The Mitochondrial Distribution and Morphology Family 33 Gene FgMDM33 Is Involved in Autophagy and Pathogenesis in Fusarium graminearum

The mitochondrial distribution and morphology family 33 gene (MDM33) regulates mitochondrial homeostasis by mediating the mitochondrial fission process in yeast. The wheat head blight Fusarium graminearum contains an FgMdm33 protein that is orthologous to Saccharomyces cerevisiae Mdm33, albeit its function remains unknown. We have reported here the roles of FgMdm33 in regulating fungal morphogenesis, mitochondrial morphology, autophagy, apoptosis, and fungal pathogenicity. The ΔFgmdm33 mutants generated through a homologous recombination strategy in this study exhibited defects in terms of mycelial growth, conidia production, and virulence. Hyphal cells lacking FgMDM33 displayed elongated mitochondria and a dispensable respiratory-deficient growth phenotype, indicating the possible involvement of FgMDM33 in mitochondrial fission. The ΔFgmdm33 mutants displayed a remarkable reduction in the proteolysis of GFP-FgAtg8, whereas the formation of autophagic bodies in the hyphal cells of mutants was recorded under the induction of mitophagy. In addition, the transcriptional expression of the apoptosis-inducing factor 1 gene (FgAIF1) was significantly upregulated in the ΔFgmdm33 mutants. Cumulatively, these results indicate that FgMDM33 is involved in mitochondrial fission, non-selective macroautophagy, and apoptosis and that it regulates fungal growth, conidiation, and pathogenicity of the head blight pathogen.


Introduction
The mitochondria play a central role in the cellular metabolism of eukaryotic cells by acting as the powerhouses to generate ATP through the tricarboxylic acid cycle and oxidative phosphorylation [1][2][3].To maintain their proper functional state, mitochondria undergo dynamic processes, involving fission, fusion, transport, and mitophagy [3,4].Fusion and fission are crucial for mitochondrial homeostasis, which regulates mitochondrial morphology, distribution, and function [5].The fusion process is beneficial for ATP generation and it maintains the mitochondrial DNA (mtDNA) levels and fidelity, whereas the fission of mitochondria influences the morphology and facilitates mitochondrial transport, mitophagy, apoptosis, and cell division [6][7][8].
In Saccharomyces cerevisiae, the mitochondrial outer membrane proteins Fzo1 and Ugo1 and inner membrane protein Mgm1 regulate mitochondrial fusion [9][10][11].The conserved dynamin GTPase Dnm1 is responsible for mitochondrial fission, which is then recruited to the mitochondrial outer membrane by the mitochondrial adaptor proteins Fis1, Mdv1, and Caf4 [12][13][14][15].Dimmer et al. reported a novel gene MDM33 that encodes a predicted membrane protein through the systematic screen, and they observed that the majority of mdm33 mutant cells harbored giant ring-like mitochondrial structures or 2-4 smaller interconnected mitochondrial rings [16,17].In addition, the structural defects of other organelles, such as the endoplasmic reticulum (ER), vacuoles, and cytoskeletons, were not determined in ∆mdm33 cells [17].As a consequence, Mdm33 is specifically required for the regulation of mitochondrial morphology.The phenotype of the ∆mdm33 mutant bears some similarities to ∆dnm1, ∆mdv1, and ∆fis1 [17][18][19][20].The overexpression of MDM33 results in a high aggregation of the mitochondria, the formation of the inner membrane septa, and the disappearance of the inner membrane cristae, which together severely affects the mitochondrial morphology [17].When combined with the location in the mitochondrial inner membrane, Mdm33 is supposedly involved in the fission of the mitochondrial inner membrane [17].
Yeast Mdm33 interacts with prohibitins, Phb1 and Phb2, which are the inner membrane complex that regulates the mitochondrial phospholipid homeostasis [21].As a result, Mdm33 modulates the phospholipids' homeostasis in the mitochondrial inner membrane by affecting the levels of phosphatidylethanolamine and cardiolipin [21].In addition, Mdm33 interacts with the subunits of the ER-mitochondria encounter structure (ERMES), which then mediates the formation of the ER-mitochondrial contact sites and promotes the exchange of phospholipids between two organelles [22,23].Mdm33 can regulate mitochondrial homeostasis via interaction with genes involved in the phospholipid metabolism and mitochondrial distribution and function.For cell growth, the overexpression of MDM33 causes growth arrest [24].However, the molecular biological function of MDM33 in regulating mitochondrial morphology or hyphal growth has so far not been reported in filamentous fungi.
Fusarium graminearum is a phytopathogen responsible for Fusarium head blight (FHB), a destructive fungal disease in wheat-growing areas worldwide, which leads to serious yield losses as well as decreases in the quality of cereals due to the contamination of mycotoxins [25][26][27].This disease caused by F. graminearum is difficult to control because of the lack of effective wheat cultivars with high resistance and the complexity of interaction between F. graminearum and wheat [25,28].As a hemibiotrophic fungus, F. graminearum developed infectious hyphae in subcuticular and intercellular plant tissues at the initial stages of infection, and colonized plant cells intracellularly after causing their death at later stages of infection [25,29].In the processes, numerous genes related to the pathogenesis have been revealed [30].Previous studies showed that some genes such as FgNdk1 and Fgporin affecting the mitochondrial morphology play crucial roles in pathogenicity [31,32].In this study, we identified FgMDM33 (FGSG_09659) encoding a putative mitochondrial distribution and morphology family 33 protein.Targeted gene deletion mutants of Fg-MDM33 exhibited a significant reduction in mycelial growth, conidia production, and virulence.The ∆Fgmdm33 mutants exhibited a high sensitivity to H 2 O 2 and induced the upregulation of the apoptosis-inducing factor (AIF).The deletion of this gene causes defects in the mitochondrial morphology and non-selective macroautophagy.Hence, in F. graminearum, FgMdm33 plays a crucial role in regulating fungal development, pathogenicity, and mitochondrial morphology mediated by fission, apoptosis, or autophagy.

Fungal Strains and Culture Conditions
F. granminearum wild-type strain PH-1 [33] served as a progenitor for generating the deletion mutants of FgMDM33.PH-1, the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants, and the complemented strain c∆Fgmdm33 were cultured on potato dextrose agar (PDA) plates at 25 • C with 12 h light/12 h dark cycles.

Gene Deletion and Complementation
A homologous recombination strategy was employed for FgMDM33 deletion, as described elsewhere [34].The plasmid pCAMBIA1300-neo was used for constructing the complementation vector.FgMDM33 along with its native promoter was cloned into a pCAMBIA1300-neo vector using the CloneExpress II One Step Cloning Kit (Nanjing Vazyme Biotech Co., Ltd., Nanjing, China).All PCR primers used in this study are listed in Table S1.The protocols for genetic transformation in F. graminearum are described elsewhere [35].Gene deletion mutants and complemented transformants were confirmed via PCR and a Southern blot analysis using the digoxigenin (DIG) high-prime DNA labeling and detection starter Kit I (Roche, Mannheim, Germany).

Phenotypic Analysis
For the colony growth assay, 5 mm mycelial plugs from the edge of the 3-day-old colony were, respectively, placed on PDA, complete medium (CM, 1 g/L yeast extract, 1 g/L casamino acid, 2 g/L peptone, 10 g/L D-glucose, 5% 20× nitrate salts, 0.1% trace elements, and 1% vitamin solution), and minimal medium (MM) plates and incubated at 25 • C for 3 days.For the conidiation assay, five 5 mm mycelial plugs from each strain were inoculated into 20 mL of a carboxylmethyl cellulose (CMC) liquid medium for 4 days at 25 • C in the light.For the investigation of conidial germination, freshly harvested spores of each strain were cultured in a 2% sucrose solution and examined under a light microscope (Olympus, Tokyo, Japan) after 24 h.These experiments, with 3 replicates, were repeated thrice.

Pathogenicity Assays
For Fusarium head blight symptom evaluation on wheat (the susceptible cultivar Jimai33), the middle spikelets of at least 10 flowering wheat heads were inoculated with mycelial plugs [33].Symptomatic spikelets were determined after incubation for 14 days.For the infection assays on the wheat leaf, 5 mm mycelial plugs were cut from the edge of a 3-day-old colony and inoculated on at least 6 leaves of 10-day seedlings of Jimai33.After inoculation, the leaves were kept at 25 • C with 100% humidity in a plant growth chamber.Photographs were taken at 3 days of inoculation.The infection experiments were conducted thrice independently.

Sensitivity of Mycelial Growth to Oxidative Stress
To examine the sensitivity of each strain to oxidative stress, the prepared 5 mm mycelial plugs were placed on CM and CM supplemented with 0.05% H 2 O 2 or a 200 ppm paraquat and cultured for 3 days at 25 • C. The colony diameters were measured by the crossintersection method.The experiments were conducted in triplicates and repeated thrice.

Mitochondrial Fission Assay
Mycelial growth was measured on CM plates supplemented with glucose or lactate as the sole carbon source, which were incubated for 3 days at 25 • C and then photographed.One piece of mycelial agar from each strain was cultivated in 20 mL of a liquid CM medium at 25 • C for 24 h, and the hyphae were collected and stained with MitoTracker TM Red (Invitrogen, Waltham, MA, USA), followed by observation under the Zeiss LSM 780 laser confocal microscope (Carl Zeiss, Jena, Germany).Moreover, the harvested hyphae were treated for transmission electron microscopy (TEM) assays and observed under the JEM-1230 electron microscope (JEOL, Tokyo, Japan), operating at 70 kV.

Western Blot Analysis
Total protein was extracted as described previously [33].Briefly, the protein samples from each extract were fractionated by SDS-PAGE and then immunoblotted with an anti-GFP antibody (Abmart, M20004, Shanghai, China) at the recommended dilutions.HRP-AffiniPure Goat Anti-Mouse IgG (H + L) (Fdbio Science, Shanghai, China) was used as the secondary antibody at a 1:10,000-1:50,000 dilution.FDbio-Femto ECL Kit (Fdbio Science, Shanghai, China) was used to detect the chemiluminescent signals.

Autophagy Assays
For the mitophagy assays, the hyphae of each strain were cultured in a liquid CM medium for 24 h and then shifted to the basal medium with glycerol (BM-G, 1.6 g/L yeast nitrogen base, 2 g/L asparagine, 1 g/L NH 4 NO 3 , and 1.5% glycerol) for 30 h, followed by 6 h of growth in the MM-N medium with 3 mM phenymethysulfonyl fluoride (PMSF).The harvested mycelium was observed under TEM, as described previously [36].Non-selective macroautophagy was analyzed as previously described [33].

qRT-PCR Analysis
The total RNA of each sample was isolated from the harvested hyphae by using the PureLink TM RNA Mini Kit (Invitrogen, Waltham, MA, USA), and cDNA was synthesized with reverse transcription by the HiScript ® II RT SuperMix for qPCR (+ gDNA wiper) (Nanjing Vazyme Biotech Co., Ltd., Nanjing, China).For qRT-PCR assays, ChamQ SYBR qPCR Master Mix (Nanjing Vazyme Biotech Co., Ltd., Nanjing, China) and the BIO-RAD CFX96 TM Real-Time System (Hercules, CA, USA) were used.The expression level of the target gene was calculated by the 2 −∆∆Ct method [37] using FgACTIN (FGSG_07335) as the internal reference.The experiment was conducted with at least 3 replicates and was repeated thrice independently.

Statistical Analyses
Statistical data were analyzed with SPSS Inc. (IBM) software (https://www.ibm.com/spss, accessed on 30 July 2009) and presented as the means ± standard deviations.Statistical significance was determined by ANOVA.

FgMDM33 Is Involved in Vegetative Growth and Conidiogenesis
The BLASTP analysis using S. cerevisiae Mdm33 as a query revealed that the F. graminearum genome contains an Mdm33 ortholog, which was named FgMdm33 (FGSG_09659).FgMdm33 encoding a 512 aa protein possesses the remarkable She9_MDM33 domain and a transmembrane region.To investigate the biological functions of this gene in F. graminearum, we generated targeted gene deletion mutants of FgMDM33 via a homologous recombination strategy.The resulting transformants with the hygromycin-resistance gene were preliminarily screened by the PCR analysis using the primer pairs listed in Table S1, which were then confirmed by Southern blotting hybridization (Figure S1).
Thus, we conducted a series of phenotypic analyses of the mutants by using conventional biological methods.∆Fgmdm33-3 and ∆Fgmdm33-5 mutants grew significantly slower relative to the wild-type strain PH-1 on PDA, CM, and MM plates at 25 • C for 3 days (Figure 1A,B).When compared to the thick aerial mycelia of PH-1, the aerial mycelia of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants were sparse and attached to the surface of the medium.The defects in the mycelial growth of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants were recovered by the reintroduction of FgMDM33 (Figure 1A,B).When culturing in the 1% MBL medium, the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants, respectively, produced (2.14 ± 0.23) × 10 5 conidia/mL and (2.26 ± 0.31) × 10 5 conidia/mL after 4 days, demonstrating a significant decrease when compared with (4.38 ± 0.18) × 10 5 conidia/mL and (4.32 ± 0.06) × 10 5 conidia/mL produced by PH-1 and the complemented strain c∆Fgmdm33, respectively (Figure 1C).We also observed the germination of conidia produced by PH-1, ∆Fgmdm33-3, ∆Fgmdm33-5, and c∆Fgmdm33 under light microscopy.The microscopic examination revealed no significant difference in the germination rate of the conidia among these strains, of which >85% of the conidia could germinate in a 2% sucrose solution (Figure 1D).These results suggested that FgMDM33 plays an important role in regulating the vegetative growth and conidia production, but not conidial germination.
spectively (Figure 1C).We also observed the germination of conidia produced by PH-1, ΔFgmdm33-3, ΔFgmdm33-5, and cΔFgmdm33 under light microscopy.The microscopic examination revealed no significant difference in the germination rate of the conidia among these strains, of which >85% of the conidia could germinate in a 2% sucrose solution (Figure 1D).These results suggested that FgMDM33 plays an important role in regulating the vegetative growth and conidia production, but not conidial germination.

FgMDM33 Is Required for Full Virulence in F. graminearum
To investigate the role of FgMDM33 in F. graminearum, we performed a virulence analysis by inoculating flowering wheat heads with mycelial plugs from each strain.The wild-type strain PH-1 and the complemented strain cΔFgmdm33 led to the development of typical scab symptoms in inoculated and nearby spikelets of wheat heads after 14 days.

FgMDM33 Is Required for Full Virulence in F. graminearum
To investigate the role of FgMDM33 in F. graminearum, we performed a virulence analysis by inoculating flowering wheat heads with mycelial plugs from each strain.The wild-type strain PH-1 and the complemented strain c∆Fgmdm33 led to the development of typical scab symptoms in inoculated and nearby spikelets of wheat heads after 14 days.Rachis of the inoculated wheat heads became blighted and the spikelets dried (Figure 2A).Moreover, the grains of the infected spikelets by PH-1 and c∆Fgmdm33 strains became shriveled and bleached (Figure 2A).In contrast, the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants triggered slight scab symptoms only in the point-inoculated spikelet, but not in any nearby spikelets.Spikelets above the inoculation point remained green, and the rachis turned brown only at the inoculated point.In addition, only the grain at the pointinoculated spikelet became shriveled, and the other grains from nearby spikelets remained plump (Figure 2A).To further testify the attenuated pathogenicity of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants, we inoculated the 10-day-old wheat leaves with mycelial plugs.After 3 days of inoculation, the whole leaves inoculated by PH-1 and c∆Fgmdm33 were observed to be covered with mycelia and had decayed, while the leaves inoculated with the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants were only partially covered with mycelia and had decayed only around the inoculation sites (Figure 2B).These results indicated that FgMDM33 is required for full virulence in F. graminearum.As the deletion of the gene led to the development of severe mycelial growth defects, the attenuated virulence of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants was likely to be the result of growth defects.
turned brown only at the inoculated point.In addition, only the grain at the point-inoculated spikelet became shriveled, and the other grains from nearby spikelets remained plump (Figure 2A).To further testify the attenuated pathogenicity of the ΔFgmdm33-3 and ΔFgmdm33-5 mutants, we inoculated the 10-day-old wheat leaves with mycelial plugs.After 3 days of inoculation, the whole leaves inoculated by PH-1 and cΔFgmdm33 were observed to be covered with mycelia and had decayed, while the leaves inoculated with the ΔFgmdm33-3 and ΔFgmdm33-5 mutants were only partially covered with mycelia and had decayed only around the inoculation sites (Figure 2B).These results indicated that FgMDM33 is required for full virulence in F. graminearum.As the deletion of the gene led to the development of severe mycelial growth defects, the attenuated virulence of the ΔFgmdm33-3 and ΔFgmdm33-5 mutants was likely to be the result of growth defects.The mycelial plugs of each strain were inoculated onto the detached wheat leaves.The photographs were taken after 3 days of inoculation.

FgMDM33 May Be Involved in Mitochondrial Fission
In S. cerevisiae, Mdm33 controls mitochondrial morphology by regulating the fission of the mitochondrial inner membrane.The phenotype of the ∆mdm33 mutant displays no mtDNA loss and no acquisition of a respiratory-deficient growth, which bears some similarities to the deletion mutants of DNM1, MDV1, and FIS1 genes, which are known as the critical regulators affecting mitochondrial fission [17].To examine whether FgMDM33 is involved in mitochondrial fission, we determined the response of the mutants to a non-fermentative carbon source.In consistence with the wild-type strain PH-1 and the complemented strain c∆Fgmdm33, the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants grew well on CM plates using lactate as the sole carbon source (Figure 3A), which implies that the  To further determine the role of FgMDM33 in mitochondrial fission, we observed the mitochondrial morphology in hyphal cells of each strain by staining mitochondria with the MitoTracker TM Red dye.Under laser scanning confocal microscopy, the mitochondria in the hyphal cells of PH-1 and c∆Fgmdm33 were observed to be punctuated and short rod-shaped, while the mitochondria of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants presented an elongated shape (Figure 3B).Moreover, we examined the morphology of the mitochondria in different strains by TEM after culturing in the CM medium for 24 h.The elongated mitochondria were observed in the ∆Fgmdm33-5 mutant (Figure 3C).The elongated mitochondrial morphology of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants reminds us that FgMDM33 assists in maintaining the mitochondrial morphology by regulating the mitochondrial fission in F. graminearum.

Loss of FgMDM33 Caused Partial Defects in Autophagy
Mitochondria with abnormal morphology can be degraded by mitophagy [38], and the deletion of FgMDM33 can result in defects of the mitochondrial morphology.Thus, we speculated that FgMDM33 may be involved in mitophagy.We previously reported that FgAtg20 is involved in multiple autophagic processes but not required for mitophagy [33].Thus, the ∆Fgatg20 mutant was used as the negative control in the mitophagy analysis.Each strain was cultured in the glycerol medium for 30 h after culturing in the CM medium for 24 h, and then shifted to the MM-N medium for another 12 h in the presence of 2 mM phenylmethanesulfonylfluoride fluoride (PMSF).Autophagic bodies in vacuoles were observed in the wild-type strain PH-1, the ∆Fgmdm33-5 mutant, and the complemented strain c∆Fgmdm33 by TEM, as in the ∆Fgatg20 mutant (Figure 4A).This result suggested that FgMDM33 was not involved in mitophagy in F. graminearum.Then, we attempted to determine whether FgMDM33 is involved in the non-selective macroautophagy.The previously constructed GFP-FgAtg8 vector was, respectively, transformed into PH-1 and the ∆Fgmdm33-5 mutant [33].In the nutrient-rich CM medium, both GFP-FgAtg8 and a free GFP band could be detected in PH-1 and the ∆Fgmdm33-5 mutant (Figure 4B).When being induced in the MM-N medium for 16 h, the level of full-length GFP-FgAtg8 decreased and an increasingly stronger free GFP band was detected in both strains.Compared to the nearly invisible GFP-FgAtg8 band in PH-1, the GFP-FgAtg8 band in the ∆Fgmdm33-5 mutant remained strong under the condition of nitrogen starvation for 16 h (Figure 4B).The GFP-FgAtg8 proteolysis assay confirmed that the non-selective macroautophagy was partially impaired in the ∆Fgmdm33-5 mutant.Cumulatively, FgMDM33 is important for non-selective macroautophagy, but not for mitophagy in F. graminearum.

FgMDM33 Is Involved in the Apoptosis of F. graminearum
We analyzed the responses of each strain to several stresses and found that the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants displayed a high sensitivity to oxidative stress.When compared to the wild-type strain PH-1 and the complemented strain c∆Fgmdm33 with the mycelial growth inhibition of 59.28 ± 6.18% and 66.41 ± 0.34%, respectively, ∆Fgmdm33-3 and ∆Fgmdm33-5 presented 96.93 ± 2.39% and 99.29 ± 0.36% mycelial growth inhibition, respectively, and were more sensitive to 0.05% H 2 O 2 (Figure 5A,B).Paraquat is also regarded as the agent causing oxidative stress [39]; hence, the sensitivity of each strain to paraquat was also tested.However, the sensitivity to 200 ppm paraquat displayed no significant difference among the tested strains.qRT-PCR was further conducted to investigate the expression levels of peroxidase genes involved in the regulation of oxidative stresses.Several catalase genes, FCA1, FCA2, and FCA4, in response to oxidative stress were found to be significantly upregulated in the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants when compared to PH-1 and c∆Fgmdm33, although the expression level of SOD1, which is responsible for the detoxification of paraquat, displayed no significant difference among all strains (Figure 5C).This result suggested that the increased sensitivity of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants to H 2 O 2 was not attributable to the deficiency of the expression of catalase genes.It has been reported that H 2 O 2 can induce apoptosis [40]; therefore, we speculated that the enhanced sensitivity of the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants to H 2 O 2 may be attributable to the activation of apoptotic cell death after the deletion of FgMDM33.AIF is the main mediator of caspase-independent apoptosis-like cell death, and its expression is upregulated during the process of apoptosis [41].The qRT-PCR analysis in the present study showed that the expression of FgAIF1 (FGSG_02433), the AIF in F. graminearum (Figure S2), was significantly upregulated in the ∆Fgmdm33-3 and ∆Fgmdm33-5 mutants relative to that with PH-1 and c∆Fgmdm33 (Figure 5D).These results indicated that FgMDM33 may negatively regulate the apoptosis in F. graminearum.Each strain was cultured in liquid CM at 25 °C for 24 h, after which it was shifted to BM-G for 30 h and starved in MM-N for another 24 h.Scale bar = 0.5 µm.(B) GFP-FgAtg8 proteolysis assays.Immunoblot assays with the total lysates from CM-or MM-N-cultured strains with an anti-GFP antibody or anti-GAPDH antibody.

FgMDM33 Is Involved in the Apoptosis of F. graminearum
We analyzed the responses of each strain to several stresses and found that the ΔFgmdm33-3 and ΔFgmdm33-5 mutants displayed a high sensitivity to oxidative stress.When compared to the wild-type strain PH-1 and the complemented strain cΔFgmdm33 with the mycelial growth inhibition of 59.28 ± 6.18% and 66.41 ± 0.34%, respectively, deletion of FgMDM33.AIF is the main mediator of caspase-independent apoptosis-like cell death, and its expression is upregulated during the process of apoptosis [41].The qRT-PCR analysis in the present study showed that the expression of FgAIF1 (FGSG_02433), the AIF in F. graminearum (Figure S2), was significantly upregulated in the ΔFgmdm33-3 and ΔFgmdm33-5 mutants relative to that with PH-1 and cΔFgmdm33 (Figure 5D).These results indicated that FgMDM33 may negatively regulate the apoptosis in F. graminearum.

Discussion
We have described that FgMdm33 was required for the maintenance of proper mitochondrial morphology in F. graminearum.This role of FgMdm33 was determined through the analysis of the mitochondrial structure in ∆Fgmdm33-mutant cells, which displayed mitochondrial aggregation and extension as elongated tubules (Figure 3).In S. cerevisiae, the ∆mdm33 mutant bearing defects in the mitochondrial fission did not acquire a respiratorydeficient growth phenotype [17].In F. graminearum, the ∆Fgmdm33 mutants could grow on CM plates with lactate as the sole carbon source (Figure 3A), displaying a similar phenotype to the ∆mdm33 mutant of S. cerevisiae, suggesting that the ∆Fgmdm33 mutants may bear the mitochondrial fission defect.Moreover, the elongated mitochondria in ∆Fgmdm33 mutants provided further evidence to confirm the role of FgMDM33 in mitochondrial fission.
Mitochondria with an abnormal morphology can be degraded by mitophagy [42]; therefore, we speculated that FgMDM33 regulates mitochondrial morphology by mi-

Figure 1 .
Figure 1.ΔFgmdm33 mutants are defective in mycelial growth and conidiation.(A), Colony morphology of the wild-type strain PH-1, the ΔFgmdm33 mutants, and the complemented strain cΔFgmdm33 grown on PDA, CM, and MM plates at 25 °C for 3 days.(B), The colony growth of each strain in A. Error bars represent the standard deviation.Different small letters indicate a significant difference (p < 0.05).(C), A bar chart depicting the statistical analysis of conidiation.Conidia produced by each strain were quantified with a hemacytometer after incubation in the CMC liquid medium for 4 days in the light.Different small letters indicate a significant difference (p < 0.05).(D), The percentage of conidial germination.The conidia of each strain were cultured in a 2% sucrose solution and examined under a light microscope after 24 h.The same small letters indicate no significant difference (p < 0.05).

Figure 1 .
Figure 1.∆Fgmdm33 mutants are defective in mycelial growth and conidiation.(A), Colony morphology of the wild-type strain PH-1, the ∆Fgmdm33 mutants, and the complemented strain c∆Fgmdm33 grown on PDA, CM, and MM plates at 25 • C for 3 days.(B), The colony growth of each strain in A. Error bars represent the standard deviation.Different small letters indicate a significant difference (p < 0.05).(C), A bar chart depicting the statistical analysis of conidiation.Conidia produced by each strain were quantified with a hemacytometer after incubation in the CMC liquid medium for 4 days in the light.Different small letters indicate a significant difference (p < 0.05).(D), The percentage of conidial germination.The conidia of each strain were cultured in a 2% sucrose solution and examined under a light microscope after 24 h.The same small letters indicate no significant difference (p < 0.05).

Figure 2 .Figure 2 .
Figure 2. FgMDM33 is required for full virulence.(A), The dissection of wheat heads infected by the wild-type strain PH-1, the ΔFgmdm33 mutants, and the complemented strain cΔFgmdm33.Inoculated wheat heads were examined and dissected after 14 days of inoculation.(B), Wheat cut-leaf Figure 2. FgMDM33 is required for full virulence.(A), The dissection of wheat heads infected by the wild-type strain PH-1, the ∆Fgmdm33 mutants, and the complemented strain c∆Fgmdm33.Inoculated wheat heads were examined and dissected after 14 days of inoculation.(B), Wheat cut-leaf assays.The mycelial plugs of each strain were inoculated onto the detached wheat leaves.The photographs were taken after 3 days of inoculation.

Figure 3 .
Figure 3. FgMDM33 may be involved in mitochondrial fission.(A), Colony growth of the wildstrain PH-1, the ΔFgmdm33 mutants, and the complemented strain cΔFgmdm33 on CM plates taining glucose or lactate.(B), The mitochondria in the hyphal cells of the ΔFgmdm33 mutant elongated.Each strain was grown in a CM liquid medium for 24 h at 25 °C, after which the hy were harvested and stained with MitoTracker TM Red for microscopic observation.Scale bar = 5 (C), Ultrastructural morphology of the mitochondria in the hyphal cells of each strain was visua by transmission electron microscopy.M, mitochondria.Scale bar = 0.5 µm.

Figure 3 .
Figure 3. FgMDM33 may be involved in mitochondrial fission.(A), Colony growth of the wildtype strain PH-1, the ∆Fgmdm33 mutants, and the complemented strain c∆Fgmdm33 on CM plates containing glucose or lactate.(B), The mitochondria in the hyphal cells of the ∆Fgmdm33 mutant were elongated.Each strain was grown in a CM liquid medium for 24 h at 25 • C, after which the hyphae were harvested and stained with MitoTracker TM Red for microscopic observation.Scale bar = 5 µm.(C), Ultrastructural morphology of the mitochondria in the hyphal cells of each strain was visualized by transmission electron microscopy.M, mitochondria.Scale bar = 0.5 µm.

Figure 4 .
Figure 4.The deletion of FgMDM33 caused defects in non-selective macroautophagy.(A) FgMDM33is dispensable for mitophagy in F. graminearum.The observation of autophagic bodies in the vacuoles of wild-type strain PH-1, the ΔFgmdm33 mutants, and the complemented strain cΔFgmdm33.Each strain was cultured in liquid CM at 25 °C for 24 h, after which it was shifted to BM-G for 30 h and starved in MM-N for another 24 h.Scale bar = 0.5 µm.(B) GFP-FgAtg8 proteolysis assays.Immunoblot assays with the total lysates from CM-or MM-N-cultured strains with an anti-GFP antibody or anti-GAPDH antibody.

Figure 4 .
Figure 4.The deletion of FgMDM33 caused defects in non-selective macroautophagy.(A) FgMDM33is dispensable for mitophagy in F. graminearum.The observation of autophagic bodies in the vacuoles of wild-type strain PH-1, the ∆Fgmdm33 mutants, and the complemented strain c∆Fgmdm33.Each strain was cultured in liquid CM at 25 • C for 24 h, after which it was shifted to BM-G for 30 h and starved in MM-N for another 24 h.Scale bar = 0.5 µm.(B) GFP-FgAtg8 proteolysis assays.Immunoblot assays with the total lysates from CM-or MM-N-cultured strains with an anti-GFP antibody or anti-GAPDH antibody.

Figure 5 .
Figure 5. FgMDM33 may be involved in apoptosis.(A) Colonies of the wild-type strain PH-1, the ΔFgmdm33 mutants, and the complemented strain cΔFgmdm33 on CM with 0.05% H2O2 or 200 ppm paraquat for 3 days at 25 °C.(B) The statistical analysis of the percentage of mycelial growth inhibition.Line bars in each column represent the standard deviation.Different small letters indicate a

Figure 5 .
Figure 5. FgMDM33 may be involved in apoptosis.(A) Colonies of the wild-type strain PH-1, the ∆Fgmdm33 mutants, and the complemented strain c∆Fgmdm33 on CM with 0.05% H 2 O 2 or 200 ppm paraquat for 3 days at 25 • C. (B) The statistical analysis of the percentage of mycelial growth inhibition.Line bars in each column represent the standard deviation.Different small letters indicate a significant difference (p < 0.05).(C) The transcriptional levels of FgFCA1, FgFCA2, FgFCA4, and FgSOD1 by qRT-PCR.Line bars in each column represent the standard deviation.Different small letters indicate a significant difference (p < 0.05).(D) The expression of FgAIF1 was upregulated in the ∆Fgmdm33 mutants.Line bars in each column represent the standard deviation.Different small letters indicate significant differences (p < 0.05).