Functional Characterization of the M36 Metalloprotease FgFly1 in Fusarium graminearum

Fungalysin metallopeptidase (M36), a hydrolase, catalyzes the hydrolysis of alanine, glycine, etc. Normally, it is considered to play an important role in the progress of fungal infection. However, the function of fungalysin metallopeptidase (M36) in Fusarium graminearum has not been reported. In this study, we explored the biological functions of FgFly1, a fungalysin metallopeptidase (M36) of F. graminearum. We found that ΔFgFly1 did not affect the ability to produce DON toxin, although it inhibited spore germination during asexual reproduction and reduction in pathogenicity compared with PH-1. Therefore, we speculated that FgFly1 affects the pathogenicity of F.graminearum by affecting pathways related to wheat disease resistance. Target protein TaCAMTA (calmodulin-binding transcription activator) was selected by a yeast two-hybrid (Y2H) system. Then, the interaction between FgFly1 and TaCAMTA was verified by bimolecular fluorescent complimentary (BiFC) and luciferase complementation assay (LCA). Furthermore, compared with wild-type Arabidopsis thaliana, the morbidity level of ΔAtCAMTA was increased after infection with F. graminearum, and the expression level of NPR1 was significantly reduced. Based on the above results, we concluded that FgFly1 regulated F. graminearum pathogenicity by interacting with host cell CAMTA protein.


Introduction
Wheat, an important food crop worldwide, is threatened by various diseases during its life cycle. Fusarium head blight (FHB) is a worldwide disease caused by Fusarium fungi, which can result in a yield loss of more than 30% in epidemic years [1][2][3]. Moreover. the development of FHB is accompanied by the accumulation of two main mycotoxins, deoxynivalenol (DON) and zearalenone (ZEA) [4,5], which can significantly reduce the quality of grains, threatening the health of humans and livestock [6]. At present, management of FHB remains challenging, mainly because a germplasm with complete resistance or immunity to F. graminearum is lacking, and the main control method for F. graminearum is currently the use of azole fungicides during the flowering stage of wheat. However, these methods are costly and time-consuming. Therefore, it is essential to explore the pathogenic and molecular mechanisms of F. graminearum [7].
When a pathogen invades a host plant, the pathogen secretes effectors into the plant, which induces a series of disease-resistance responses [8]. In a narrow sense, the effectors in phytopathogenic fungi are the proteins secreted from pathogens into extracellular and intracellular spaces of host plants [9][10][11][12]. A generalized definition of effectors is "proteins and small molecules that alter host cell structures and functions, thereby facilitating the colonization of pathogens" [13].
A common strategy of pathogenic fungi is to degrade host structural barriers by secreting extracellular enzymes [14][15][16]. Elastase metalloproteinase plays an important role M36 metalloproteinase in F. graminearum and wheat CAMTA protein, we conducted a related study.
In this study, we investigated the function of M36 metalloprotease FgFly1 in F. graminearum. We found that FgFly1 interacted with wheat TaCAMTA to reduce the disease resistance of the host wheat to achieve the purpose of infection. In addition to its virulence function, FgFly1 also affected the spore germination of F. graminearum asexual reproduction and the susceptibility of metal ion stress. We believe that FgFly1 plays an important role in the growth, development and infection of F. graminearum.

Culture Conditions for Plants and Fungi
Nicotiana benthamiana plants were grown in a glasshouse (24 • C, 70% relative humidity, 16 h light photoperiod). F. graminearum strain PH-1 (NRRL 31084), as wild type, was used to construct gene-deletion mutants. Wild-type strains, gene-deletion mutants and complementary strains were cultured on potato dextrose agar medium (PDA), trace-element minimal medium (MM) and complete medium (CM) at 25 • C, and mycelial growth was observed. In a spore reproduction experiment, 5 young fungus dishes from the periphery of 3-day-old colonies were inoculated into 50 mL conical flasks containing 30 mL CMC medium. After 4 days of culture in a shaker (180 rpm) at 25 • C, the number of conidia in each flask was determined under a microscope with a hemocytometer [37]. Each experiment was repeated 3 times. All strains were stored as mycelium in 15% glycerol at −80 • C.

Construction of Gene-Deletion and Complementary Mutants
Construction of gene-deletion and complementary vectors and subsequent transformation of F. graminearum was performed by double-joint PCR techniques [38] and polyethylene glycol (PEG)-mediated protoplast transformation [39]. Gene-deletion mutants were identified by PCR detection using relevant primers and further analyzed by Southern blot. We used a designable primer to amplify fragments (Table S1).

Construction of Green Fluorescent Protein (GFP) Vector
In order to construct a FgFly1-GFP vector, the PCR product and the pYF11-GFP-GEN vector digested by XhoI were cotransferred into yeast XK1-25, and the yeast plasmid pYF11-FgFly1-GFP-GEN was obtained. The yeast plasmid pYF11-FgFly1-GFP-GEN was extracted and transformed into E. coli DH5α for large-scale amplification of the recombinant plasmid [39]. The FgFly1-GFP fusion vector was transformed to observe the localization of FgFly1. The transformed complementary strains were screened and inoculated on PDA medium with 100 mg/L G418 sulfate. The corresponding primers of the complementary strains were analyzed by PCR and Southern blot. Finally, the green fluorescent protein (GFP) of the obtained strain was observed with a confocal microscope.

Sexual Reproduction and Vegetative Growth Assays
The activated strains were inoculated on carrot agar medium, and 500 µL of 2.5% Tween-20 solution was added to the aerial hyphae of each inoculated strain. The formation of an ascus membrane and the number of perithecia were measured after 15 d of culture under 25 • C black light lamps [40]. In order to determine the susceptibility of strains to stress, the strains were inoculated on PDA medium with different agents and cultured at 25 • C for 3 d; the colony diameter was measured, and the inhibition rate was calculated. The strain mycelial growth inhibition rate (MGIR) was calculated according to the formula MGIR = [(N − C)/C] × 100, where C is the diameter of the control colony, and N is the diameter of the treated colony. Each experiment was independently repeated three times.

Pathogenicity and DON Production Assays
The pathogenicity of corn filaments was analyzed, and the activated strains were inoculated in the middle of young corn filaments and cultured at 25 • C for 4 d with moderate humidity. For pathogenicity analysis of wheat spikelets, conidia of wild-type PH-1, knockout mutants and complementary strains cultured in CMC medium for 4 d were filtered and centrifuged, and the conidia concentration was adjusted to 1 × 10 5 /mL with sterile water. During the flowering stage of wheat, an equal amount of 10 µL of the conidia suspension was injected into the florets of the central part of the susceptible cultivar Xiaoyan 22 with a pipette, and 20 replicates were designed for each strain [41,42]. The wheat was placed neatly on wet filter paper and cultured for 2d at 25-30 • C until the wheat grew 2-3 cm buds. The concentration of spores was adjusted to 1 × 10 5 /mL, and the top 3-5 mm of the wheat buds was cut off with a sharp blade. The absorbent cotton was torn into pieces the size of soybeans, soaked in spore suspension for 5 s and wrapped on the wound on top of the wheat. After moisturizing the culture for 48 h, the absorbable cotton was removed, and the outer epidermis of the wheat coleoptile was removed for microscope observation or continued culture for 5 d to observe the disease situation [41,43]. An emetic toxin ELISA kit (Jiangsu Enzyme Immunity Industry Co., Ltd., YanCheng, China) was used to measure DON production.

qRT-PCR Assays
Tri genes were induced with TBI medium and treated in the dark at 25 • C and 180 rpm for 48 h. The mycelia were collected and ground in liquid nitrogen. Total RNA was extracted using RNAiso reagent (TaKaRa Co., Dalian, China), and 1 mg of each RNA sample was subjected to reverse transcription using a HiScriptII QRT Super Mix qPCR kit (Vazyme Biotech, Nanjing, China). The expression level of each gene was determined by quantitative real-time PCR with the primers listed in Table S1. Each experiment was repeated three times independently.

Yeast Two-Hybrid Assays
The gene DNA fragment was inserted into the pBT3-SUC plasmid as a bait vector. The bait and prey vector pPR3-N-cDNA library (Oebiotech, Shanghai, China) was cotransferred into yeast strain NMY51 according to the yeast protocol manual (Clontech, Shanghai, China), and positive clones on SD/-LEU-TRP-HIS-ADE/3-AT medium were selected. In order to confirm the interaction between gene and target, the recombinant pBT3-SUC gene and pPR3-N target were cotransferred into yeast strain NMY51 and grown in SD/-Leu-Trp-His-Ade/3-AT.

BiFC
The gene and target were cloned into BiFC vectors C-YFP and N-YFP, respectively. The constructed bimolecular fluorescent complementary vector was transformed into Agrobacterium tumefaciens GV3101 (pSoup-p19) competence, and the correct single clone detected by PCR was cultured in LB medium containing kanapenicillin and rifampicin resistance. The A. tumefaciens solution was kept at 28 • C, shaken at 200 rpm and centrifuged at 4000 rpm for 10 min to collect the bacteria, and the supernatant was discarded, and 10 mM MgCl 2 was added to treat the Agrobacterium solution. AS buffer was used to mix c-YFP bacterial solution with N-YFP bacterial solution and P19 bacterial solution according to the OD 600 ratio of 0.5:0.5:0.3 and then placed in the dark for 2 h before injecting N. benthamiana. Fluorescence microscopy was performed after 48-72 h [44].

LUC
The gene and target gene were constructed into n-LUC and c-LUC vectors, respectively, by one-step cloning method. The plasmid was transformed to GV3101 (pSoup-P19), and the monoclonal clones detected by PCR were cultured in LB medium containing kanapenicillin and rifampicin resistance. The A. tumefaciens solution was kept at 28 • C, shaken at 200 rpm until the OD 600 was 1.0-1.3 and centrifuged at 4000 rpm for 10 min to collect the bacteria. Then, the supernatant was discarded, 10 mM MgCl 2 was added and shaken slightly, followed by the addition of an appropriate amount of infection solution and mixing by pipetting. Then, the solution was kept in the dark for 2 h. The A. tumefaciens infection solution was mixed evenly, with a final concentration of 0.2. N. benthamiana leaves were injected, marked, cultured in the dark for 12 h and subsequently placed under normal light conditions for 24 h. Then, the N. benthamiana leaves injected with Agrobacterium were cut and spread out on the prepared 4% agar plates, and firefly luciflucase substrate was coated against light, and imaging was performed after standing in darkness for 5 min. The exposure time on the plant living molecular labeling imaging instrument was 10 min to collect fluorescence signals [45].

Statistical Analyses
Data are presented as the mean of triplicates. Significance was determined by Fisher exact test at p = 0.05.

Phylogenetic Analysis and Sequence Alignment of Different Fungal Fly1 Proteins and Adcquisition of FgFly1 Deletion Mutants
The protein encoded by the FgFly1 gene (FGSG_03467) of F. graminearum contains 631 amino acids. Phylogenetic analysis was performed using MEGA software, and sequence alignment was performed using DNAMAN. Species names and accession numbers are as follows: F. graminearum, XP_011322248.  Figure 1A). In addition, signal peptide prediction (http://www.cbs.dtu. dk/services/SignalP/ accessed on 2 December 2021) Fly1 protein has a 23aa N-terminal signal peptide. Homology alignment (http://pfam.xfam.org/ accessed on 2 December 2021) showed that the FTP domain and peptidase M36 domain, as well as a characteristic HEXXH active site for metalloproteinases, were similar to that of other fungi ( Figure 1B).
We constructed a FgFly1 knockout vector by double-joint PCR technique and knockedout by protoplast transformation. Afterwards, ∆FgFly1 was identified by PCR and Southern blot ( Figure S1).

Deletion of FgFly1 Impairs Asexual Reproduction and Spore Germination of F. graminearum
To investigate the role of FgFly1 in asexual growth, we inoculated WT PH-1, mutant ∆FgFly1 and ∆FgFly1-C on PDA, CM and MM agar media for culture, respectively. The growth after 3 d is shown in Figure 2A, and the specific data are shown in Table S2. the growth rate of ∆FgFly1 was slightly lower than that of WT PH-1; the hyphae of mutant ∆FgFly1 were more densely branched according to observation with a microscope of the hyphae grown on cellophane ( Figure 2B). In addition, the conidial production of each strain was measured after 5 d by inducing asexual conidia on CMC medium, as shown in Figure 2C. The molecular conidial production of mutant ∆FgFly1 was significantly lower than that of wild-type PH-1, and this phenomenon was recovered in the complementary strain ∆FgFly1-C. The spores were enriched and added to YEPD for germination, and 100 spores were randomly selected at 0, 2, 4, 6 and 8 h to calculate the germination rate. The spore germination rate of WT PH-1 was about 15% at 2 h, and the germination rate of mutant ∆FgFly1 was 3%; the spore germination rate of WT PH-1 was about 72%, and the germination rate of mutant ∆FgFly1 was 31% at 4 h. At 6 h, the spore germination rate of WT ph-1 was about 89%, and that of mutant ∆FgFly1 was 55%. The germination rate of WT PH-1 mutant ∆FgFly1 was 77% at 8 h.   The results in Figure 2D show that the spore germination of the mutant ∆FgFly1 afte 4 h, 6 h and 8 h was significantly lower and even slightly deformed compared with tha of WT PH-1. These results suggest that FgFly1 is essential for the production of conidia and the process of spore germination in F. graminearum.
When assayed for sexual reproduction on carrot agar plates, the ΔFgFly1 produced perithecia that same as those of wild-type strain PH-1 and ΔFgFly1-C ( Figure S2).

∆FgFly1 Mutants Have Enhanced Tolerance to Metal Cation Stress and Cell-Wall-Related Stress
To verify the effect of FgFly1 on F. graminearum tolerance, we examined the suscep tibility of ∆FgFly1 mutants to metal cations and cell-wall-damaging agents. WT PH-1, mu tant ∆FgFly1 and complementary strain ∆FgFly1-C were grown on PDA medium contain ing 0.5 mol/L CaCl2, 4 mmol/L CuSO4·5H2O, 0.2 g/L Congo Red, 0.02% SDS and 0.75 g/L caffeine for 3 d. As shown in ( Figure 3A,B), compared with WT PH-1 and ∆FgFly1-C ∆FgFly1 exhibited enhanced tolerance to metal cation stress and cell wall disruptors. The results in Figure 2D show that the spore germination of the mutant ∆FgFly1 after 4 h, 6 h and 8 h was significantly lower and even slightly deformed compared with that of WT PH-1. These results suggest that FgFly1 is essential for the production of conidia and the process of spore germination in F. graminearum.
When assayed for sexual reproduction on carrot agar plates, the ∆FgFly1 produced perithecia that same as those of wild-type strain PH-1 and ∆FgFly1-C ( Figure S2).

∆FgFly1 Mutants Have Enhanced Tolerance to Metal Cation Stress and Cell-Wall-Related Stress
To verify the effect of FgFly1 on F. graminearum tolerance, we examined the susceptibility of ∆FgFly1 mutants to metal cations and cell-wall-damaging agents. WT PH-1, mutant ∆FgFly1 and complementary strain ∆FgFly1-C were grown on PDA medium containing 0.5 mol/L CaCl 2 , 4 mmol/L CuSO 4 ·5H 2 O, 0.2 g/L Congo Red, 0.02% SDS and 0.75 g/L caffeine for 3 d. As shown in ( Figure 3A,B), compared with WT PH-1 and ∆FgFly1-C, ∆FgFly1 exhibited enhanced tolerance to metal cation stress and cell wall disruptors.

Deletion of FgFly1 Leads to Reduced Pathogenicity of F. graminearum
To investigate the biological function of FgFly1 during wheat infection, we f formed a spore inoculation assay on wheat ears in the flowering stage in the field 15th day of inoculation, the ∆FgFly1 mutant only exhibited symptoms of infection patches around the inoculation site, but the infection symptoms did not spread areas. The wild type and complementary strain ∆FgFly1-C showed severe spikele fection, which spread to other spikelets of the same plant ( Figure 4A). As shown i 4B, the diseased spikelets in the field were isolated and purified, and the grow graminearum was the same as that of the wild type, mutant ∆FgFly1 and ∆FgFly also carried out infection experiments on maize filaments. When cultured at 25 °C the WT PH-1 and the complementary strain ∆FgFly1-C had already infected th maize filaments, but the mutant ∆FgFly1 did not spread significantly ( Figure 4C dition, we also performed coleoptile infection experiments in which the ectoderm inoculated with PH-1 and ∆FgFly1-C turned significantly dark brown after 7 d p ulation. In contrast, the lesion size of the deletion mutant ∆FgFly1 was relative ( Figure 4D). The specific data from Figure 4A,C,D are presented in Table S2. Co with PH-1, the hyphal growth rate of ∆FgFly1 decreased by only 20%, wherea was active during infection, indicating that the reduced growth rate was not the k associated with the reduced virulence of the modified mutant.
Deoxynivalenol (DON) produced by Fusarium is considered a key virulence F. graminearum [46]. Plant infection experiments showed that deletion of FgFly1 the virulence of F. graminearum; therefore, we measured DON produced by t strains with a DON toxin ELISA kit. Compared with wild type, the ability of ∆FgFly1 to produce DON toxin did not change significantly ( Figure 4F). To furt firm this result, we determined the expression levels of key biosynthetic genes o celene by qRT-PCR, and the results are shown in Figure 4E. Compared with w the expression levels of the Tri gene family in ∆FgFly1 were not significantly cha conclusion, FgFly1 may affect the pathogenicity-related pathways of wheat dis

Deletion of FgFly1 Leads to Reduced Pathogenicity of F. graminearum
To investigate the biological function of FgFly1 during wheat infection, we first performed a spore inoculation assay on wheat ears in the flowering stage in the field. On the 15th day of inoculation, the ∆FgFly1 mutant only exhibited symptoms of infection in small patches around the inoculation site, but the infection symptoms did not spread to large areas. The wild type and complementary strain ∆FgFly1-C showed severe spikelets of infection, which spread to other spikelets of the same plant ( Figure 4A). As shown in Figure 4B, the diseased spikelets in the field were isolated and purified, and the growth of F. graminearum was the same as that of the wild type, mutant ∆FgFly1 and ∆FgFly1-C. We also carried out infection experiments on maize filaments. When cultured at 25 • C for 7 d, the WT PH-1 and the complementary strain ∆FgFly1-C had already infected the whole maize filaments, but the mutant ∆FgFly1 did not spread significantly ( Figure 4C). In addition, we also performed coleoptile infection experiments in which the ectodermal stems inoculated with PH-1 and ∆FgFly1-C turned significantly dark brown after 7 d postinoculation. In contrast, the lesion size of the deletion mutant ∆FgFly1 was relatively small. ( Figure 4D). The specific data from Figure 4A,C,D are presented in Table S2. Compared with PH-1, the hyphal growth rate of ∆FgFly1 decreased by only 20%, whereas FgFly1 was active during infection, indicating that the reduced growth rate was not the key factor associated with the reduced virulence of the modified mutant.
Deoxynivalenol (DON) produced by Fusarium is considered a key virulence factor of F. graminearum [46]. Plant infection experiments showed that deletion of FgFly1 reduced the virulence of F. graminearum; therefore, we measured DON produced by the three strains with a DON toxin ELISA kit. Compared with wild type, the ability of mutant ∆FgFly1 to produce DON toxin did not change significantly ( Figure 4F). To further confirm this result, we determined the expression levels of key biosynthetic genes of trichocelene by qRT-PCR, and the results are shown in Figure 4E. Compared with wild type, the expression levels of the Tri gene family in ∆FgFly1 were not significantly changed. In conclusion, FgFly1 may affect the pathogenicity-related pathways of wheat disease resistance.

The Signal Peptide of FgFly1 Has a Secretory Function, and Overexpression Causes Allergic Necrosis in N. benthamiana
In order to verify whether the FgFly1 signal peptide has a secretory function, we amplified the full-length signal peptide (amino acids 1-23), constructed pSUC2-FgFly1 sp and used the sucrose invertase secretion system of the YTK12 strain to verify the N-terminal signal peptide of FgFly1. It has a secretory function [47]. As shown in Figure 5A, the YTK12 strain was unable to synthesize tryptophan with sucrose invertase, so it is not able to grow on CMD-W medium lacking tryptophan and YPRAA, which contains only raffinose as a carbon source. Because the recombinant plasmid replaced the signal peptide of the yeast sucrose invertase with the target signal peptide FgFly1 SP and only the secreted sucrose invertase was able to use the raffinose on YPRAA, the transformed strain was able to grow on YPRAA and induced the TTC red reaction, indicating that the N-terminal signal peptide of FgFly1 has a secretory function. In order to study the function of FgFly1, we ligated its full-length coding sequence into a pBin vector and transiently expressed the FgFly1 gene in N. benthamiana leaves using A. tumefaciens infiltration technology. We found that transient expression of FgFly1 promoted N. benthamiana cell death, as well as the apoptosispromoting gene Bax ( Figure 5B).

Interaction between FgFly1 and TaCAMTA
We obtained the candidate target protein TaCAMTA of FgFly1 by screening the wheat yeast library. To further verify the interaction between FgFly1 and TaCAMTA, we cotransformed the plasmid pPR3-N-CAMTA/pBT3-SUC-Fly1 into an NMY51 yeast-competent cell for Yeast two-hybrid verification. (pNubG-Fe65/pTSU2-APP) was used as positive control, and pBT3-SUC-Fly1/pPR3-N was used as negative control. Single colonies growing on SD/-Trp/-Leu-deficient medium were picked and diluted at 10 6 , 10 5 and 10 4 gradients, respectively, and then spotted on SD/-Trp/-Leu/-His/-Ade/AbA plates. The tested yeasts were found to grow normally ( Figure 6A), indicating that the candidate target protein TaCAMTA and the effector protein FgFly1 could interact through the Yeast two-hybrid system. A. tumefaciens containing C-YFP-Fly1 and N-YFP-TaCAMTA vectors were constructed, mixed in equal proportions and coinjected into N. benthamiana leaves by A. tumefaciens infiltration technology. The 2 d infected leaves were observed by laser confocal microscopy for YFP fluorescence signal, and C-YFP/N-YFP-TaCAMTA was used as a negative control. The results are shown in Figure 6B; no fluorescence signal was detected in the negative control under laser confocal microscopy, and a YFP fluorescence signal was detected in C-YFP-Fly1/N-YFP-TaCAMTA, indicating that there was an interaction between FgFly1 and TaCAMTA. A. tumefaciens containing C-Luc-FgFly1 and N-Luc-TaCAMTA were constructed, mixed in equal proportions and coinjected into N. ben-

Interaction between FgFly1 and TaCAMTA
We obtained the candidate target protein TaCAMTA of FgFly1 by screening the wheat yeast library. To further verify the interaction between FgFly1 and TaCAMTA, we cotransformed the plasmid pPR3-N-CAMTA/pBT3-SUC-Fly1 into an NMY51 yeast-competent cell for Yeast two-hybrid verification. (pNubG-Fe65/pTSU2-APP) was used as positive control, and pBT3-SUC-Fly1/pPR3-N was used as negative control. Single colonies growing on SD/-Trp/-Leu-deficient medium were picked and diluted at 10 6 , 10 5 and 10 4 gradients, respectively, and then spotted on SD/-Trp/-Leu/-His/-Ade/AbA plates. The tested yeasts were found to grow normally ( Figure 6A), indicating that the candidate target protein TaCAMTA and the effector protein FgFly1 could interact through the Yeast two-hybrid system. A. tumefaciens containing C-YFP-Fly1 and N-YFP-TaCAMTA vectors were constructed, mixed in equal proportions and coinjected into N. benthamiana leaves by A. tumefaciens infiltration technology. The 2 d infected leaves were observed by laser confocal microscopy for YFP fluorescence signal, and C-YFP/N-YFP-TaCAMTA was used as a negative control. The results are shown in Figure 6B; no fluorescence signal was detected in the negative control under laser confocal microscopy, and a YFP fluorescence signal was detected in C-YFP-Fly1/N-YFP-TaCAMTA, indicating that there was an interaction between FgFly1 and TaCAMTA. A. tumefaciens containing C-Luc-FgFly1 and N-Luc-TaCAMTA were constructed, mixed in equal proportions and coinjected into N. benthamiana leaves by A. tumefaciens infiltration technology, and the infected leaves were taken for 36 h using a plant living molecular imaging system. The results are shown in Figure 6C. C-Luc-FgFly1/N-Luc-TaCAMTA has the same fluorescence as the positive control, but no fluorescence appeared in the negative control, indicating that there was an interaction between FgFly1 and TaCAMTA.

CAMTA Functional Verification
CAMTA plays an important regulatory role in plant biotic stress, abiotic stress, and plant growth and development as a CaM-binding protein in response to a series of environmental stresses, such as drought, salt and cold and hormonal signals, such as abscisic acid, ethylene and growth hormone [48]. Therefore, we used the A. thaliana ∆CAMTA mutant to verify the function of CAMTA in disease resistance. A. thaliana was infected with wild-type PH-1 spore fluid in the flowering stage, and the results are shown in Figure 7A. The disease resistance of CAMTA-deficient A. thaliana was enhanced relative to wild-type A. thaliana. The disease spot lengths of 10 A. thaliana plants were counted; the CAMTAdeficient mutant had significantly shorter spot lengths than the wild type, as shown in Figure 7B. CAMTA has been reported to negatively regulate NPR1-mediated immune re-

CAMTA Functional Verification
CAMTA plays an important regulatory role in plant biotic stress, abiotic stress, and plant growth and development as a CaM-binding protein in response to a series of environmental stresses, such as drought, salt and cold and hormonal signals, such as abscisic acid, ethylene and growth hormone [48]. Therefore, we used the A. thaliana ∆CAMTA mutant to verify the function of CAMTA in disease resistance. A. thaliana was infected with wild-type PH-1 spore fluid in the flowering stage, and the results are shown in Figure 7A. The disease resistance of CAMTA-deficient A. thaliana was enhanced relative to wild-type A. thaliana. The disease spot lengths of 10 A. thaliana plants were counted; the CAMTA-deficient mutant had significantly shorter spot lengths than the wild type, as shown in Figure 7B. CAMTA has been reported to negatively regulate NPR1-mediated immune response [49], so we collected A. thaliana at different infection stages of F. graminearum to determine the expression levels of NPR1-1, NPR1-3 and NPR1-4. Compared to WT, AtNPR1-1, AtNPR1-3 and AtNPR1-4 were found to be significantly up regulated in the ∆AtCAMTA mutant after 24 h. In conclusion, CAMTA is a susceptibility gene that plays a negative regulatory role in mediating the expression of the disease-resistance gene NPR1.

Discussion
In the process of plant-pathogen coevolution, a complex interaction relationship is formed. Plants establish many recognition and resistance mechanisms to organize and limit the infection of pathogens. Pathogens also form a variety of pathogenic mechanisms in order to avoid or overcome plant disease-resistance mechanisms, such as the formation of special infection structures, the secretion of a variety of hydrolytic enzymes, the production of host-selective toxins and the detoxification of plant-resistant substances [50][51][52].
Among fungal pathogens, metalloproteases (MEPs) are related to the autotoxicity of fungal pathogens [53]. Secreted metalloproteases are important pathogenic factors for many fungal diseases and cause damage to hosts to varying degrees. Sanz-Martin et al. (2016) found that the fungal hemolysin metalloproteinase gene Cgfl of Colletotrichum graminicola was also involved in fungal virulence to maize [28]. Avr-Pita, a metalloproteinase in Magnaporthe grisea, has been proven to be a specific pathogenic factor that can directly bind to plant resistance gene products, trigger signal cascade and thus produce resistance [29]. Although F. graminearum has been thoroughly studied, it is still unclear whether the

Discussion
In the process of plant-pathogen coevolution, a complex interaction relationship is formed. Plants establish many recognition and resistance mechanisms to organize and limit the infection of pathogens. Pathogens also form a variety of pathogenic mechanisms in order to avoid or overcome plant disease-resistance mechanisms, such as the formation of special infection structures, the secretion of a variety of hydrolytic enzymes, the production of host-selective toxins and the detoxification of plant-resistant substances [50][51][52].
Among fungal pathogens, metalloproteases (MEPs) are related to the autotoxicity of fungal pathogens [53]. Secreted metalloproteases are important pathogenic factors for many fungal diseases and cause damage to hosts to varying degrees. Sanz-Martin et al. (2016) found that the fungal hemolysin metalloproteinase gene Cgfl of Colletotrichum graminicola was also involved in fungal virulence to maize [28]. Avr-Pita, a metalloproteinase in Magnaporthe grisea, has been proven to be a specific pathogenic factor that can directly bind to plant resistance gene products, trigger signal cascade and thus produce resistance [29]. Although F. graminearum has been thoroughly studied, it is still unclear whether the M36 metalloprotease of F. graminearum plays key a role in host resistance. This study revealed the function of M36 metalloprotease in the development and infection of F. graminearum.
Based on homology BLAST search, a single-copy gene encoding M36 metallopeptidase was found in F. graminearum. FgFly1, similarly to other fungi, contains signal peptide, fungalysin propeptide motif, fungalysin metallopeptidase (M36) and HEXXH structure.
The reported M36 metalloprotease of maize powdery mildew affects the differentiation and virulence of conidia of maize powdery mildew and has the ability to cleave the maize chitinase [54]. In Fusarium verticillioides, FvFly1 was also found to have the ability to truncate chitinase [55]. In our study, the deletion of F. graminearum FgFly1 affected asexual reproduction, as well as virulence (Figures 2 and 4). The sensitivity of F. graminearum to cell-wall-related stress decreased after the deletion of FgFly1 (Figure 3). Therefore, we suspected that FgFly1 also interacted with wheat type-IV chitinase and carried out pointto-point verification, although the verification was not successful. We did not rule out the possibility of FgFly1 interacting with other types of chitinases, so we screened the wheat cDNA library, but no chitinase-related targets were found in the results. Interestingly, we found TaCAMTA in selected targets and validated the interaction ( Figure 6). CAMTA is a kind of important calmodulin-binding transcription factor. As a CaM-binding protein in multicellular eukaryotes, CAMTA responds to a series of environmental stress and hormone signals and plays an important regulatory role in plant stress, abiotic stress, and plant growth and development. Studies have shown that the CAMTA gene family is associated with plant innate immune signals [36]. In rice, oScbT (CAMTA member) plays a role in resistance to Xanthomonas oryzae pv. oryzae and Magnaporthe grisea [56]. Peiguo Yuan et al. (2020) found that CAMTA directly regulated the transcription of the NPR1 gene and played a key role in SA-mediated immune response [49]. Similarly to our results, the NPR1 gene was significantly upregulated in CAMTA-deficient Arabidopsis mutants, and the disease resistance of ∆AtCAMTA was significantly enhanced compared with wild type (Figure 7). Consistent with previous results, CAMTA is a susceptible gene that can negatively regulate the expression of NPR1.
In summary, we found that FgFly1 was involved in the asexual reproduction of F. graminearum, reducing plant disease resistance and enhancing its infection ability through interaction with CAMTA. This is the first instance of identification of the target CAMTA of F. graminearum M36 metalloprotease. In contrast to the functions of M36 metalloprotease in other fungi, these results may help to elucidate other filamentous fungal infection hosts.

Conclusions
The results presented here indicate that F. graminearum M36 metalloproteinase (FgFly1) is an effector with certain functions. As an effector within F. graminearum, ∆FgFly1 affected sexual reproduction in F. graminearum and influenced the sensitivity to Ca 2+ , Cu 2+ and cell-wall-related stresses. Interestingly, ∆FgFly1 had no significant effect on DON toxin production or the expression level of the Tri gene family, but pathogenicity was significantly reduced relative to PH-1 and ∆FgFly1-C, so we concluded that FgFly1 affected wheat, enhancing its pathogenicity. We then screened TaCAMTA for interactions with FgFly1 by Y2H and verified the interactions using Bi-FC and Luc. Inoculation of A. thaliana CAMTA deletion mutant with F. graminearum on pistil revealed that the disease resistance of A. thaliana was enhanced, and the expression level of A. thaliana NPR1 was significantly upregulated, proving that CAMTA is a disease-susceptibility gene. We tentatively concluded that FgFly1 interacts with the susceptibility gene CAMTA to promote its expression to enhance pathogenicity. In the present study, we explored the function of FgFly1 at the transcriptional level to further understand the pathogenesis of the effector FgFly1, which is important for the exploration of the pathogenesis of F. graminearum.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/jof8070726/s1, Table S1: PCR primers used in this study; Table S2. Growth rate, average disease severity and plant infection between different types of strains. Figure  S1: PutativeFly1 mutant was screened by PCR analysis, and putative Fly1 mutants were screened by Southern blot analysis; Figure S2: Analysis of ascospores discharge from PH-1, ∆FgFly1 and ∆FgFly1-C on carrot plate for 15 d.