Mitochondrial dynamics caused by QoIs and SDHIs fungicides depended on FgDnm1 in Fusarium graminearum

KANG Jin-bo,ZHANG Jie,LIU Yin-kai,SONG Ji-chang,OU Jian-lin,TAO Xian,ZHOU Ming-guo,DUAN Ya-bing

College of Plant Protection,Nanjing Agricultural University,Nanjing 210095,P.R.China

Abstract Fusarium head blight (FHB) caused by Fusarium graminearum is a devastating fungal disease on small grain cereal crops,because it reduces yield and quality and causes the mycotoxin contamination to the grain.Dynamins and dynamin-related proteins (DRPs) are large GTPase superfamily members,which are typically involved in the budding and division of vesicles in eukaryotic cells,but their roles in Fusarium spp.remain unexplored.Here,we found that FgDnm1,a DRP and homolog to Dnm1 in Saccharomyces cerevisiae,contributes to the normal fungal growth,sexual reproduction and sensitivity to fungicides.In addition,we found FgDnm1 co-localizes with mitochondria and is involved in toxisome formation and deoxynivalenol (DON) production.Several quinone outside inhibitors (QoIs) and succinate dehydrogenase inhibitors (SDHIs) cause fragmentated morphology of mitochondria.Importantly,the deletion of FgDnm1 displays filamentous mitochondria and blocks the mitochondrial fragmentation induced by QoIs and SDHIs.Taken together,our studies uncover the effect of mitochondrial dynamics in fungal normal growth and how such events link to fungicides sensitivity and toxisome formation.Thus,we concluded that altered mitochondrial morphology induced by QoIs and SDHIs depends on FgDnm1.

Keywords: Fusarium graminearum,FgDnm1,mitochondrial dynamics,fungicides

1.Introduction

Fusarium graminearumis a destructive fungal pathogen which causes Fusarium head blight (FHB).This disease results in severe yield losses and quality reduction,posing a great threat to human health owning to its ability of mycotoxin production (Champeilet al.2004;Qiuet al.2014).So far,the primary approach to control FHB is the use of chemical fungicides.In recent years,respiratory chain inhibitor fungicides have developed rapidly and possess a good application prospect in the control of FHB due to their novel targets and wide spectrum (Xuet al.2019;Duanet al.2020).

Mitochondria are universally semi-autonomous subcellular organelles involved in energy supplies,cellular physiology,cellular differentiation,cellular signaling,apoptosis and autophagy.Mitochondria are highly dynamic organelles undergoing persistent fusion and fission under different life stages or external factors (Pernaset al.2016).According to the theory of mitochondrial aging,the accumulation of reactive oxygen species (ROS)produced in the mitochondrial respiratory chain can cause mutations and damages of mitochondrial DNA (mtDNA).The gradual accumulation of damaged mtDNA may lead to the decreased activity of respiratory chain complex,which leads to cell apoptosis (Balabanet al.2005).However,damaged mtDNA and depolarized mitochondrial membrane could be compartmented through fission,and cleared through autophagy or other ways,so as to maintain the normal function of mitochondria (Tanakaet al.2010).This suggests a tight link between mitochondrial fission and respiratory chain function,and genes involved in mitochondrial fission may have a specific mechanism of response to respiratory chain inhibitor fungicides.

Dynamins are the proteins typically composed of five domains,including GTPase domain,middle domain,pleckstrin homology domain (PHD),GTPase effector domain (GED),and proline rich domain (PRD).However,dynamin-related proteins (DRPs) lack one or more of these domains or contain additional domains (Zhonget al.2016).DRPs and dynamins belong to large GTPase superfamily proteins,which are indispensable for the budding of transport vesicles and scission of organelles in eukaryotic cells (Praefcke and McMahon 2004).The biological functions of DRPs and dynamins have been studied in many organisms.InSaccharomyces cerevisiae,Dnm1 and Mgm1 mediate mitochondrial fission and fusion(Martinez-Guzmanet al.2020).Vps1,a conserved fungal DRP,plays important roles in vacuolar fission (Raymondet al.1992).InMagnaporthe oryzae,MoDnm1,MoFis and MoMdv1 regulate mitochondrial and peroxisomal fission,pexophagy and mitophagy,and are also required for appressorium function (Zhonget al.2016).Mitochondrial fusion is also a conserved cellular process.InS.cerevisiae,the GTPase Fzo1,Mgm1,and Ugo1,links the outer and inner membrane fusion machineries and plays critical roles in mitochondrial fusion (Rapaportet al.1998;Sesaki and Jensen 2001;Martinez-Guzmanet al.2020).These studies suggest that dynamins and DRPs are critical for mitochondrial dynamics and morphology.

Although some dynamins and DRPs were explored in many organisms (Ramachandranet al.2018),their roles were not fully understood inFusariumspp.In this study,FgDnm1(FGSG_01320),the mitochondrial fission gene,was screened and identified inF.graminearum.FgDnm1knockout/restore mutant was constructedviahomologous recombination strategy,and its function was characterized by multi-phenotypic analyses.The susceptibility of the mutant to several respiratory chain inhibitor fungicides was determined.We also observed the mitochondrial dynamics and morphology underFgDnm1deletion and several fungicides treatment.Consequently,our study initially uncovered the underlying correlation between mitochondrial morphology and toxisome formation inF.graminearum.Our results showed that FgDnm1 plays critical roles in mycelial growth,sexual reproduction,fungicides sensitivity and toxisome formation inF.graminearum.Furthermore,the significance of altered mitochondrial morphology caused by fungicides depends on FgDnm1 inF.graminearummay also explain the pharmacological mechanisms causing mitochondrial fragmentation.

2.Materials and methods

2.1.Fungal strains and culture conditions

TheF.graminearumstrain PH-1 was used as the progenitor for the transformation assays in this study.PH-1 and the resultant mutants were grown on potato dextrose agar (PDA) for hyphal growth examination.Sexual reproduction was assayed on carrot agar medium in accordance with a previous study (Qinet al.2020).Toxisome observation and trichothecene production were assayed by culturing each strain in liquid trichothecene biosynthesis inducing medium (TBI) at 28°C in a shaker (175 r min-1) in the dark.The experiments were conducted three times.For the observation of mitochondrial dynamics and morphology,all the strains or mutants were grown in yeast extracts peptone dextrose medium (YEPD) for 3 d at 25°C.

2.2.FgDnm1 phylogenetic tree

Phylogenetic tree of FgDnm1 homologues from several other species were constructed by the neighbour-joining tree method using the MEGA7 Programs.Species names and GenBank accession numbers are as follows:XP_011317106 (F.graminearumFgDnm1);NP_013100(Saccharomyces cerevisiaeDnm1p);XP_018231645(Fusarium oxysporumFoDnm1);XP_018744325.1(Fusarium verticillioidesFvDnm1);XP_003717217(M.oryzaeMoDnm1);XP_024546963 (Botrytis cinereaBcDnm1);XP_001597414 (Sclerotinia sclerotiorumSsDnm1);and XP_964068 (Neurospora crassaNcDnm1).

2.3.Construction of FgDnm1 deletion and complemented mutants

TheF.graminearumprotoplast preparation and fungal transformation were performed according to the previous reports (Liuet al.2013;Yanget al.2018).Briefly,FgDnm1gene was replaced with two marker genes.The bacterial hygromycin B (hyg B) phosphotransferase gene(hph) was used as a positive selectable marker,and the herpes simplex virus thymidine kinase gene (hsv-tk) was used as a conditional negative selection marker.The flanking sequences were amplified and fused with two marker genes,then transformed into the wild-type strain PH-1 (Appendix A).PutativeFgDnm1deletion mutants were further confirmed by PCR,quantitative PCR and Southern blotting assays (Appendices B and C).For the complemented assays,theFgDnm1gene with the flanking sequences was amplified and transformed into theFgDnm1deletion mutant to replace the two marker genes,and the transformants were verified by sequencing,PCR and Southern blotting assays.

2.4.Observation of subcellular localization,mitochondrial dynamics and fusarium toxisomes

To investigate the subcellular localization of FgDnm1,the gene fused with a GFP tag in the C-terminus was transformed into PH-1.FgDnm1-GFP mutant was stained with MitoTracker Red CMXRos (Yeasen,Shanghai,China),then observed with Leica TCS SP8 Confocal Microscope (Wetzlar,Hessen,Germany).

For the observation of mitochondrial dynamics,the mitochondria was tagged with GFP (ATP2-GFP) (Appendix D),and the resulting GFP tagged mutants were cultured in YEPD medium at 25°C for 48 h.Fungicides were added after culturing for 24 h in YEPD.The fluorescent intensity of mitochondria was observed with Leica TCS SP8 Confocal Microscope (Wetzlar,Hessen,Germany).

For the observation of toxisome formation patterns,the sequence ofFgTri1tagged with GFP was introduced into the corresponding strain using the strategy previously reported (Zhouet al.2020a).All the resulting strains were observed under Leica TCS SP8 Confocal Microscope(Wetzlar).

2.5.Fungicide sensitivity to quinone outside inhibitors (QoIs) and succinate dehydrogenase inhibitors (SDHIs)

The sensitivity of the wild-type strain and the derived mutants to fluopyram and boscalid was tested on yeast bacteo peptone agar (YBA) medium amended with 2.5,5,10,20,40 μg mL-1of fluopyram and 31.25,62.5,125,250,500 μg mL-1of boscalid.Pyraclostrobin and coumoxystrobin sensitivity assays were tested on alkyl ester agar (AEA) medium amended with 0.0025,0.005,0.01,0.02,and 0.04 μg mL-1of pyraclostrobin and 0.0008,0.004,0.02,and 0.1 μg mL-1of coumoxystrobin.After culturing for 3 d at 25°C in the dark,colony diameters were measured in two perpendicular directions,and the EC50values were calculated as previously described(Zhenget al.2018).The experiments were repeated twice with three replicates for each concentration.

2.6.Mycelial growth and sexual reproduction

To test the mycelial growth of each strain,mycelial plugs(5 mm in diameter) taken from the edge of a 3-d-old colony of each strain were inoculated on PDA.After incubation for 4 d at 25°C in the dark,the diameter of each strain was measured,and the experiments were conducted three times.

Sexual reproduction was examined on carrot medium according to a previous report (Bianet al.2020).Briefly,mycelial plug taken from a 3-d-old colony was cultured on carrort medium for 7 d,the aerial hyphae was then scraped with sterile 0.1% Tween-20 (Wanget al.2011;Zhenget al.2016).

2.7.Conidiation and virulence assays

For conidiation tests,one mycelial plug of each strain was inoculated in a 50-mL flask containing 30 mL mung bean broth (MBB) medium before incubation for 3 d at 25°C in a shaker (175 r min-1).The number of conidia in MBB was determined using a haemacytometer.The experiments were repeated twice.

For virulence assays,we conducted infection assays using wheat coleoptiles as previously described (Wanget al.2011;Zhenget al.2016).Briefly,wheat seed(Huaimai 33) was planted in plates containing wet filter paper.The plates were placed in a plastic case containing filter paper to retain moisture.After incubation for 3 d,the top 2 to 3 mm of the coleoptiles were cut out and 2 μL of fresh conidia (1×106) suspensions were pipetted into the wound of seedlings.After incubation at 25°C for 7 d,the lengths of disease lesions were measured.At least 60 wheat seedlings were used for each treatment.The experiments were performed three times.

2.8.Gene expression level analysis

To determine the expression level ofFgDnm1andFgFzo1,we extracted total RNA individually from PH-1 and ΔFgDnm1 mutants according to a previous report(Duanet al.2020).The two genes were then quantified by qPCR andactingene was used as a reference.Relative expression levels of each gene were calculated using the 2-ΔΔCtmethod (Livak and Schmittgen 2001).

2.9.Analysis of DON production

To quantify deoxynivalenol (DON) production,each strain was cultured in trichothecene biosynthesis inducing medium (TBI) at 28°C for 7 d.The mycelia and liquid were then collected.The liquid was assayed to quantify DON using enzyme linked immunosorbent assay (ELISA)and the mycelia were dried and weighted.

3.Results

3.1.FgDnm1 is a dynamin-related protein

Dynamins are highly conserved,so we searched available genomes ofF.graminearumby BLAST using theS.cerevisiaedynamin Dnm1 as the reference and identified FGSG_01320 encoding the Dnm1 homolog FgDnm1.A large GTPase domain,a middle coiled coil domain and a GED domain of FgDnm1 were identified using SMART online by domain analysis,which shares high sequence conservation withS.cerevisiaeDnm1p(Fig.1-A).Phylogenetic analysis was performed using Mega7 Software with the neighbor-joining method.The results showed that FgDnm1 was homologous to those from other fungi.The deduced amino acid sequence of FgDnm1 was 93% identical to that ofFusarium oxysporumandFusarium verticillioides(GenBank accession XP_018231645 and XP_018744325.1) (Fig.1-B).The percentages of identity of the amino acid sequences were high enough to consider the conservative evolutionary trend of Dnm1 inFusariumspp.

Fig.1 Bioinformatics analysis of dynamin-related proteins in Fusarium graminearum.A,protein domain analysis of FgDmn1.DYNc,large GTPase domain;GED,GTPase effector domain.B,phylogenetic tree analysis of FgDmn1 protein and its homologous proteins.The identities between the amino acid sequence of FgDnm1 in F.granminearum and that in other fungi are as follows:Fusarium oxysporum,93% identity;Fusarium verticillioides,93% identity;Neurospora crassa,84% identity;Botrytis cinerea,93%identity;Sclerotinia sclerotiorum,93% identity;Magnaporthe oryzae,91% identity;Saccharomyces cerevisiae,91% identity.

3.2.FgDnm1 and FgFzo1 are mainly localized to mitochondria

Dnm1 and Fzo1 are the main components of mitochondrial fission and fusion machine,which has been confirmed to mediate mitochondrial dynamics (Rapaportet al.1998;Tieu and Nunnari 2000).To further explore the functions of FgDnm1 and FgFzo1,we observed their subcellular localization.A GFP tag was fused to the C-terminus of each gene,and expressed in PH-1.FgDnm1 presented punctate pattern and FgFzo1 showed diffused pattern(Fig.2).To test if the two proteins located in mitochondria,we stained the mitochondria with a far red-fluorescent dye MitoTracker Red CMXRos.Green and red fluorescence signal was detected and overlapped in both mutants(Fig.2),indicating that FgDnm1 and FgFzo1 are mainly localized to mitochondria.

Fig.2 FgDnm1 and FgFzo1 co-localize with mitochondria.A,subcellular localization of FgDnm1-GFP in conidia (right panel)and hyphae (left panel) of PH-1 cultured in YEPD medium for 48 h.Bar=5 μm.B,the localization of FgFzo1-GFP in conidia (right panel) and hyphae (left panel) of PH-1 cultured in YEPD medium for 48 h.BF,bright field.Bar=5 μm.

3.3.Deletion of FgDnm1 promotes FgFzo1 expression level

InS.cerevisiae,Fzo1p has been confirmed to mediate the fusion of mitochondrial inner membrane and that Dnm1-Fzo1 double deletion mutant restored to normal mitochondrial morphology (Sesaki and Jensen 1999).The filamentous network of mitochondria caused byFgDnm1deletion may due to enhanced mitochondrial fusion.To determine whether FgDnm1 could affect mitochondrial fusion,we obtained Fzo1 homolog (FgFzo1) by BLAST using theS.cerevisiaeFzo1 as the reference.We found that expression level ofFgFzo1in ΔFgDnm1 upregulated markedly as compared with wide type strain PH-1 and complemented strain (Fig.3).These results suggest thatFgDnm1is involved in a negative regulation toFgFzo1inF.graminearum.

Fig.3 The deletion of FgDnm1 promotes FgFzo1 expression level.Total RNA extracts from the wild type,ΔFgDnm1 and complementary strain were analyzed by qPCR.Data are mean±SD (n=3).Means of values followed by different letters are significantly different according to Fisher’s least significant difference (P＜0.05).

3.4.FgDnm1 is required for mycelial growth and sexual reproduction

To gain an insight of the functions of dynamins inF.graminearum,we try to generate deletion mutant ofFgDnm1andFgFzo1.Maybe for the reason of impaired mtDNA mentioned in previous reports (Sesakiet al.2001;Guillouet al.2005),theFgFzo1deletion mutants could not be obtained despite more than 2 000 transformants were screened.It suggests that FgFzo1 is essential inF.graminearum.However,ΔFgDnm1 mutant showed significantly attenuated hyphal growth in either PDA medium or minimal medium (MM)(Fig.4-A;Appendix E),and displayed obvious reduction in perithecium production in comparison to the wildtype strain (Fig.4-B).These results indicated that FgDnm1 is involved in vegetative growth and sexual reproduction.

Fig.4 Effects of FgDnm1 on mycelial growth and sexual reproduction.A,the wild type,deletion mutants and complementary strain were cultured on PDA and MM for 72 h,ΔFgDnm1 mutants showed a reduced mycelial growth rate and aerial hyphae amount.B,reduced perithecia production in ΔFgDnm1 compared with wild type and complementary strain on carrot medium.

To test whether FgDnm1 is also required for conidiation and virulence,we measured conidiation by MBB medium and conducted infection assays on wheat coleoptiles.Surprisingly,no significant difference was seen between the wild-type and ΔFgDnm1 mutant (Appendix F),suggesting that FgDnm1 is dispensable for conidiation and virulence inF.graminearum.

3.5.FgDnm1 contributes to fungicides sensitivity

To explore the sensitivity of ΔFgDnm1 mutants to respiratory inhibitors,we first assessed the sensitivity of ΔFgDnm1 to boscalid and fluopyram.The results showed that ΔFgDnm1 exhibits significantly increased sensitivity to boscalid and fluopyram (Table 1;Fig.5).Furthermore,we also tested the sensitivity of ΔFgDnm1 to pyraclostrobin and coumoxystrobin,and a significantly decreased sensitivity was observed (Table 1;Fig.5).These results suggest that FgDnm1 contributes to fungicide sensitivity.

Fig.5 FgDnm1 contributes to fungicides sensitivity.A,colonies were photographed after 72 h at 25°C on YBA medium for fluopyram(left panel) and boscalid (right panel).B,colonies were photographed after 72 h at 25°C on AEA medium for pyraclostrobin (left panel) and coumoxystrobin (right panel).

Table 1 Sensitivity of strains to QoIs and SDHIs fungicides

3.6.Fluopyram,pyraclostrobin,boscalid and coumoxystrobin induces mitochondrial fragmentation

Since QoIs and SDHIs are both respiratory inhibitors,which target to mitochondrial complex III and complex II(Arakawaet al.2018;Desbordeset al.2020;Duanet al.2020),we assume that they may have an impact on mitochondrial dynamics and morphology.To address this possibility,we constructed a strain bearing mitochondrial green fluorescence (ATP2-GFP).Then,the resultant strain was treated with various concentrations of fungicides.As shown in Fig.6,mitochondria morphology changed from aggregated to fragmented with increasing concentrations of fungicides.For some unknown reasons,fluopyram and boscalid have significant impacts on mitochondrial dynamics with the concentration of only 2×EC50and 1×EC50,but the effect of pyraclostrobin and coumoxystrobin is limited in comparison.Taken together,these results strongly indicated that QoIs and SDHIs have an impact on mitochondrial dynamics and morphology.

Fig.6 Pyraclostrobin,fluopyram,coumoxystrobin and boscalid induces mitochondrial fragmentation,respectively.PH-1 with a mitochondrial marker was treated with different fold EC50 of fungicides.The solvent DMSO was used as the control.BF,bright field.Data showed the similar results obtained in three times independent experiments performed for each treatment.Bar=5 μm.

3.7.FgDnm1 deletion mutants are defective in mitochondrial fission

InS.cerevisiae,Dnm1 is involved in mitochondrial fission,and deletion ofDnm1causes the defect of fission(Bleazardet al.1999;Kennedyet al.2020).To better reveal the roles of FgDnm1 in mitochondrial fission,PH-1 and ΔFgDnm1 harboring the tagged ATP2-GFP were examined.The mitochondria become filamentous network in ΔFgDnm1 compared with that fragmented in wild-type PH-1 (Fig.7).These results suggest that FgDnm1 plays an important role in the mitochondrial fission process inF.graminearum.

Fig.7 Effects of FgDnm1 on mitochondrial morphology.PH-1 and ΔFgDnm1 strains expressing FgAtp2-GFP were cultured in YEPD medium for 48 h and observed under a Laser Scanning Confocal Microscope.BF,bright field.Bar=5 μm.

3.8.Altered mitochondrial morphology caused by fungicides depends on FgDnm1

Since several fungicides have been proved to induce mitochondrial fragmentation,and Dnm1 and its homologs contribute to mitochondrial fission machinery,we considerthat mitochondrial fragmentation caused by fungicides may be mediated by FgDnm1.To verify this possibility,we treated ΔFgDnm1 and PH-1 tagged with ATP2-GFP individually with the corresponding fungicides of specific concentrations sufficient to induce mitochondrial fragmentation.Namely,1×EC50of boscalid,2×EC50of fluopyram,10×EC50of pyraclostrobin and 10×EC50of coumoxystrobin.As expected,the mitochondria remain filamentous in ΔFgDnm1 instead of fragmented in the wild type (Fig.8).These results indicate that FgDnm1 is a key factor mediating mitochondrial fission and the fragmentation of mitochondria induced by fungicides may depend on FgDnm1.

Fig.8 Altered mitochondrial morphology caused by fungicides relies on FgDnm1.A,effects of DMSO on mitochondrial morphology of PH-1 and ΔFgDnm1 mutants.B,effects of pyraclostrobin on mitochondrial morphology of PH-1 and ΔFgDnm1 mutants.PH-1 and ΔFgDnm1 mutants expressing FgAtp2-GFP were cultured in YEPD medium for 24 h at 25°C and then treated with 10×EC50 of pyraclostrobin for 12 h at 25°C followed by the observation under a Laser Scanning Confocal Microscope.C,effects of boscalid on mitochondrial morph of PH-1 and ΔFgDnm1 mutants.PH-1 and ΔFgDnm1 mutants expressing FgAtp2-GFP were cultured in YEPD for 24 h at 25°C and then treated with 10×EC50 of boscalid for 12 h at 25°C followed by the observation under a Laser Scanning Confocal Microscope.D,effects of fluopyram on mitochondrial morphology of PH-1 and ΔFgDnm1 mutants.PH-1 and ΔFgDnm1 mutants expressing FgAtp2-GFP were cultured in YEPD for 24 h at 25°C and then treated with 2×EC50 of fluopyram for 12 h at 25°C followed by the observation under a Laser Scanning Confocal Microscope.E,effects of coumoxystrobin on mitochondrial morphology of PH-1 and ΔFgDnm1 mutants.PH-1 and ΔFgDnm1 mutants expressing FgAtp2-GFP were cultured in YEPD for 24 h at 25°C and then treated with 100×EC50 of boscalid for 12 h at 25°C followed by the observation under a Laser Scanning Confocal Microscope.BF,bright field.Bar=5 μm.

3.9.FgDnm1 is involved in toxisome formation and DON production

Given that the toxisomes are critical for mycotoxin biosynthesis inF.graminearum,and this structure stems from endoplasmic reticulum remodeling (Tanget al.2018;Zhouet al.2021).In addition,the mitochondria and endoplasmic reticulum form tight functional contacts that regulate several crucial cellular processes (Gomez-Suagaet al.2017a,b;Veereshet al.2019;Zhouet al.2020b).Therefore,the mitochondrial dynamin proteins may play a role in toxisome formation and mycotoxin production.To test this hypothesis,we expressed Tri1-GFP in the wild type and ΔFgDnm1 mutants.Microscopic observations showed the Tri1-GFP fluorescent signals reduced markedly in the mycelia of ΔFgDnm1 in comparison with those in wild type (Fig.9-A).Correspondingly,DON in the mycelia of ΔFgDnm1 was below the level of the wild type (Fig.9-B).Taken together,these results indicate that FgDnm1 is involved in toxisome formation and DON production and the correlation between mitochondrial dynamics and toxisome formation deserve further interest.

Fig.9 FgDnm1 is related to toxisome formation and deoxynivalenol (DON) production.A,effects of FgDnm1 deletion on toxisome formation.Hyphae of PH-1 and ΔFgDnm1 expressing Tri1-GFP were cultured in TBI before fluorescent signal examination.BF,bright field.Bar=5 μm.B,effects of FgDnm1 deletion on DON production.DON was extracted from each strain cultured in TBI for 7 d at 28°C,and then assayed by enzyme linked immunosorbent assay (ELISA).Data are mean±SD (n=3).The different letters on the bars indicates significant difference according to Fisher’s least significant difference test (P＜0.01).

4.Discussion

Dynamin-related protein Dnm1p mediating mitochondrial fission in complex with Mdv1 and Fis1 was firstly identified by researchers inS.cerevisiae,with highly conserved structure and function (Bleazardet al.1999;Griffinet al.2005).In many cases,Dnm1p exists in the form of polymers in cytosol and transfer to the fission site of mitochondrial outer membraneviathe assistant of chaperonin Mdv1 and Fis1 (Mozdyet al.2000;Tieu and Nunnari 2000;Bossyet al.2003;Yoonet al.2003).Mdv1 was recruited from cytoplasm and bound with Dnm1p by Fis1 for activating Dnm1p to form Dnm1-GTP oligomer and enrich in mitochondrial potential cleavage site during outer membrane division event.Furthermore,Dnm1-GTP oligomers will continue to form ring structure,altering the distance or angles between molecules through the hydrolysis of GTP,gradually compressing and cutting offouter membrane to make it split.Dnm1p will return to cytosol after the fission event (Zhang and Chan 2007;Lackneret al.2009).InM.oryzae,MoDnm1 function as a regulatory molecule through indirect recruiting MoFis1viaan adapter MoMdv1.MoDnm1 deletion mutant exhibited attenuated vegetative growth,asexual reproduction and virulence inM.oryzaeand the absence of Fis1 and Dnm1 accounted for the defects of mitosis and ascospore differentiation inPodospora anserine(Zhonget al.2016;Navarro-Espíndolaet al.2020).While the mitochondrial fission events mediated by Dnm1 have been systematically confirmed inS.cerevisiae,M.oryzaeandP.anserina,but not inF.graminearum.In this study,we have shown that FgDnm1 contributes to several biological phenotypes and revealed the correlation between mitochondrial morphology and normal respiratory chain function.Previous studies laid more emphasis on some typical biological phenotypes related to the functions of DRPs and to fungal growth and development (Praefckeet al.2004;Zhonget al.2016),our studies emphasize not only biology property but also respiratory chain function,fungicides sensitivity and toxisome formation linked to DRPs.Therefore,this study may establish a novel correlation between mitochondrial morphology and fungal fungicides sensitivity and toxisome formation.Additionally,our studies try to provide new strategies for the development of species-specific compound against pathogenic fungi.

It has been demonstrated that Dnm1p or its homologs are localized on mitochondrial inS.cerevisiaeandM.oryzae(Mearset al.2011;Zhonget al.2016).In current study,we found that the majority of FgDnm1 are localized on mitochondria while a few of them could be also observed in cytosol.These results indicate that most of FgDnm1 may be recruited to mitochondria fission sites along with the cycles of mitochondrial dynamics and a few remain in cytosol.

Fzo1,the core components of mitochondrial fusion machinery,has been proved to regulate mitochondrial dynamics with numerous associated proteins (Kouet al.2019;Gao and Hu 2021).ΔFgDnm1 mutants displayed filamentous network of mitochondria compared with PH-1.Therefore,we speculated that the absence ofFgDnm1may enhanceFgFzo1expression level,resulting in more filamentous network of mitochondria.To verify our hypothesis,we assayed the expression level ofFgDnm1andFgFzo1in PH-1 and ΔFgDnm1 mutants by quantitative PCR.As expected,the expression level of FgFzo1 markedly increased in ΔFgDnm1 mutants compared with that in wild type.These results suggest that the expression level of FgFzo1 is enhanced byFgDnm1deletion,which is in agreement with our hypothesis.Our studies further demonstrate the tight interaction of dynamic change of mitochondrial morphology on transcriptional level.

Since mitochondrial functions are closely related to its morphology,we generated deletion and complemental mutants by homologous replacement,and screened by PCR and confirmed by Southern blot analysis.Previous studies demonstrates thatMoDnm1deletion mutant exhibited attenuated vegetative growth,asexual reproduction and virulence inM.oryzaeand the absence of Fis1 and Dnm1 accounted for the defects of mitosis and ascospore differentiation inP.anserine(Zhonget al.2016;Navarroet al.2020).Therefore,we identified the biological phenotype of ΔFgDnm1 and found that ΔFgDnm1 exhibited remarkably reduction in mycelial growth,aerial hyphae amount and perithecium production,consistent with the findings inM.oryzaeandP.anserine.However,the conidiation and virulence ability were not affected byFgDnm1deletion.Such discrepancy may due to species differences.Taken together,these results show that the vegetative growth and sexual reproduction are regulated by FgDnm1.

Mitochondrial respiratory chain complex II and complex III have been regarded as the most promising targets of agricultural fungicide in recent years (Casida and Durkin 2017;Heet al.2018;Huanget al.2019;Fisheret al.2020).SDHIs and QoIs,which target to complexes II and complexes III,are commonly applied in agricultural practice (Xuet al.2019;Duanet al.2020).However,desired efficacy of these fungicides was still absent in the management of FHB.We tested the sensitivity of wild type and ΔFgDnm1 to several QoIs and SDHIs.Compared with wild type and complemented strain,ΔFgDnm1 presented reduced sensitivity to QoIs and increased sensitivity to SDHIs.Our findings for the first time raise the possibility that fungicides sensitivity is associated with mitochondrial morphology.However,the antagonistically altered sensitivity to QoIs and SDHIs in ΔFgDnm1 mutant may attributes to different targets of these fungicides.

In eukaryotic cells,mitochondrial fission event is mediated by a complex composed of several proteins,including Dnm1,Fis1,Mdv1,and Caf4 (Nagotuet al.2008;Guoet al.2012;Zhonget al.2016).Furthermore,QoIs and SDHIs are members of respiratory inhibition fungicides,which may affect the normal function of mitochondria (Gisiet al.2002;Kampet al.2021).Previous report has shown that mitochondrial morphology is crucial for its function (Valenteet al.2017;Spieret al.2019).In the current study,FgAtp2,the F1F0 ATP synthase subunit beta (FGSG_04312),has been identified as an effectively unequivocal fluorescent biomarker for mitochondria inF.graminearum(Appendix D).Importantly,we found that QoIs and SDHIs fungicides induced mitochondrial fragmentation in wild type,but not in ΔFgDnm1 mutant.These results suggest that altered mitochondrial morphology caused by fungicides may depends on FgDnm1.However,the level of mitochondrial morphology change is diverse under different fungicides treatment,this may result from different EC50and active ingredients,further studies are required to clarify this discrepancy.In addition,we also conclude that FgDnm1 might be one of the targets of QoIs and SDHIs fungicides.In summary,our current data support a model as shown in Fig.10.In the wild type ofFusarium graminearum,FgDnm1 mediating mitochondrial fission was recruited to the fission site of mitochondria under QoIs and SDHIs treatment.In the ΔFgDnm1 mutant,although under the same fungicides treatment,the mitochondrial fission is completely blocked.Thus,FgDnm1 plays essential roles for fungicides-induced mitochondrial fragmentation.

Fig.10 A proposed model illustrating the role of FgDnm1 in fungicides-induced mitochondrial fragmentation.

DON is the main mycotoxin inF.graminearum.During infection,DON is mainly produced by a novel cellular structure named “fusarium toxisomes”,which is remodeled from endoplasmic reticulum (Tanget al.2018;Zhouet al.2020a,2021).Previous studies have reported that ER and mitochondria could form functional contactsvia“tethering proteins” (Hirabayashiet al.2017).In our experiments,compared with wild type,ΔFgDnm1 was defective in toxisome formation and DON production.However,there was no significant virulence reduction in ΔFgDnm1 although DON contributes to virulence,implying another parallel pathways contribute to virulence may be enhanced by FgDnm1 deletion.Nevertheless,these results suggest that FgDnm1 is involved in DON biosynthesis,implying a potential link betweenEndoplasmic reticulumand FgDnm1.

5.Conclusion

Taken together,our study demonstrated for the first time that mitochondrial dynamics affected by QoIs and SDHIs fungicides depends on FgDnm1 inF.graminearum.Importantly,FgDnm1 may be a novel target for QoIs and SDHIs fungicides.In addition,our data initially show that the mitochondrial dynamics also affect DON production and toxisome formation,but the potential mechanism behind them awaits further investigation.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31772190),the Jiangsu Agriculture Science and Technology Innovation Fund,China (JASTIF)(CX(21)2037),and the Postgraduate Research &Practice Innovation Program of Jiangsu Province,China(KYCX21_0631).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on http://www.ChinaAgriSci.com/V2/En/appendix.htm