E. were derived from three independent biological repeats with three replications. (C) Phosphorylation of SsHog1 was enhanced in ss1sln1∆ mutant cells. The fresh haploid sporidia were grown in YePSA medium at 28 • C for 30 h. The total protein of sporidia was extracted with lysis buffer. Western blots depicting the levels of SsHog1 phosphorylation in the indicated strains. Above: blots were probed for phosphorylated SsHog1 (Hog1-P). Below: blots were probed for total SsHog1 (Hog1). Coomassie blue staining of total proteins served as loading control.  Percentage (%) of 'black whip'/total seedlings was indicated. Bar chart depicts the statistical difference among the mean values ( p < 0.05, p < 0.01, p < 0.001). Mean ± S.E. were derived from two independent biological repeats with three replications.

Phosphorylation Level of SsHog1 Positively Mediates Mating and Virulence of S. scitamineum
An Atf1 ortholog named SsAtf1 was identified and characterized in S. scitamineum. Deletion mutants of ss1atf1∆ in MAT-1 and ss2atf1∆ in MAT-2, and reintegrated mutants of Ss1ATF1-COM in ss1atf1∆ and Ss2ATF1-COM in ss2atf1∆ background, were individually obtained by homologous recombination approach as described before [38]. Mutants were confirmed by PCR amplification (Figure S1A,B) and Southern blot ( Figure S2A). The results of RT-qPCR analysis revealed the complete deletion of the SsATF1 gene in ss1atf1∆ and ss2atf1∆, and it was fully recovered in the reintegrated strains Ss1ATF1-COM and Ss2ATF11-COM, respectively ( Figure S2C).
To characterize the potential influence of SsAtf1 on the phosphorylation level of SsHog1, the levels of SsHog1 phosphorylation in MAT-1, ss1atf1∆ and Ss1ATF1-COM sporidia were assessed. As expected, the SsHog1 was activated and phosphorylated in ss1atf1∆ mutants as compared to wild-type and Ss1ATF1-COM ( Figure 5A). The phosphorylation of Atf1 by M. oryzae Hog1 ortholog has been reported [29]. Therefore, a Ss1Atf1:eGFP fusion strain was generated in MAT-1 or ss1hog1∆ to test its role in SsAtf1 nuclear localization. To study the localization, the Ss1Atf1:eGFP and Ss1Atf1:eGFP(hog1∆) sporidia were stained with DAPI and observed under a fluorescence microscope. The results indicate that SsAtf1:eGFP fusion protein was mainly accumulated in the nucleus of both MAT-1 and ss1hog1∆ strains ( Figure 5B), suggesting that SsHog1 did not affect the nuclear localization of SsAtf1. Furthermore, the mating with wild-type, deletion of SsATF1 and reintegrated strains was also assessed. The results depict an increase in the formation of white and fuzzy colonies in ss1atf1∆ × ss2atf1∆ combination in comparison to wild-type and reintegrated strains ( Figure 5C). The relative fungal biomass continuously and significantly increased for up to 3 days in seedling stems inoculated with ss1atf1∆ × ss2atf1∆, as compared to that MAT-1 × MAT-2 or Ss1ATF1-COM × Ss2ATF1-COM ( Figure 5D).
These data demonstrate that SsAtf1 contributed to the dephosphorylation of SsHog1, enhanced SsHog1 phosphorylation level and facilitated the mating or virulence of S. scitamineum.
3.6. SsSln1 Negatively Regulates the Transcription of SsPRF1 by SsHog1 Phosphorylation, but Not through the cAMP/PKA Signaling Pathway Previously, our study showed that the pheromone response factor SsPRF1 gene plays a key role in mating and pathogenicity [39], and was significantly down-regulated in cAMP/PKA defective mutants [38]. Therefore, RT-qPCR analysis was performed for assessing the expression of the SsPRF1 gene in MAT-1, ss1hog1∆, Ss1PTP1-OE, Ss1PTP2-OE, ss1sln1∆ and ss1atf1∆ strains grown on YePSA medium. The results show that transcriptional expression of the SsPRF1 gene was significantly (p < 0.05) reduced in ss1hog1∆ and Ss1PTP2-OE, whereas it slightly decreased in Ss1PTP1-OE mutant, compared to wild type ( Figure 6A). However, the transcription level of the SsPRF1 gene was significantly (p < 0.05) up-regulated in ss1sln1∆ mutant, whereas it slightly increased in ss1atf1∆ mutant ( Figure 6A). To evaluate whether the reduced mating in ss1hog1∆ mutants was caused by the down-regulation of the SsPRF1 gene, a Ss1PRF1-OE(hog1∆) mutant was generated, which overexpressed the SsPRF1 gene in ss1hog1∆ mutants. The mating was noted to be fully restored in Ss1PRF1-OE(hog1∆) mutant compared to the ss1hog1∆ mutant ( Figure 6B), and the transcriptional expression of SsPRF1 gene in Ss1PRF1-OE(hog1∆) mutant was close to the level of wild-type MAT-1 ( Figure S2F). Furthermore, the transcriptional profiling showed that SsGPA3, SsUAC1 and SsADR1 genes were not significantly different in wildtype and ss1sln1∆ strains grown in YePSA medium ( Figure 6C). Moreover, SsSln1 did not interact with the regulatory subunit SsUbc1 or the catalytic subunit SsAdr1 of cAMPdependent protein kinase A (PKA) through yeast two-hybrid analysis, indicating that SsSln1 also did not interact with the pheromone response factor SsPrf1 in yeast two-hybrid analysis ( Figure 6D). ss1atf1Δ × ss2atf1Δ, as compared to that MAT-1 × MAT-2 or Ss1ATF1-COM × Ss2ATF1-COM ( Figure 5D).
These data demonstrate that SsAtf1 contributed to the dephosphorylation of SsHog1, enhanced SsHog1 phosphorylation level and facilitated the mating or virulence of S. scitamineum.  with an equal volume of the compatible WT strain or mutants. The MAT-1 × MAT-2, ss1atf1∆ × ss2atf1∆, and Ss1ATF1-COM × Ss2ATF1-COM combination was equally inoculated into the sugarcane seedlings of variety ROC22 to incubate at 28 • C for 3 days. Relative fungal biomass was measured by RT-qPCR with the total DNA isolated from infected sugarcane stems. The fungal ACTIN gene was used for the estimation of relative fungal biomass through the −∆∆Ct method with the plant GADPH gene as an internal control. Bar chart depicts the statistical difference among the mean values ( p < 0.01). Mean ± S.E. were derived from two independent biological repeats with three replications. Mating assay of mutants. Sporidia from MAT-1, ss1hog1Δ and SsPRF1-OE(hog1Δ) strains were separately mixed with an equal volume of MAT-2 sporidia and spotted to YePSA medium to incubate at 28 °C. Images were taken after 40 h of cultivation. (C) Transcriptional profile of SsGPA3, SsUAC1 and SsADR1 gene in the MAT-1 and ss1sln1Δ mutant. The relative gene expression level was calculated according to the -ΔΔCt method with ACTIN as an internal control. The NS represents no significance. Mean ± S.E. were derived from three independent biological repeats with three replications. (D) The yeast two-hybrid assay. Simultaneous co-transformation of positive-control vectors (pGBKT7-p53 and pGADT7-LargeT), negative-control vectors (pGADT7-LargeT and pGBKT7-LaminC), empty vectors (pGADT7 and pGBKT7) and pGBKT7-SsSln1 (bait vectors, BD) with pGADT7-SsAdr1, pGADT7-SsUbc1 and pGADT7-SsPrf1 into the Y2H Gold strain. The transformant was grown on SD-Trp-Leu (lacking tryptophan and leucine) and SD-Trp-Leu-His-Ade (lacking tryptophan, leucine, histidine and adenine) synthetic medium to incubate at 30 °C. Images were taken after 3 days of cultivation. Images are representative of n = 2 biological replications of the experiment.

Cross-Talk between the SsSln1 and cAMP/PKA Pathways Antagonistically Regulates Mating and Virulence by SsPRF1 Transcription
To investigate the potential relationship of SsSln1 with the cAMP/PKA pathway during mating and virulence of S. scitamineum, SsGPA3, SsUAC1 and SsADR1 deletion strains were individually generated in ss1sln1Δ mutants by homologous recombination and named as ss1sln1Δgpa3Δ, ss1sln1Δuac1Δ and ss1sln1Δadr1Δ ( Figures S1C,D and S2G,H). Transcriptional profiling revealed that the expression of pheromone response factor gene SsPRF1 was significantly reduced in the cAMP/PKA defective mutants ( Figure 7A), which is consistent with previous findings [38]. However, SsPRF1 gene expression was not significantly reduced in ss1sln1Δgpa3Δ, ss1sln1Δuac1Δ and ss1sln1Δadr1Δ mutants, as compared to the wild type ( Figure 7A). To test the changes in the mating of double-deletion p < 0.01). Mean ± S.E. were derived from three independent biological repeats with three replications. (B) Mating assay of mutants. Sporidia from MAT-1, ss1hog1∆ and SsPRF1-OE(hog1∆) strains were separately mixed with an equal volume of MAT-2 sporidia and spotted to YePSA medium to incubate at 28 • C. Images were taken after 40 h of cultivation. (C) Transcriptional profile of SsGPA3, SsUAC1 and SsADR1 gene in the MAT-1 and ss1sln1∆ mutant. The relative gene expression level was calculated according to the −∆∆Ct method with ACTIN as an internal control. The NS represents no significance. Mean ± S.E. were derived from three independent biological repeats with three replications. (D) The yeast two-hybrid assay. Simultaneous co-transformation of positive-control vectors (pGBKT7-p53 and pGADT7-LargeT), negative-control vectors (pGADT7-LargeT and pGBKT7-LaminC), empty vectors (pGADT7 and pGBKT7) and pGBKT7-SsSln1 (bait vectors, BD) with pGADT7-SsAdr1, pGADT7-SsUbc1 and pGADT7-SsPrf1 into the Y2H Gold strain. The transformant was grown on SD-Trp-Leu (lacking tryptophan and leucine) and SD-Trp-Leu-His-Ade (lacking tryptophan, leucine, histidine and adenine) synthetic medium to incubate at 30 • C. Images were taken after 3 days of cultivation. Images are representative of n = 2 biological replications of the experiment.
These results collectively suggest that SsSln1 negatively regulates the transcription of SsPRF1 by SsHog1 phosphorylation, but not through the cAMP/PKA signaling pathway.

Cross-Talk between the SsSln1 and cAMP/PKA Pathways Antagonistically Regulates Mating and Virulence by SsPRF1 Transcription
To investigate the potential relationship of SsSln1 with the cAMP/PKA pathway during mating and virulence of S. scitamineum, SsGPA3, SsUAC1 and SsADR1 deletion strains were individually generated in ss1sln1∆ mutants by homologous recombination and named as ss1sln1∆gpa3∆, ss1sln1∆uac1∆ and ss1sln1∆adr1∆ ( Figures S1C,D and S2G,H). Transcriptional profiling revealed that the expression of pheromone response factor gene SsPRF1 was significantly reduced in the cAMP/PKA defective mutants ( Figure 7A), which is consistent with previous findings [38]. However, SsPRF1 gene expression was not significantly reduced in ss1sln1∆gpa3∆, ss1sln1∆uac1∆ and ss1sln1∆adr1∆ mutants, as compared to the wild type ( Figure 7A). To test the changes in the mating of double-deletion strains, we assessed the mating of MAT-1, ss1sln1∆ and cAMP/PKA single-deletion mutants (ss1gpa3∆, ss1uac1∆ and ss1adr1∆) and double-deletion strains (ss1sln1∆gpa3∆, ss1sln1∆uac1∆ and ss1sln1∆adr1∆) after mixing with MAT-2 sporidia on minimal medium. The results display that the mating capability of ss1sln1∆gpa3∆, ss1sln1∆uac1∆ and ss1sln1∆adr1∆ mutants could be partially restored in the ss1gpa3∆, ss1uac1∆ and ss1adr1∆ mutants by mixing with compatible wild-type MAT-2 ( Figure 7B). strains, we assessed the mating of MAT-1, ss1sln1Δ and cAMP/PKA single-deletion mutants (ss1gpa3Δ, ss1uac1Δ and ss1adr1Δ) and double-deletion strains (ss1sln1Δgpa3Δ, ss1sln1Δuac1Δ and ss1sln1Δadr1Δ) after mixing with MAT-2 sporidia on minimal medium. The results display that the mating capability of ss1sln1Δgpa3Δ, ss1sln1Δuac1Δ and ss1sln1Δadr1Δ mutants could be partially restored in the ss1gpa3Δ, ss1uac1Δ and ss1adr1Δ mutants by mixing with compatible wild-type MAT-2 ( Figure 7B).
Pathogenicity analysis was also performed for assessing the virulence in wild-type, cAMP/PKA-defective mutants, and double-deletion strains. The results show that the cAMP/PKA-defective mutants failed to induce the disease symptoms in sugarcane seedlings, as previously described [38]. Contrarily, typical disease 'black whip' symptoms were observed in the double-deletion strains and WT-infected seedlings ( Figure 7C). Statistical analysis depicted that the double-deletion strains presented the disease symptoms in more than 40% of the seedlings, whereas 55% of wild-type seedlings exhibited disease symptoms ( Figure 7D).
Taken together, these data suggest that cross-talk between histidine kinase SsSln1 and cAMP/PKA pathways regulates the mating and virulence by affecting the transcription of the SsPRF1 gene.  The RT-qPCR assay was performed to assess the expression of SsPRF1 gene in MAT-1, ss1gpa3∆, ss1sln1∆gpa3∆, ss1uac1∆, ss1sln1∆uac1∆, ss1adr1∆, and ss1sln1∆adr1∆ strains under sporidial growth on YePSA plate for 30 h, respectively. The relative gene expression was calculated by following the −∆∆Ct method with ACTIN as an internal control. Bar chart depicts the statistical difference among the mean values ( p < 0.01). The NS represents no significance. Mean ± S.E. were derived from three independent biological repeats with three replications. (B) Mating assay of mutants. Sporidia from MAT, ss1sln1∆, ss1gpa3∆, ss1sln1∆gpa3∆, ss1uac1∆, ss1sln1∆uac1∆, ss1adr1∆ and ss1sln1∆adr1∆ strains were separately mixed with MAT-2 sporidia of equal volume and spotted onto minimal medium to incubate at 28 • C. Images were taken after 30 h of cultivation. Scale bar = 0.5 mm. (C) Pathogenicity assay of mutants. Sporidia from MAT-1, ss1gpa3∆, ss1sln1∆gpa3∆, ss1uac1∆, ss1sln1∆uac1∆, ss1adr1∆ and ss1sln1∆adr1∆ strains were separately mixed with an equal volume of MAT-2 sporidia and inoculated into the sugarcane seedlings of variety ROC22. The infection assays were performed with at least 15 seedlings. The symptoms of 'black whip' were documented and photographed at about six months post inoculation. The red dotted box regions were enlarged for a better view of whip symptoms. The symptoms of 'black whip' are denoted by red arrows. (D) Bar chart depicts the quantification of infection as shown in (C). Percentage (%) of 'black whip'/total seedlings is indicated. Bar chart depicts the statistical difference among the mean values ( p < 0.05, p < 0.001). Mean ± S.E. were derived from two independent biological repeats with three replications.
Pathogenicity analysis was also performed for assessing the virulence in wild-type, cAMP/PKA-defective mutants, and double-deletion strains. The results show that the cAMP/PKA-defective mutants failed to induce the disease symptoms in sugarcane seedlings, as previously described [38]. Contrarily, typical disease 'black whip' symptoms were observed in the double-deletion strains and WT-infected seedlings ( Figure 7C). Statistical analysis depicted that the double-deletion strains presented the disease symptoms in more than 40% of the seedlings, whereas 55% of wild-type seedlings exhibited disease symptoms ( Figure 7D).
Taken together, these data suggest that cross-talk between histidine kinase SsSln1 and cAMP/PKA pathways regulates the mating and virulence by affecting the transcription of the SsPRF1 gene.

Discussion
The two-component histidine kinases of fungi function as sensor proteins to mediate signal transduction events related to morphogenesis, cell growth, mycelium development, cell wall regulation, osmotic adaptation and virulence [7,15,46,47]. Ten histidine kinases have been reported in M. oryzae [6], one (Sln1) in S. cerevisiae [3,48] and three (Sln1, Hk1 and Nik1) in C. albicans [3,14]. During this study, a histidine kinase Sln1 was identified and functionally characterized in S. scitamineum. However, SLN1 is not essential for the growth of S. scitamineum, which contradicts the previous reports of S. cerevisiae [7,8,49].
Interestingly, our results show that the SsSln1 null mutant exhibited a stronger mating activity and increased virulence in S. scitamineum, which is opposite to what has been reported in C. albicans [14]. One question raised by this study is why enhanced mating and virulence occur in the SsSln1 null mutant. The model for regulating the mating and virulence after the loss of SsSln1 is summarized in Figure 8. In this study, an increased level of SsHog1 phosphorylation in the ss1sln1∆ mutant was found, indicating that SsSln1 negatively regulates SsHog1 phosphorylation in S. scitamineum. This is in line with previous reports about S. cerevisiae [7,48], which revealed the activation of Hog1 kinase through disruption of the SLN1 gene. However, ss1sln1∆ mutant displayed enhanced SsHog1 phosphorylation, mating and virulence that was similar to the Ypd1 null mutant of C. albicans, which blocked the YPD1 gene to enhance Hog1 phosphorylation and virulence [13]. This suggests that SsSln1-mediated SsHog1 might regulate the mating and virulence of S. scitamineum. The Hog1 kinase was reported to regulate virulence in various pathogenic fungi, such as C. albicans [49], C. neoformans [50,51] and Aspergillus fumigatus [9,52,53], but not in M. oryzae [54]. This study also elaborated that the elimination or reduction of SsHog1 phosphorylation affected the mating and virulence of S. scitamineum. Atf1 ortholog was also identified and characterized during the study, which was reported in M. oryzae [29,55] and C. neoformans [30] to suppress the hyperphosphorylation of Hog1 ortholog. As expected, the deletion of SsATF1 enhanced the phosphorylation of SsHog1, as compared to the wild-type and Ss1ATF1-COM strains. In M. oryzae, the Atf1 is required for virulence [29,55]. We found that SsAtf1 null mutant enhanced the mating and increased relative fungal biomass in inoculated seedlings in comparison to wild type. Nevertheless, the potential role of SsAtf1 in enhancing S. scitamineum virulence should be further investigated before assessing the pathogenicity with inoculation. In short, the results of this study demonstrate that the phosphorylation of SsHog1 was required for the mating and virulence of S. scitamineum.
On the other hand, several signal transduction pathways may interact with each other in fungi [37]. A significant down-regulation of pheromone response factor SsPRF1 gene in cAMP/PKA-defective mutants and sskpp2∆ mutant was reported in our previous study [38,56]. Moreover, the exogenous addition of cAMP could partially restore the mating defect in sskpp2∆ mutant [56], suggesting that there are essential connections between the cAMP/PKA and Kpp2 MAPK pathways in S. scitamineum. In this study, our findings reveal an intriguing underlying relationship between the histidine kinase SsSln1 and cAMP/PKA pathways during mating and virulence. We found that the histidine kinase SsSln1 did not interact with the regulatory subunit SsUbc1, catalytic subunit SsAdr1 of cAMP-dependent protein kinase A (PKA) and pheromone response factor SsPrf1 by yeast two-hybrid analy-sis. This is inconsistent with the results reported in M. oryzae, where interaction between Sln1 and the regulatory subunit of cAMP-dependent protein kinase A (Sum1) was demonstrated by yeast two-hybrid analyses and co-immunoprecipitation [15]. In addition, the abolition of SsHog1 phosphorylation led to the down-regulated transcription of the SsPRF1 gene. Contrarily, the enhanced SsHog1 phosphorylation up-regulated SsPRF1 transcription, suggesting that histidine kinase SsSln1 negatively regulated SsHog1 phosphorylation to activate the transcription of the SsPRF1 gene. However, the mechanism of SsHog1 phosphorylation that regulates the transcription of the SsPRF1 gene requires further elaboration. The pheromone response factor SsPrf1 was reported to play a key role in the virulence factor of S. scitamineum [39]. During this study, we found that the mating could be fully restored in Ss1PRF1-OE(hog1∆) mutant, implying that the down-regulation of the SsPRF1 gene decreased the mating ability of ss1hog1∆ mutant. The transcription of the SsPRF1 gene was also partly restored in the double-deletion strains (ss1sln1∆gpa3∆, ss1sln1∆uac1∆ and ss1sln1∆adr1∆), as compared to down-regulation in cAMP/PKA-defective mutants, as previously described [38]. Meanwhile, the double-deletion strains exhibited remarkably enhanced mating and virulence capabilities in comparison to the cAMP/PKA-defective mutants. Taken together, we infer that cross-talk between the SsSln1 and cAMP/PKA pathways antagonistically regulates mating and virulence by balancing the transcription of the SsPRF1 gene in S. scitamineum. two-hybrid analysis. This is inconsistent with the results reported in M. oryzae, where interaction between Sln1 and the regulatory subunit of cAMP-dependent protein kinase A (Sum1) was demonstrated by yeast two-hybrid analyses and co-immunoprecipitation [15]. In addition, the abolition of SsHog1 phosphorylation led to the down-regulated transcription of the SsPRF1 gene. Contrarily, the enhanced SsHog1 phosphorylation up-regulated SsPRF1 transcription, suggesting that histidine kinase SsSln1 negatively regulated SsHog1 phosphorylation to activate the transcription of the SsPRF1 gene. However, the mechanism of SsHog1 phosphorylation that regulates the transcription of the SsPRF1 gene requires further elaboration. The pheromone response factor SsPrf1 was reported to play a key role in the virulence factor of S. scitamineum [39]. During this study, we found that the mating could be fully restored in Ss1PRF1-OE(hog1Δ) mutant, implying that the downregulation of the SsPRF1 gene decreased the mating ability of ss1hog1Δ mutant. The transcription of the SsPRF1 gene was also partly restored in the double-deletion strains (ss1sln1Δgpa3Δ, ss1sln1Δuac1Δ and ss1sln1Δadr1Δ), as compared to down-regulation in cAMP/PKA-defective mutants, as previously described [38]. Meanwhile, the double-deletion strains exhibited remarkably enhanced mating and virulence capabilities in comparison to the cAMP/PKA-defective mutants. Taken together, we infer that cross-talk between the SsSln1 and cAMP/PKA pathways antagonistically regulates mating and virulence by balancing the transcription of the SsPRF1 gene in S. scitamineum. In short, this study provides evidence that histidine kinase SsSln1 negatively regulates SsHog1 phosphorylation, which is essential for mating and virulence in S. scitamineum. Furthermore, we also reveal a novel mechanism by which histidine kinase SsSln1 and cAMP/PKA pathways antagonistically regulate mating and virulence via affecting the transcription of the SsPRF1 gene. In short, this study provides evidence that histidine kinase SsSln1 negatively regulates SsHog1 phosphorylation, which is essential for mating and virulence in S. scitamineum. Furthermore, we also reveal a novel mechanism by which histidine kinase SsSln1 and cAMP/PKA pathways antagonistically regulate mating and virulence via affecting the transcription of the SsPRF1 gene.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/jof7080610/s1, Figure S1: Identification of mutants by PCR amplification, Figure S2: Southern blot and RT-qPCR analysis of the mutant, Table S1: The primers and sequences used in this study.

Informed Consent Statement: Not applicable.
Data Availability Statement: All data required to understand this article are presented in the study or the Supplementary Materials. Any raw data further requested will be provided by the corresponding authors.