The Sch9 Kinase Regulates Conidium Size, Stress Responses, and Pathogenesis in Fusarium graminearum

Fusarium head blight caused by Fusarium graminearum is an important disease of wheat and barley worldwide. In a previous study on functional characterization of the F. graminearum kinome, one protein kinase gene important for virulence is orthologous to SCH9 that is functionally related to the cAMP-PKA and TOR pathways in the budding yeast. In this study, we further characterized the functions of FgSCH9 in F. graminearum and its ortholog in Magnaporthe oryzae. The ΔFgsch9 mutant was slightly reduced in growth rate but significantly reduced in conidiation, DON production, and virulence on wheat heads and corn silks. It had increased tolerance to elevated temperatures but became hypersensitive to oxidative, hyperosmotic, cell wall, and membrane stresses. The ΔFgsch9 deletion also had conidium morphology defects and produced smaller conidia. These results suggest that FgSCH9 is important for stress responses, DON production, conidiogenesis, and pathogenesis in F. graminearum. In the rice blast fungus Magnaporthe oryzae, the ΔMosch9 mutant also was defective in conidiogenesis and pathogenesis. Interestingly, it also produced smaller conidia and appressoria. Taken together, our data indicate that the SCH9 kinase gene may have a conserved role in regulating conidium size and plant infection in phytopathogenic ascomycetes.


Introduction
Fusarium graminearum (teleomorph Gibberella zeae) is one of the major causal agents of wheat and barley head blight or scab in North America and other parts of the world [1,2]. It also infects corn and other small grains. Like many other Fusarium species, F. graminearum produces several mycotoxins that are harmful to human and animals [3]. Deoxynivalenol (DON) is a trichothecene mycotoxin produced by this pathogen that is of major health concerns and monitored by many governments.
As a potent inhibitor of eukaryotic protein synthesis, DON is also a phytotoxin. In fact, the first virulence gene identified in F. graminearum is TRI5 that encodes the trichodiene synthase essential for DON biosynthesis [4,5]. Since the release of its genome sequence [6], molecular genetics and genomics studies of F. graminearum have advanced significantly. Expression profiling studies have identified a variety of genes with different expression levels during plant infection [7,8]. Genes of various biochemical or biological functions have been identified as important virulence or pathogenicity factors in F. graminearum, including other genes involved in DON biosynthesis [9,10], components of key signal transduction pathways [11,12,13,14], transcription factors of different DNA-binding domains [15,16,17], and several enzymes required for primary metabolism [18,19,20]. Like many other filamentous ascomycetes, F. graminearum has three mitogenactivated protein (MAP) kinase pathways, and all of them play critical roles in pathogenesis and DON production [21,22]. It also has the well-conserved cAMP signaling pathway that is involved in the switching from saprophytic growth to infectious growth [23]. Two genes encoding the catalytic subunits of PKA have distinct and overlapping functions in regulating developmental and plant infection processes in F. graminearum [24].
A total of 42 protein kinase genes were found to be important for plant infection by systemic characterization of the F. graminearum kinome [21]. One of them, FGSG_00472, is orthologous to SCH9 of the budding yeast. The Sch9 protein kinase shares sequence similarity with the catalytic subunits of PKA and it is functionally related to cAMP signaling in response to nutrient availability in Saccharomyces cerevisiae. It inhibits PKA activity by regulating the localization of Tpk1/2/3 and stability of Tpk2 [25]. Disruption of SCH9 increases PKA activities [25] and stress tolerance [26]. Similar to mutations in RAS2 and CYR1, down-regulation of glucose signaling by deletion of SCH9 increases longevity and resistance to oxidative stress and heat shock [27]. In S. cerevisiae, SCH9 also is a master regulator of protein synthesis. It is phosphorylated by TORC1 to regulate TORC1-dependent cellular processes, such as ribosome produc-tion and translation [28,29]. The Sch9 kinase also is involved in the regulation of autophagy together with the TORC1 and cAMP-PKA pathways [30].
Although SCH9 orthologs are well conserved in plant pathogenic fungi or filamentous ascomycetes, none of them have been functionally characterized. Considering the diverse functions of SCH9 in S. cerevisiae and the importance of cAMP signaling in F. graminearum, in this study we further characterized the Fgsch9 deletion mutant generated in the systemic characterization of the F. graminearum kinome [21]. Although it was only slightly reduced in growth rate, the DFgsch9 mutant was significantly reduced in DON production and virulence. It had increased tolerance to elevated temperatures but increased sensitivities to oxidative, hyperosmotic, cell wall, and membrane stresses. The DFgsch9 deletion mutant also was defective in conidiogenesis and produced smaller conidia. In the rice blast fungus Magnaporthe oryzae, the DMosch9 deletion mutant also was defective in conidiogenesis and pathogenesis. Interestingly, it also produced smaller conidia and appressoria. These results indicate that the SCH9 kinase gene may have a conserved role in regulating conidium size and plant infection in plant pathogenic fungi.

Deletion of the FgSCH9 gene reduces hyphal growth and conidiation
In the functional study of the F. graminearum kinome, we noticed that the mutant deleted of FGSG_00472.3 was defective in plant infection [21]. This gene (named FgSCH9) encodes a serine/threonine protein kinase that is highly similar to yeast SCH9. It is well conserved in filamentous fungi and shares limited homology with MEK kinases of the MAP kinase pathways [31]. In this study, we first confirmed the Fgsch9 mutant generated in an earlier study [21] by Southern blot analyses. When hybridized with a FgSCH9 fragment amplified with primers F5 and R6, the wild type had the 4.4-kb EcoRI band but the DFgsch9 mutant SD1 had no hybridization signals. When probed with the hph fragment, PH-1 had no hybridization signals. Mutant SD1 had a 2.8-kb band, which is the expected size derived from the gene replacement event (Fig. S1).
The DFgsch9 mutant was reduced in growth rate and conidiation (Table 1). On PDA agar plates, mutant SD1 produced less aerial hyphae than PH-1 (Fig. 1A). We also noticed that mutant SD1 was reduced in hyphal branching at the edges of colonies formed on 1/2 CM for 36 h in comparison with PH-1 (Fig. 1B). In addition, 70.467.9% of the DFgsch9 conidia had morphological defects. Instead of forming typical Fusarium conidia, the DFgsch9 mutant produced conidia that lacked foot cells or had fewer septa (Fig. 1C) The DFgsch9 mutant has increased tolerance to elevated temperatures To test whether the DFgsch9 mutant has increased tolerance to elevated temperatures, we assayed germ tube growth after incubation in YEPD for 18 h at 30uC. Similar to incubation at 25uC ( Fig. 2A), both PH-1 and DFgsch9 mutant SD1 germinated and produced germ tubes at 30uC. However, germ tubes of PH-1 often had apical and intercalary swollen bodies when incubated at 30uC but not at 25uC (Fig. 2B). Germ tube growth was normal in the DFgsch9 mutant SD1 at either 25 or 30uC. These observations suggest that deletion of FgSCH9 increased tolerance to elevated temperatures, which is similar to what was observed in S. cerevisiae [26].
The DFgsch9 mutant is defective in responses to various environmental stresses To test whether deletion of FgSCH9 affects responses to other stresses, we assayed colonial growth of the DFgsch9 mutant on CM plates with four different chemicals that are representative of oxidative, cell wall, hyperosmotic, and membrane stresses. In comparison with the wild type, mutant SD1 produced significantly smaller colonies after incubation for 3 days in the presence of 0.05% H 2 O 2 , 300 mg/ml Congo red, 0.7 M NaCl, or 0.01% SDS (Fig. 3A). The DFgsch9 mutant had restricted hyphal growth, particularly on CM with 0.01% SDS. These results indicated that the DFgsch9 mutant had increased sensitivities to oxidative, membrane, hyperosmotic, and cell wall stresses. FgSCH9 must be involved in responses to general environmental stresses in F. graminearum. In the budding yeast, the DFgsch9 mutant has increased sensitivities to oxidative and osmotic stresses [26,32].
To confirm its defects in response to hyperosmotic stress, we assayed conidium germination and germ tube growth in the DFgsch9 mutant in YEPD with different concentrations of NaCl. After incubation at 25uC for 12 h, germ tube growth was similar between mutant SD1 and PH-1 in the presence of 0.3 M NaCl (Fig. 3B). However, although germination was not inhibited, germ tube growth was significantly reduced by 0.7 M or 1 M NaCl in mutant SD1 but not in PH-1 (Fig. 3B). Therefore, germ tube growth, similar to hyphal growth, is sensitive to hyperosmotic stress in the DFgsch9 mutant.
New hyphal growth is observed inside dead hyphae of the DFgsch9 mutant Interestingly, old hyphae of the DFgsch9 mutant often had empty, dead intercalary compartments, which may be related to its defects in cell wall integrity. In hyphae from YEPD or PDA cultures, we were able to observe new hyphal growth inside the empty compartments (Fig. 4). Close examination indicated that new hyphal growth emerged through the septal pore area from flanking compartments that were still alive (Fig. 4). These observations indicate that the DFgsch9 mutant must be able to plug up the septal pore when some hyphal compartments were damaged or became empty. However, this plug up is incomplete or reversible and the DFgsch9 mutant can re-establish polarized growth through the plugged septal pore area.

The DFgsch9 mutant is defective in plant infection
To determine the role of FgSCH9 in pathogenesis, we inoculated flowering wheat heads with conidia from PH-1 and mutant SD1. The spikelets of susceptible wheat cultivar Xiao-yan22 drop-inoculated with the DFgsch9 mutant still developed typical wheat scab symptoms. However, mutant SD1 was defective in spreading from the inoculated kernels to other spikelets on the same wheat heads. At 14 days post-inoculation (dpi), the average disease index (diseased spikelets per head) was 5.6 for mutant SD1 and 14.7 for PH-1 (Table 1), which was approximately a 62% reduction in virulence. Because it is an important virulence factor [4,5], we also assayed DON production in diseased wheat kernels harvested at 14 dpi. In comparison with PH-1, the DFgsch9 mutant also was reduced in DON production ( Table 1).
F. graminearum also is a corn pathogen and it can cause Gibberella ear rot and stalk rot. In infection assays with corn silks, both PH-1 and mutant SD1 were able to colonize and cause discoloration. However, the extent of discoloration was significantly reduced in the DFgsch9 mutant in comparison with the wild type (Fig. 5B). On corn silks inoculated with mutant SD1, discoloration was restricted to the inoculation site. Based on data from three independent experiments, the average discolored region on corn silks inoculated with PH-1 and the Fgsch9 mutant was 47.664.8 mm and 16.763.7 mm, respectively. These results further indicate that FgSCH9 is an important virulence factor in F. graminearum.

Expression of FgSCH9-GFP fully complemented the defects of mutant SD1
For complementation assays, we generated the FgSCH9-GFP fusion construct under its own promoter control and introduced it into the DFgsch9 mutant SD1. Strain N9 was one of the resulting DFgsch9/FgSCH9-GFP transformants identified by PCR assays. It was normal in growth rate, conidiation, and conidium morphology ( Fig. 1; Table 1). The defects of mutant SD1 in stress responses and plant infection also were rescued in transformant N9 ( Fig. 3;  Fig. 5). These results indicate that deletion of FgSCH9 is directly responsible for all the phenotypes of the DFgsch9 mutant.
To determine the expression and subcellular localization of FgSCH9 in different growth and developmental stages, we examined GFP signals in transformant N9 by epifluorescence microscopy. In conidia and germ tubes, GFP signals of similar strength were observed in the cytoplasm (Fig. 6). Similar results were obtained in hyphae and ascospores. These data indicate that FgSCH9 was expressed in all the cell types examined and it had no distinct subcellular localization pattern in F. graminearum, which is similar to the localization of Sch9 in S. cerevisiae [33].
The Mosch9 mutant also was reduced in conidium size and virulence To verify the function of FgSCH9 in controlling cell size and virulence, we identified its ortholog in the rice blast fungus M. oryzae, MGG_14773.6 that was named MoSCH9 in this study. The MoSCH9 gene replacement construct was generated by the ligation-PCR approach [34] and transformed into protoplasts of strain Ku80 [35]. The DMosch9 deletion mutant also was reduced in conidium size (Fig. 7A). Whereas the average size of wild-type conidia was 18.161.566.060.7 mm (length6width), conidia of the DMosch9 mutant were 15.561.765.160.6 mm. Interestingly, appressoria formed by the DMosch9 mutant on hydrophobic surfaces also were smaller than those of strain Ku80 (Fig. 7A). On average, the diameter of mutant appressoria was 6.960.4 mm.
The average size of wild-type appressoria was 8.460.5 mm. In infection assays with rice leaves, the DMosch9 mutant also was reduced in virulence (Fig. 7B). In comparison with Ku80, the mutant produced fewer lesions (Fig. 7B). These results indicate that the SCH9 orthologs may have conserved functions in regulating cell size and plant infection in fungal pathogens.

Discussion
In the budding yeast, the Sch9 kinase is functionally related to the cAMP-signaling and TORC1 pathways [25,28,30]. These two well-conserved pathways recently were shown to be involved in various development and infection processes in F. graminearum [23,24,36]. Although Sch9 orthologs are well conserved in filamentous ascomycetes, none of them have been functionally characterized. In this study, we found that the DFgsch9 mutant   The Sch9 Gene in the Wheat Scab Fungus PLOS ONE | www.plosone.org had a reduced growth rate and produced smaller conidia, which is similar to the sch9 deletion mutant of S. cerevisiae that exhibits slow growth and small cell size [33,37,38,39]. In M. oryzae, conidia of the DMosch9 mutant also were smaller than those of the wild type. Even the appressoria produced by the DMosch9 mutant were smaller. In C. albicans, the cell size of the CaSCH9 deletion mutant was significantly reduced in comparison with that of the wild type strain [40]. Therefore, SCH9 orthologs may have a conserved function in regulating conidium size in filamentous ascomycetes. In S. cerevisiae, SCH9 and SFP1 are required for carbon-source modulation of cell size [33].
The TOR complex is well-conserved from yeast to human and it is involved in regulating growth-related signaling. It controls growth in response to nutrients by regulating translation, transcription, ribosome biogenesis, nutrient transport, and autophagy [29,30,41,42]. In S. cerevisiae, SCH9 functions downstream from the TOR pathways for properly regulating ribosome biogenesis, translation initiation, and entry into G 0 phase [28]. In F. graminearum, the TOR pathway plays critical roles in regulating vegetative differentiation and virulence [36]. It is likely that FgSCH9 is also functionally related to the TOR pathway in F. graminearum. Unlike S. cerevisiae that has TOR1 and TOR2, two TOR kinase genes, F. graminearum has a single essential TOR kinase gene [21]. It will be interesting to determine the functional relationship between FgTor1 and FgSch9 by assaying the phosphorylation level or activity of FgSch9 in rapamycintreated samples.
In the budding yeast, the Sch9 kinase is functionally related to the Ras-cAMP signaling pathway and it shares a large number of phosphorylation targets with PKA, such as Hog1 and Pfk2 [43].
Sch9 inhibits PKA activity and disruption of SCH9 increases PKA activities [25]. In the fission yeast Schizosaccharomyces pombe, Sck1 and Sck2 are homologous to Sch9. Overexpression of the sck1 or sck2 gene can suppress the loss of PKA activity [44,45]. F. graminearum has only one Sch9 ortholog but two genes encoding catalytic subunits of PKA. Whereas deletion of CPK2 had no detectable phenotype, the cpk1 mutant was reduced in growth, virulence, DON production, and conidiogenesis [24]. The cpk1 cpk2 double mutant had more severe defects in growth and was non-pathogenic [24]. In addition, Sch9 and PKA control parallel pathways that converge on the Rim15 kinase in S. cerevisiae [41,46]. In M. oryzae, the rim15 deletion mutant has a slower growth rate, slightly increased sensitivity to hyperosmotic stress, and reduced hyphal melanization and virulence [47]. In F. graminearum, the Fgrim15 mutant was reduced in conidiation and virulence [21].
In S. cerevisiae, SCH9 also is involved in the signal transduction facilitator function of the Hsp90 chaperone complex that is required for the maturation of hundreds of diverse client proteins [48]. The sch9 mutant has increased stress resistance [26]. In human pathogen Cryptococcus neoformans, the sch9 deletion mutant also had increased thermal tolerance [49]. In F. graminearum, the DFgsch9 mutant had increased tolerance to elevated temperatures during germ tube growth. Germ tubes of the DFgsch9 mutant were normal but the wild type produced apical and intercalary swollen bodies when incubated at 30uC.
In F. graminearum, the DFgsch9 mutant had increased sensitivity to 0.7 M NaCl, suggesting that FgSCH9 is important for response to hyperosmotic stress. In the budding yeast, Sch9 is involved in response to hyperosmotic stress via its association with Hog1 and co-regulation of subsets of genes such as GRE2 and CTT1 [32]. Nevertheless, the DFgsch9 mutant had increased sensitivity to SDS and Congo red as well. In addition, we found that the DFgsch9 mutant had the hyphae-in-hyphae phenotype and increased tolerance to elevated temperatures. Therefore, it is likely that FgSCH9 is required for general stress responses in F. graminearum.
Unlike the cpk1 cpk2 double mutant [24], the DFgsch9 mutant was still pathogenic. However, it was significantly reduced in virulence on flowering wheat heads and corn silks. Although SCH9 orthologs are well-conserved in plant pathogenic fungi or filamentous ascomyctes, none of them have been characterized. However, its ortholog is known to be important for virulence in human pathogens C. albicans and C. neoformans. The CaSCH9 deletion mutant was attenuated in virulence in a mouse mode of systemic candidiasis due to its defects in yeast growth and filamentation [40]. In C. neoformans, the SCH9 ortholog functions both independently of and in conjunction with the cAMP-PKA pathway in pathogenesis [49]. In F. graminearum, DON is an important virulence factor and the DFgsch9 mutant was significantly reduced in DON production in infected wheat kernels. In addition, the DFgsch9 mutant had a reduced growth rate and increased sensitivity to oxidative and other stresses. All these factors may contribute to the defects of the DFgsch9 mutant in plant infection.
The DFgsch9 mutant had increased sensitivity to cell wall stresses, indicating defects in cell wall integrity. However, unlike the mgv1 mutant [22], deletion of FgSCH9 had no effect on hyphal fusion in F. graminearum. Interestingly, old hyphae of the DFgsch9 mutant often had empty, dead intercalary compartments, likely due to its defects in cell wall integrity. In some of them, new, narrow hyphae were produced from the middle of septa and grew into the empty, dead old hyphae. It appears that the DFgsch9 mutant can plug up the septal pore when hyphal compartments become damaged. However, it is defective in this process and somehow can regain polarized growth through the plugged septal pore areas. To our knowledge, this phenomenon of new hyphal growth inside old hyphal compartments has not been reported in F. graminearum. Recently, we found that the tub2 deletion mutant also had similar defects (Zhang and Xu, unpublished). It will be interesting to determine the molecular mechanisms involved in the regulation of septal pore plugging and further growth inside empty hyphal fragments or re-establishment of hyphal tip growth at the septal pore.

Strains and culture conditions
The wild-type strain PH-1 and all the transformants of F. graminearum generated in this study were routinely cultured on PDA agar plates [50]. Growth rate and conidiation were assayed as described [24,51]. DNA was extracted from vegetative hyphae harvested from liquid YEPD (1% yeast extract, 2% peptone, 2% glucose). Protoplast preparation and PEG-mediated transformation were performed as described [22]. Hygromycin B (Calbiochem, La Jolla, CA) and geneticin (Sigma, St. Louis, MO) were added to the final concentrations of 300 and 350 mg/ml, respectively, to the TB3 medium (0.3% yeast extract, 0.3% casamino acids, 3% glucose) for transformant selection. To test sensitivity against various stresses, vegetative growth was assayed on CM plates with 0.05% H 2 O 2 , 0.01% SDS, 300 mg/ml Congo Red, or 0.7 M NaCl as described [21,52]. For assaying conidium germination and germ tube growth, conidia were shaken in YEPD (5610 4 conidia/ml) at 175 rpm for 12 h before examination.

Generation of FgSCH9-GFP fusion constructs and complementation strains
To generate the SCH9-GFP fusion constructs, PCR products containing the genomic fragments of the target gene and its promoter sequence were amplified with primers GFPNAT/F and GFPNAT/R (Table S1) and cloned into pFL2 by the yeast gap repair approach [55,56]. The FgSCH9-GFP fusion construct was confirmed by sequencing analysis and transformed into protoplasts of the DFgsch9 mutant. Geneticin-resistant transformants harboring the transforming FgSCH9-GFP construct were identified by PCR and confirmed by the presence of GFP signals. The DFgsch9/gSCH9-GFP complemented transformant N9 was selected for phenotype analysis and subcellular localization analysis of GFP signals.

Infection with F. graminearum and DON production assays
For infection assays with flowering wheat heads of cultivars XiaoYan 22 or Norm, conidia were harvested from 5-day-old CMC cultures and re-suspended in sterile distilled water to 2.0610 5 conidia/ml [57,58]. The fifth spikelet from the base of the spike was inoculated with 10 ml of conidium suspensions as described [59]. Inoculated wheat heads were capped with a plastic bag to maitain humidity for 48 h. After removing the bags, wheat plants were cultured for another 12 days before examination for disease symptoms. The disease index for each strain was estimated by counting diseased spikelets per wheat head at 14 dpi. The inoculated kernels with typical disease symptoms were harvested and assayed for DON production as described [60]. Infection assays with corn silks of cultivar Pioneer 2375 were conducted as described [18]. The extent of discoloration on infected corn silks was measured after incubation at 25uC in a moisture chamber for 5 days [18].

Assays for conidium morphology and size
Conidia were harvested from 5-day-old CMC cultures as described [22] and examined by DIC microscopy. The ones that lacked foot cells or had fewer than four septa were considered to be defective in conidium morphology. For both PH-1 and the Fgsch9 mutant, conidium morphology assays were repeated three times and 400 conidia were examined in each replicate. To compare their differences in conidium size, the length and width of conidia of PH-1 and the Fgsch9 mutant were measured in three independent experiments with at least 80 conidia measured per replicate. The average number of septa in conidia was calculated with data from three experiments of counting septa in 150 conidia per replicate. The resulting data were subjected to analyses of variance (ANOVA) with the SPSS17.0 statistics software package (SPSS Inc., Chicago, IL). Comparisons of the means were analyzed with the protected Fisher's Least Significant Difference (LSD) test (P = 0.05).

Targeted deletion of the MoSCH9 gene in M. oryzae
To delete the MoSCH9 gene, its upstream and downstream flanking sequences were amplified with primer pairs MoSCH9-1F/MoSCH9-2R and MoSCH9-3F/MoSCH9-4R, digested with FesI and AscI, respectively. The resulting PCR products were digested and ligated with the hygromycin resistance cassette (hph) released from pCX63 as described [34]. The gene replacement construct was amplified with nest primers MoSCH9-1F'/ MoSCH9-4R' (Table S1) and directly transformed into protoplasts of Ku80. Hygromycin-resistant transformants were screened by PCR with primers MoSCH9-NF/MoSCH9-NR (Table S1) and confirmed by Southern blot analysis.

Appressorium formation and plant infection with the DMosch9 mutant
Conidia harvested from 10-day-old oatmeal agar cultures were re-suspended to 10 5 conidia/ml in distilled water for appressorium formation assays on plastic coverslips as described [61,62]. Twoweek-old seedlings of rice cultivar CO-39 were used for infection assays with conidia re-suspended 0.25% gelation as described [63]. Disease symptoms were observed 7 dpi.

Supporting Information
Fig. S1 Generation and Southern blot analysis of the DFgsch9 mutant. A. The genomic region of the FgSCH9 gene and PCR fragments used for generation of the gene replacement construct and Southern blot hybridization. PCR primers were marked with small arrows. E, EcoRI. B. Southern blots of EcoRI-digested genomic DNA of the wild-type strain (PH-1) and DFgsch9 mutant (SD1) were hybridized with an FgSCH9 (probe 1) or hph fragment (probe 2). Whereas probe 1 hybridized to a 4.4-kb band in PH-1 but not in SD1, probe 2 detected a 2.8-kb band in SD1 that was diagnostic of the gene replacement event. (TIF)