GONG Xiao-dong,LlU Yu-wei,Bl Huan-huan,YANG Xiao-rong,HAN Jian-min,DONG Jin-gao, ,GU Shou-qin

1 State Key Laboratory of North China Crop Improvement and Regulation/Key Laboratory of Hebei Province for Plant Physiology and Molecular Pathology,College of Life Sciences,Hebei Agricultural University,Baoding 071001,P.R.China

2 College of Plant Protection,Hebei Agricultural University,Baoding 071001,P.R.China

3 Botou Vocational College,Cangzhou 062150,P.R.China

Abstract Homologous recombination (HR) and nonhomologous end joining (NHEJ) are considered the two main double-strand break(DSB) repair approaches in eukaryotes. Inhibiting the activities of the key component in NHEJ commonly enhances the efficiency of targeted gene knockouts or affects growth and development in higher eukaryotes. However,little is known about the roles of the NHEJ pathway in foliar pathogens. Here we identified a gene designated StKU80,which encodes a putative DNA end-binding protein homologous to yeast Ku80,in the foliar pathogen Exserohilum turcicum. Conserved domain analysis showed that the typical domains VWA,Ku78 and Ku-PK-bind are usually present in Ku70/80 proteins in eukaryotes and are also present in StKu80. Phylogenetic analysis indicated that StKu80 is most closely related to Ku80(XP_001802136.1) from Parastagonospora nodorum,followed by Ku80 (AGF90044.1) from Monascus ruber. Furthermore,the gene knockout mutants ΔStKU80-1 and ΔStKU80-2 were obtained. These mutants displayed longer septas,thinner cell walls,smaller amounts of substances on cell wall surfaces,and more mitochondria per cell than the wild-type (WT) strain but similar HT-toxin activity. The mutants did not produce conidia and mature appressoria. On the other hand,the mutants were highly sensitive to H2O2,but not to ultraviolet radiation. In summary,the StKU80 plays devious roles in regulating the development of E.turcicum.

Keywords:Exserohilum turcicum,pathogenicity,gene knockout,growth and development

1.lntroduction

DNA,serving as the most important hereditary material,is generally damaged during cell growth and development or in organisms under diverse stresses (Jackson and Bartek 2009). Previous reports showed that approximately 10-100 chromosomal DNA double-strand breaks (DSBs) occur in a cell every 24 h and that DSBs commonly occur during normal cell physiological processes or are caused by endogenous or exogenous DNA-damaging agents such as enzymes,free radicals,ionizing radiation and ultraviolet radiation (Daleyet al.2005). Thus,lack of responses to this damage or errors in their repair will lead to gene mutations.

To avoid DSB damage,organisms have developed a variety of DSB repair pathways (Aylon and Kupiec 2004;Daleyet al.2005; Haber 2006; Shrivastavet al.2007; Kitoet al.2008; Thompson 2012). Among them,homologous recombination (HR) and nonhomologous end joining (NHEJ)pathways are the most common. HR is an accurate DNA repair pathway that uses an identical chromosome as a template and occurs only in S phase and G2 phase,while NHEJ is the fundamental DSB repair mode,which the cell uses to form a specific base-pairing (the shortest pairing is 1-5 bp) between two complementary bases,not requiring any templates (van Gentet al.2001; Luet al.2016). On the other hand,NHEJ induces mismatching between prominent single strands at the broken ends to achieve direct connections between two broken double strands and occurs during the whole cell cycle. Moreover,some reports showed that HR and NHEJ are independent and competitive and that the proportions in which they are used differ between organisms. For example,HR is a main DNA repair pathway in bacteriaand yeast,while NHEJ is dominant in human,plant,insect and fungi (Pastinket al.2001; Gonget al.2005;Shuman and Glickman 2007; Maruyama and Kitamoto 2008;Villalbaet al.2008; Kück and Hoff 2010). The activity of NHEJ is higher than that of HR in most of the filamentous fungi (Pastinket al.2001; Kück and Hoff 2010). Therefore,low gene editing efficiency commonly occurs in fungi because NHEJ is the dominant repair pathway (Maruyama and Kitamoto 2008; Villalbaet al.2008).

Like most DNA repair processes,the NHEJ repair pathway requires nucleases to resect damaged DNA,polymerases to fill in new DNA fragments,and ligases to restore the integrity of DNA strands (Lieber 2010). The multi-enzyme complex Ku70/80,a heterodimer,serves as a node at which nucleases,polymerases and ligases can dock (Lieber 2008). Ku70/80 is extremely sensitive to DNA damage,rapidly recognizes gaps or discontinuous fragments in DSBs and is considered a sensor of DSBs in the NHEJ pathway.Furthermore,due to their strong affinity and lack of sequence specificity,Ku proteins are involved in not only DSB repair but also many crucial biological process,including the regulation of DNA replication,gene expression,nucleic acid metabolism,telomere structure maintenance,cell proliferation,and anti-apoptosis (Fisher and Zakian 2005).When DSBs occur,the Ku80/Ku70 complex binds broken DNA ends to avoid their digestion by DNase,recruits and activates a DNA-dependent protein kinase catalytic subunit(DNA-PKcs) to form the corresponding DNA-dependent protein kinase (DNA-PK) and completes DNA repair with the assistance of XRCC4 and DNA ligase IV (Walkeret al.2001; Hefferin and Tomkinson 2005). Moreover,the C-terminus of Ku80 recognizes variations in DNA structure and activates different DNA-PKs to different degrees,and the recognition of DNA structure changes is carried out by the interaction between DNA-PKcs and DNA (Woodset al.2015). Therefore,Ku proteins are crucial to the initiation and regulation of the NHEJ pathway,and a deficiency in Ku80 can decrease the efficiency of DNA repair in the NHEJ pathway.

Most reports focused on the functional research of Ku systems examining human and animal systems,and the Ku proteins were related to oncogenesis and high incidences of lymphadenoma and thymic tumors(Guet al.1997; Vogelet al.1999). Some other reports showed thatKUgenes are highly expressed in human,including in the tissues of nasopharyngeal carcinoma,ovarian serous tumor,mammary cancer,pancreatic cancer and laryngocarcinoma (Guet al.1997; Vogelet al.1999).KU80splay various roles in different fungi and mainly affect their growth and development,cell proliferation,and their resistance to ultraviolet rays,mutagens and DNA-damaging agents. For example,inNeurospora crassa,KU80knockout mutants show higher sensitivity to methyl methanesulfonate,ethyl methanesulfonate,and bleomycin than the wild-type (WT) strain but are insensitive to UV light,4-nitroquinoline 1-oxide,camptothecin,and hydroxyurea (Ninomiyaet al.2004). InUstilago maydis,the Ku heterodimer functions in the initial step of the NHEJ pathway. In addition,Ku proteins play important roles in cellular processes,including telomere maintenance and cell proliferation (De Sena-Tomáset al.2015). Some studies indicate that the inactivation ofKUgenes contributes to improving the efficiency of gene editing. For example,inRhodosporidium toruloides,the deletion ofKU70increases the gene replacement frequency to 90% when the length of the flanking sequence is ≥1 kb (Kohet al.2014). InMonascus ruber,the deletion of eitherKU70orKU80elevates the gene editing efficiency from 15 to 61.5% (Heet al.2013). InN.crassa,the deletion of eithermus-51ormus-52,genes homologous to humanKU70/KU80,increases the gene replacement frequency from 10-30 to 100% (Ninomiyaet al.2004). A similar phenomenon is observed in other fungi such asMagnaporthe grisea,Colletotrichum higginsianum,Botrytis cinerea,Parastagonospora nodorum,andClaviceps purpurea(Choqueret al.2008; Haarmannet al.2008;Villalbaet al.2008; Ushimaruet al.2010; Fenget al.2012).In summary,Ku proteins act as not only regulators controlling growth and development but also crucial reparative factors involved in DNA damage repair in organisms. Therefore,KU80orKU70knockout mutants are usually employed as efficient recipients for gene targeting when the genes have no effect on the growth and development of the organisms.

The whole genome sequenceinExserohilum turcicumwas determined in April 2011 (Ohmet al.2012),and the data were announced on the JGI website (Joint Genome Institute,http://genome.jgi-psf.org) (Grigorievet al.2012).TheE.turcicumgenome sequence is a great platform for gene function research and will be the foundation for exploring new effective measures to control the northern corn leaf blight. In the present research,a gene designatedStKU80,homologous toSaccharomyces cerevisiaeKU,was found by searching theE.turcicumgenome database and cloned. Furthermore,the gene functions ofStKU80were elucidated by comparing the differences in the phenotypes and pathogenicity ofStKU80knockout mutants and the WT strain. This research will not only clarify the effects ofStKU80on growth,development,and pathogenicity but also help to identify whether theStKU80knockout mutant can be used as high-efficiency receptor strain in gene editing inE.turcicum.

2.Materials and methods

2.1.Strains and growth conditions

TheE.turcicumstrains used in the present study were 01-23 (WT) and twoStKU80knockout mutants (KOs),ΔStKU80-1andΔStKU80-2. The methods used for strain culturing were slightly modified from previously reported methods (Guet al.2014). The strains,whose genomic DNA was extracted,were cultured in a potato dextrose(PD) (20% potato (w/v) and 2% dextrose (w/v)) medium at 100 r min-1and 25°C. Protoplasts were prepared using a conidium suspension harvested with sterile water. The suspension was transferred to a PD medium,and the culture was incubated for 20 h on a rotary shaker at 150 r min-1and 25°C. Conidium counting was performed by growing the strains on potato dextrose agar (PDA) (20% potato (w/v),2% dextrose (w/v),and 1.30% agar (w/v)) medium in the dark for 14 d. KO strains were screened on PDA medium supplemented with 0.2‰ glufosinate ammonium at 25°C in the dark. Conidium counting and appressorium induction were performed by culturing the strains on PDA medium in darkness for 14 d and then harvesting with 10 mL of sterile water in each Petri dish. All experiments were repeated at least three times.

2.2.Cloning and characterization of StKU80

The DNA sequences ofStKU80were obtained by searching theE.turcicumonline database (http://genome.jgi.doe.gov/Settu1/Settu1.home.html) using BLASTN,and the sequence ofKU80fromS.cerevisiaewas used as the query.PrimersKU80-F (5´-CGGGCAAATGACAACAGACG-3´)andKU80-R (5´-CACAACTGAATGTCCTGCTGC-3´) were designed using the software Primer 5.0 to clone the DNA ofStKU80. Phylogenetic and sequence analyses were performed by the Neighbor-Joining (NJ) method using the software ClustalX 2.1 and MEGA 5.0 (Tamuraet al.2011).

2.3.Transformation-mediated gene disruption

The gene knockout strategy used in the present study was based on the method of Ruiz-Roldánet al.(2001). An upstream flanking fragment ofStKU80(StKU80-A) was amplified with a forward primer(5´-CCCTCGAGCCACCCAGCAGCAGATTGTA-3´)and a reverse primer (5´-CGGAATTCCGATGGCGG AAACAGGA-3´). A downstream flanking fragment (StKU80-B)was amplified with a forward primer (5´-TCCCCCGGGTAA GCAGCAGGACATTCAGT-3´) and a reverse primer(5´-GCTCTAGAATCCAGGTAAGAGGTTGTTG-3´). The glufosinate ammonium resistance gene (BAR) was amplified with a forward primer (BAR-L) (5´-CGGAATTCCTCG ACAGAAGATGATATTG-3´) and a reverse primer (BAR-R)(5´-TCCCCCGGGTTCTCGGTACTATGCATTAC-3´). After purification with the TaKaRa DNA Fragment Purification Kit ver.2.0 (TaKaRa,Japan),StKU80-A,StKU80-B andBARwere cloned into the vector pBS. The resulting plasmid,named pBS-A-BAR-B,containedStKU80-A,StKU80-B andBARand was used to transform WT protoplasts ofE.turcicum. The transformed protoplasts were incubated in glufosinate ammonium medium (1.2% (w/v) agar in water containing 0.2‰ (w/v) glufosinate ammonium) at 25°C for 6-7 d until glufosinate ammonium-resistant transformants were observed. All transformants were individually preserved and subjected to verification by PCR (forward primer(KU80upL):5´-ACCCGCTTCCTAGCCTACGG-3´; reverse primer (KU80upR):5´-CGAACGCACTAAATCACCAAAT-3´),RT-PCR (forward primerKU80-TL:5´-TTCGCAATCTTCAA CAGCA-3´; reverse primerKU80-TR:5´-ATCAGGGTC GTCAAAATCG-3´). Primers (forward primer:5´-GGTATC AGTATTCAGGTTGTC-3´ and reverse primer:5´-GTTAAGA CTACGACGGTATC-3´) were used for amplifying an 18S gene fragment as a reference gene. The probes (forward primer:5´-GACAAGATGGTTCATTTAGGC-3´; reverse primer:5´-TTCAGCAGGTGGGTGTAGAG-3´) were employed for amplifying theBARfragment and the method used for Southern blotting was according to the introduction of DIG High Prime DNA Labeling and Detection Starter Kit I(Roche,Swiss).

2.4.Estimation of growth ratios and conidial yield,observation of hyphal morphology and cell wall development

The colony growth ratio and conidial yield were monitored using the method described in our previous reports (Zhanget al.2012; Guet al.2014). Strains were cultured onto PDA medium and cultured for 10 d at 25°C,and hyphal morphology was observed using a microscope (Nikon eclipse,E-200,Japan) at 10× and 40× magnifications.The methods for scanning electron microscopy (SEM)and transmission electron microscopy (TEM) were those presented in our previous report (Gonget al.2017). Each of the abovementioned experiments was performed three times.

2.5.Pathogenicity assay

Mycelium suspensions of WT and KOs were prepared using a glass-glass homogenizer. Tween 20 was added to suspension to a final concentration of 0.1%. Approximately 5 mL of each suspension was inoculated into 4-wk old corn leaves (B73). The seedlings were incubated in moist chambers for 24 h and then transferred to a growth chamber,where they were grown at 25°C with a 12 h photoperiod using fluorescent lamps.

2.6. Appressorium formation and penetration assay

To imitate pathogen penetration under indoor culture conditions,WT and KOs were cultured on PDA medium covered with artificial cellophane that had been smoothed out on the medium and then incubated for 6 d at 25°C. The artificial cellophane was then removed,and after incubation for another 7 d at 25°C,the strains that had been able to penetrate the cellophane were observable. Furthermore,appressorium development was observed according to the following protocol. A small amount of mycelium was suspended in 0.5 mL of PD medium and placed onto sterile glass slides. The slides were incubated at 25°C in dishes lined with moist Whatman 3 paper and then covered and sealed with Parafilm. Appressoria were observed and photographed using phase-contrast optics after 12,24,48,72 and 96 h post-incubation (hpi). A microscope (Nikon eclipse E-200,Japan) equipped with both 10× and 40×magnifications was employed. These experiments were repeated three times with each strain to ensure reliability.

2.7. HT-toxin isolation and toxicity detection

HT-toxin was isolated according to Liet al.(2016) with minor modifications. HT-toxins from WT and KOs were dissolved in 1 mL of methanol,and their toxicity was determined.Briefly,20 μL of HT-toxin extract was dropped onto 4-wk old punctured leaves of B73 and incubated at 25°C in dishes with moist Whatman 3 paper,with methanol used as the mock control. Excised corn leaves of B73 were incubated at 25°C and observed at 48 hpi,and lesions were characterized as previously described (Liet al.2016).

2.8. Detection of sensitivity to H2O2 and ultraviolet radiation

To assess whetherStKU80is involved in responses to oxidative stress and UV radiation,WT and KOs were treated with H2O2and ultraviolet radiation,respectively. The mycelia of WT and KOs were obtained after they were cultured on PDA medium for 10 d. The treatment conditions were as follows. The mycelia were illuminated with UV light for 10,20 and 30 min,respectively,and cultured in the darkness for 10 d. Strains were cultured on PDA medium containing 5 or 10 mmol L-1H2O2. The growth ratio of WT and KO colonies were measured from the 3rd d to the 7th d after the hyphal blocks had been cultured on medium. The inhibition ratio of the strain growth was calculated by the following formula:Inhibition (I%)=(Diameter of fungal colony in control-Diameter of fungal colony in treatment)/Diameter of colony in fungal colony in control×100. All measurements were repeated three times.

3.Results

3.1. Characterization of StKU80

A gene designatedStKU80was cloned fromE.turcicumin the present study. The 2 439 bp sequenceofStKU80genomic DNA of our study was same with the published genomic sequence of St28A strain (Ohmet al.2012).Sequence analysis revealed that the gene included a 2 199 bp open reading frame that is interrupted by a 71-bp intron and encodes a putative protein with 732 amino acids. Conserved domain analysis showed that the protein had three domains:a VWA domain,which is located between the 5th and the 222nd amino acid; a Ku78 domain,which is located between the 320th and the 456th amino acid; and a Ku-PK-bind domain,which is located between the 611th and the 731st amino acid. The VWA motif served as the phosphorylation site of DNA-PKcs (Chanet al.1999),and the Ku-PK-bind domain was required for the binding of Ku and DNA-PKcs proteins (Fig.1-A).

Fig.1 Conserved domain prediction and phylogenetic analysis of StKu80. A,the model of StKu80 showing the conserved domains VWA,Ku,and Ku-PK-bind. B,phylogenetic tree showing evolutionary relationships of fungal Ku80 proteins and the Ku70 protein of Saccharomyces cerevisiae,indicating the relationship between StKu80 and other Ku80s.

The evolutionary relationships between 12 proteins from typical fungi were investigated. These proteins Ku70(NP_014011.1) and Ku80 (NP_013824.1) were fromS.cerevisiae,Ku80(AGF90044.1)fromM.ruber,Ku80(CCE30988.1) fromC.purpurea,Ku80(XP_003713796.1)fromM.oryzae,Ku80(XP_956387.3) fromN.crassa,Ku80(AIA21644.1) fromRhodotorula toruloides,Ku80(XP_011391495.1) fromU.maydis,Ku80(XP_018162680.1)fromC.higginsianum,Ku80(XP_001802136.1) fromP.nodorum,Ku80 (XP_001551373.1) fromB.cinereaand StKu80(XP_008022155.1) fromE.turcicum. Phylogenetic analysis revealed that the proteins aligned in this study were classified into two Ku clades:Ku70-homologs and Ku80-homologs; StKu80 was classified as part of the Ku80-homologous subgroup and was most closely related to Ku80 fromP.nodorum,followed by Ku80 fromM.ruber(Fig.1-B).

3.2. Disruption of StKU80 and confirmation of KOs

The gene knockout strategy used in the present study followed that of a previous study with minor changes (Ruiz-Roldánet al.2001). Upstream (StKU80-A) and downstream(StKU80-B) fragments ofStKU80andBAR(phosphinothricin acetyltransferase gene) genes were amplified with corresponding primers. Then,these fragments were linked into multiple endonuclease sites of the vector pBS. The introduction of the resulting vector,containingStKU80-A,StKU80-B andBAR,into protoplasts of WTE.turcicumwas mediated by PEG4000. Fifty-six glufosinate-resistant transformants with steady growth states were screened after three successive transfers on PDA medium supplemented with 0.2‰ (w/v) glufosinate ammonium. Among them,five resistant strains were further confirmed by PCR withBARspecific primers (forward primer,BAR-L; reverse primer,BAR-R) andStKU80upstream-specific primers (forward primer,KU80upL; reverse primer,KU80upR). Remarkably,five KOs displayed consistent results,and the analysis based on two of these strains is presented in this study.A 1 086-bp upstream-specific (data not shown) fragment and a 1 341-bpBAR-specific fragment were amplified in KOs,while no fragments were detected in WT cells(Fig.2-A). Furthermore,RT-PCR analysis revealed thatStKU80swere no longer expressed in the transformants(Fig.2-B). Southern blotting indicated that 3.47-kb bands were detected in the transformants using specific primers(Fig.2-C),while no band was detected in WT cells,indicating thatStKU80was replaced by a single copy ofBARthat was inserted into the genomic DNA ofE.turcicum.The results above validated that twoStKU80knockout mutants,ΔStKU80-1andΔStKU80-2,were obtained in the present study.

3.3. StKU80 affects hyphal morphology and conidiation but not colony growth ratio

To confirm whetherStKU80was involved in regulating the growth and development ofE.turcicum,we compared the morphology of the colonies,hyphae,and conidia in KOs with those of the WT strain. The two KOs displayed identical phenotypes that were significantly different from those of WT. The colony color of the KOs was light gray,while that of WT was dark brown. KOs presented poorly developed,flat hyphae,while WT exhibited well-developed,long aerial hyphae across the whole colony. In contrast,statistical data showed that the KO,and WT colony diameters were similar under the same culture conditions (t-test,P＞0.05)(Fig.3-A). Optical microscope observations demonstrated that hyphal diaphragms of KOs cells were obviously longer than those of WT and that the colors of KO cells were lighter (Fig.3-C). Moreover,statistical data further proved that significant differences existed between the diaphragm lengths of KO and WT hyphal cells (t-test,P＜0.05),while no significant differences were found in hyphal cell width (t-test,P＞0.05) (Fig.3-B). Additionally,the WT conidium yield was 1.41×106per Petri dish,while no conidia were detected in KOs although they were cultured on LCA medium,which is commonly used for conidium induction. Based on these observations,we concluded thatStKU80affected hyphal morphology as well as conidiation inE.turcicumbut not colony growth ratio.

Fig.2 Identification of StKU80 knockout mutants (KOs). A,Exserohilum turcicum wild type (WT) and KOs,identified by amplifying the gene fragment including the glufosinate resistance gene (BAR) and StKU80 upstream flanking fragment using specific primers(KU80upL/KU80upR). B,RT-PCR analysis of StKU80 expression levels in WT and KOs with primers (KU80-TL/KU80-TR) and Exserohilum turcicum-18S as the reference. C,identifying the number of BAR copies in KOs by Southern blotting.

Fig.3 Growth and development of StKU80 knockout mutants (KOs). A,the radial growth ratio of Exserohilum turcicum wild type(WT) and KOs that were cultured on PDA medium for 7 d at 25°C in the dark. B,determining diaphragmatic length and liameter of hyphae. C,mycelium morphology of WT and KOs that were cultured on PDA medium for 7 d at 25°C in the dark. Vertical bars represent SD (n=3). An asterisk (＊) indicates significant difference (P＜0.05,t-test).

3.4. StKU80 regulates cell wall development

We wondered whetherStKU80affected cell wall development since the mycelium color of KOs observed under an optical microscope was lighter than that of WT. Thus,cell ultra-structures of KOs and WT were observed with SEM and TEM. The SEM showed no obvious differences between the cell surfaces of KOs and WT strains,while the hypha of KOs appeared slighter bending compared to that of WT(data not shown). The TEM indicated that less floccus from the cell surface,and many more mitochondria were present in KOs than in the WT strain (Fig.4). The results suggested thatStKU80was involved in cell development,particularly cell wall development,inE.turcicum.

3.5.StKU80 is necessary for appressorium formation but not for HT-toxin virulence

In general,that ability of appressorium to penetrate is the most important factor for successful infection by pathogens.Therefore,an appressorium development assay was performed on WT and KOs. WT mycelia formed small appressoria at 12 hpi,formed a large number of mature appressoria at 24 hpi,penetrated hydrophobic surfaces and developed infecting hyphae at 48 hpi,and infecting hyphae thrived on PDA medium at 60 hpi,while no appressoria were detected in KOs under the same conditions,even at 60 hpi(Fig.5-A). These results suggest thatStKU80controls appressorium development inE.turcicum.

On the other hand,HT-toxin virulence is another crucial pathogenic factor affecting infection byE.turcicum.Consequently,the HT-toxin virulence of the KOs and WT strains were assessed. The color of the exact from the KOs was much lighter than that from WT (Fig.5-B). Further,when the HT-toxins were inoculated on the surface of the susceptible corn leaves,lesions of KOs and WT were similar in size and color at 2 d post-inoculation on detached corn leaves; by contrast,no lesion was observed in the methanol control group under the same conditions (Fig.5-C).These results showed thatStKU80was not required for the virulence of the HT-toxin ofE.turcicum. In summary,these results demonstrated thatStKU80was significant for appressorium development but not for the virulence of HT-toxin inE.turcicum.

3.6.Susceptibility of the StKU80 knockout mutants to stresses

In general,reactive oxygen species (ROS) could damage pathogenic cellular proteins,lipids and DNA. Therefore,further analysis focused on whetherStKU80affected oxidative stress reactions. TheStKU80knockout mutants,ΔStKU80-1andΔStKU80-2,were cultured on PDA medium containing 5 and 10 mmol L-1H2O2(Fig.6-A). Colony diameters were measured from the 2nd d after the strains were cultured on the media,and the growth inhibition ratio of the KOs and WT strains were compared. The growth state of the KOs was significantly inhibited compared to the WT under both 5 and 10 mmol L-1H2O2(t-test,P＜0.05,Fig.6-B).The results indicated thatStKU80positively regulated the resistance of oxidative stress inE.turcicum.

Fig.4 Transmission electron microscope observation of cell wall of Exserohilum turcicum wild type (WT) and StKU80 knockout mutants (KOs) (scale bar,500 nm). mit,mitochondrion; CW,cell wall.

Fig.5 Observation of appressorium development and determining HT-toxin virulence of StKU80 knockout mutants (KOs).A,formation and penetration of appressorium in KOs cultured on PDA medium,examined at 12,24,48 and 60 h by microscopy(40×). B,the extract of HT-toxin from Exserohilum turcicum wild-type (WT) and KOs dissolved in methanol. C,the toxins from WT and KOs caused similar necrotic lesions on excised leaves,while no necrotic lesions were produced by the control.

Further analysis focused on whetherStKU80affected the response to ultraviolet light.ΔStKU80-1andΔStKU80-2were cultured on PDA and illuminated with 180 μW cm-2for 10,20,and 30 min,respectively. Colony diameters were measured from the second day,and the growth inhibition ratio of the KOs and WT were compared. No significant difference was found between the growth inhibition ratio of WT and KOs (Fig.6-C). The results indicated thatStKU80was not necessary for ultraviolet light stress responses inE.turcicum.

4.Discussion

Fig.6 Sensitivity of StKU80 knockout mutants (KOs) to oxidative and ultraviolet (UV) light stresses. A,colony status of Exserohilum turcicum wild-type (WT) and KOs cultured on PDA medium containing 5 and 10 mmol L-1 H2O2. B,the statistics of growth inhibition ratio of WT and KOs cultured on PDA medium containing 5 and 10 mmol L-1 H2O2 for 7 d. C,the statistics of growth inhibition ratio of WT and KOs cultured for 7 d on PDA medium and illuminated with ultraviolet light at an intensity of 180 μW cm-2 for 10,20,and 30 min. Vertical bars represent SD (n=3). An asterisk (＊) indicates significant differences (P＜0.05,t-test).

Ku70/80 heterodimer has crucial functions in eukaryotic cell cycle control,response to DNA damage and stress,senescence,regulation of DNA replication,gene expression,nucleic acid metabolism,telomere structure maintenance,and cell proliferation (Downs and Jackson 2004; Fisher and Zakian 2005; Fell and Schild-Poulter 2015; De Sena-Tomáset al.2015). The relevant functions are often neglected with the extended employment of Ku70/80 mutants in fungal genetics. In the present study,we obtained twoE.turcicum ΔStKU80mutants,which were with good consistency in the phenotype and pathogenicity. We foundE.turcicumΔStKU80mutants displayed longer septa,reduced substances on cell wall surface,increased mitochondrias compared to WT strain. Moreover,the mutants did not produce conidium and mature appressorium. To our knowledge,very few previous studies report similar conclusions in filamentous fungi. Wanget al.(2011) demonstrated thatΔKU70strains inAlternaria alternatareduced the yield of conidium and the production of pigments. Nakazawaet al.(2011) reported that deletion ofKU70affected fruiting body development in the agaricomyceteCoprinopsis cinerea. It had been reported that the inactivation ofKU70strains inPenicillium digitatumappeared very similar growth,conidium production,and pigment content as WT at the optimal temperature; however,the phenotype of theKU70mutants changed as the temperature increased (Gandíaet al.2016).TheP.digitatumΔKU80andΔKU70strains andA.alternate ΔKU70strains were not affected in their pathogenicity to citrus (Xuet al.2014). InE.turcicum,although theStKU80could influence the synthesis of the melanin related substances,the HT-toxin activity was not changed(Fig.5-C). The same results were also found in theStRas2andStPKA-cmutants ofE.turcicum,which influenced melanin synthesis,but not HT-toxin activity (Shenet al.2013;Zhanget al.2013). These results indicate that melanin is inessential for the HT-toxin activity inE.turcicum. However,we found that theStKU80KOs lost their pathogenicity to susceptible maize leaves (Appendix A). One explanation for the above results is that the pathogen cannot form normal appressorium in the mesophyll cell during infection,meaning that the normal infection strategy ofE.turcicumwhich assessed by the latest histological research can’t take place in maize leaves (Kotzeet al.2019),hence it cannot form the chlorotic spots on infected maize leaves. Notably,Hoffet al.(2010) reported thatP.chrysogenumΔKU70strains were insensitive to oxidative stress,whileE.turcicumΔStKU80strains exhibited higher sensitivity to oxidative stress. All above observations verify crucial functions of the Ku70/Ku80 complex in fungal development,maintenance of genomic integrity,pathogenicity,and response to oxidative stress,which may present differences owing to the complexity of the different fungal genomes,and whose exact mechanism will need to be further studied.

Most reports of fungi indicated that compared to HR,the predominant DSB repair pathway was NHEJ,which was considered to be the main reason for low gene knockout efficiency in fungi (Krappmann 2007; Kück and Hoff 2010). Accordingly,an effective way to improve gene knockout efficiency is increasing the efficiency of HR by inhibiting or removing the NHEJ pathway. Thus,exogenous DNA can be easily integrated into the genome (Weldet al.2006; Krappmann 2007). Recent reports showed that HR efficiency was significantly increased by knocking out the components of the NHEJ pathway (Kück and Hoff 2010).For example,Cnd15,a gene encoding the sesquiterpene cyclase Cnd15 inB.cinerea,was difficult to be knocked out of the WT organism,while the gene editing efficiency ofCnd15was elevated inKU70-orKU80-knockout mutants (Choqueret al.2008). InM.grisea,theMgADE4-targeted gene replacement frequency was greater inKU80-deleted mutants than the WT (80%vs.5%),and high frequencies (＞80%) were observed at seven other loci. The average gene replacement frequency was only 7% in the WT strain (Villalbaet al.2008). In the present research,E.turcicumStKU80deletion mutants were constructed and displayed completely different phenotypes from WT in mycelial morphogenesis,conidiation,and appressorium development,indicating thatStKU80deletion mutants were not suited as recipient strains in gene editing research.

5.Conclusion

StKU80is involved in mycelial morphogenesis,conidiation,and appressorium development but does not affect the virulence of the HT-toxin ofE.turcicum. As a component of the NHEJ pathway,StKU80not only takes part in the DSB repair,but also regulates mycelial morphogenesis,conidiation,appressorium development,and affects oxidative stress resistance and ultraviolet light stress.Further studies should focus on analyzing the mechanisms by whichStKU80regulates growth,development and pathogenicity. Importantly,theStKU80gene deletion mutant has been newly suggested to not be used as an appropriate receptor inE.turcicumgene function research.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31701741 and 31671983) and the Natural Science Foundation of Hebei Province,China(C2016204164 and C2019204211).

Appendixassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm