ZHANG Li-mei,CHEN Shu-ting,Ql Min,CAO Xue-qiLlANG NanLl Qian,TANG WeiLU GuodongZHOU Jie,YU Wen-ying,WANG Zong-hua,ZHENG Hua-kun

1 State Key Laboratory of Ecological Pest Control for Fujian and Taiwan Crops,College of Life Sciences,Fujian Agriculture and Forestry University,Fuzhou 350002,P.R.China

2 Fujian Universities Key Laboratory for Plant-Microbe Interaction,College of Life Sciences,Fujian Agriculture and Forestry University,Fuzhou 350002,P.R.China

3 Meishan Vocational and Technical College,Meishan 620010,P.R.China

4 Institute of Oceanography,Minjiang University,Fuzhou 350108,P.R.China

5 National Engineering Research Center of JUNCAO Technology,College of Life Sciences,Fujian Agriculture and Forestry University,Fuzhou 350002,P.R.China

Abstract Autophagy is responsible for maintaining fundamental cellular homeostasis and is,therefore,essential for diverse development processes.This study reported that PoElp3,the putative catalytic subunit of Elongator complex,is involved in the maintenance of autophagy homeostasis to facilitate asexual development and pathogenicity in the rice blast fungus Pyricularia oryzae.It was found that the ΔPoelp3 strains were defective in vegetative growth,conidiation,stress response,and pathogenicity.The mutants exhibited hyper-activated autophagy in the vegetative hyphae under both nutrient-rich and nutrient-deficient conditions.The hyper-activation of autophagy possibly suppressed the production of vegetative hyphae in the ΔPoelp3 strains.Moreover,the ΔPoelp3 strains were found to be more sensitive to rapamycin during vegetative-and invasive-hyphal growth but have no effect on Target-of-Rapamycin (TOR) signaling inhibition.Taken together,these results demonstrated that PoElp3 is involved in asexual development and pathogenicity by regulating autophagy in the rice blast fungus.

Keywords:elongator,Elp3,autophagy,asexual development,pathogenicity,Pyricularia oryzae

1.lntroduction

Autophagy,a highly conserved catabolic process in eukaryotes,is very important for maintaining fundamental cellular homeostasis by regulating the degradation and recycling of proteins,membranes and organelles in response to different developmental signals or environmental stimuli(Reggiori and Klionsky 2002;Pinan-Lucarreet al.2003;Klionsky 2005).Among the three types of autophagy,chaperone-mediated autophagy (CMA) is selectively responsible for the degradation of cytosolic proteins containing a specific pentapeptide,while macroautophagy and microautophagy are nonselective and could degrade more substrates (Reggiori and Klionsky 2002).InPyrirularia oryzae(syn.Magnaporthe oryzae),causal agent of the devastating blast disease,non-selective autophagic cell death is essential for the establishment of appressorial turgor pressure (Veneault-Fourreyet al.2006;Kershaw and Talbot 2009;Wilson and Talbot 2009),by regulating the degradation and recycling of cytosol and organelle components,such as lipid body (Liuet al.2007).The Target-of-Rapamycin(TOR) signaling could collaborate with the autophagy pathway to fine-tune mitosis during vegetative and invasive development inP.oryzae(Fernandezet al.2014;Marroquin-Guzman and Wilson 2015;Marroquin-Guzmanet al.2017;Heet al.2018;Sunet al.2018).Deciphering the mechanisms underlying autophagy regulation could improve the understanding of the pathogenic mechanism of this notorious blast fungus.

InP.oryzae,several histone acetylation associated proteins have been reported to be involved in autophagy regulation.For example,both Hat1 and Gcn5 are involved in the regulation of the autophagy pathway by targeting different Atg proteins.MoHat1 is required for appressorial development and pathogenicity through acetylation of two of the autophagy-related proteins,Atg3 and Atg9 (Yinet al.2019).Gcn5 can also acetylate the autophagy-related proteins,including Atg7 and Atg8,as well as other proteins,such as pyruvate kinase,and acts as a negative regulator of light-and nitrogen-starvation-induced autophagy (Zhanget al.2017;Lianget al.2018).Whereas Rtt109 is a fungal-specific histone acetyltransferase contributing to the globular modification of histone H3 at Lys56 (Lawrenceet al.2016),and have been reported to be involved in asexual development and pathogenicity by maintaining genome integrity and repairing DNA damage (Kwonet al.2018).

Elp3 is the catalytic subunit of the Elongator complex which contains a Gcn5-related N-terminal histone acetyltransferase domain (Winkleret al.2002).InSaccharomyces cerevisiae,deletion of Elp3 homolog caused a decrease in acetylation of histone H3 and H4,and led to conditional lethality in the absence of histone proteins or Gcn5.Nevertheless,these defects could be partially rescued by compromising the histone deacetylase activities (Wittschiebenet al.2000).Therefore,Elp3 was supposed to promote transcription elongation by acetylating the nucleosomal histone proteins.In the filamentous fungal pathogenFusarium graminearum,the Elp3 homolog has been reported to participate in histone acetylation,asexual development and pathogenicity (Leeet al.2014).Elp3 also possesses a radicalS-adenosylmethionine (SAM)domain (Glatt and Muller 2013),which was later proved to be involved in the modification of wobble uridine residues in eukaryotic cytosolic transfer RNA (tRNAs) (Selvaduraiet al.2014;Daudenet al.2019;Xuet al.2019).Except for the modification of histone and tRNA,Elp3 has been reported to acetylate α-tubulin,bruchpilot,and connexin-43 in animal cells (Creppeet al.2009;Miskiewiczet al.2011;Laguesseet al.2017).Elp3 is also involved in diverse biological processes in different organisms,such as the cell motility and motor-based trafficking in neurons (Creppeet al.2009;Yanget al.2016),tumorigenicity and migration of melanoma cells (Closeet al.2012),transcriptional silencing and DNA repairing inS.cerevisiae(Liet al.2009).Recently,studies in fission yeast have shown that Elp3 could regulate the TORC1/TORC2 signaling pathway,and TORC2 pathway could in turn activate Elongator by down-regulating the Gsk3 (glycogen synthase kinase 3)-dependent inhibitory phosphorylation of Elongator (Candiracciet al.2019),suggestive of a role of Elp3 in autophagy regulation.This study reported that PoElp3,the putative catalytic subunit of Elongator complex,was required for asexual development and pathogenicity inP.oryzaeby regulating autophagy.

2.Materials and methods

2.1.Manipulation of fungal strains

The wild-type and mutant strains were manipulated as described previously,with minor modifications (Zhenget al.2016).The evaluation of the mutants’ conidiation characteristics relative to the wild-type and the complemented strains were assayed on rice bran medium (2% rice bran and 1.5% agar,pH 6.0).For testing the fungal growth in nutrient deprivation conditions,strains were grown on minimal medium (MM) (Fosteret al.2003).Sensitivity assessment studies were performed by culturing the strains on complete medium (CM) containing 0.5 mol L-1NaCl for salt stress assay,0.005% sodium dodecyl sulfate (SDS),200 µg mL-1Congo red (CR),and 50 µg mL-1calcofluor white (CFW) for cell wall stress assay,1 mol L-1sorbitol for osmotic stress assay and 10 mmol L-1H2O2for oxidative stress assay.For rapamycin inhibition assay,the individual strains were grown on MM,MM-N or CM II medium supplemented with or without 1 µg mL-1rapamycin and incubated under 25°C(Heet al.2018).For all the plate growth assays,tiny plugs of each strain with the same size were transferred from CM to appropriate medium plates and photographed at 7 days post-inoculation (dpi).All the assays were repeated three times with at least three replicates in each biological repeat.

2.2.Targeted gene deletion and complementation of PoElp3

To generatePoELP3deletion mutant,this study amplified an 1 180-bp upstream fragment (A) and an 1 101-bp downstream fragment (B) of the targeted open reading frame (ORF) from the genomic DNA of the wild-type strain Guy11 using the primer setsELP3AF&ELP3AR,andELP3BF&ELP3BR,respectively.The purified upstream and downstream fragments were fused to the 5´-terminal (HY) and 3´-terminal(YG) of hygromycin B phosphotransferase cassette (HPH),respectively,to obtain the ‘A-H’ and ‘H-B’ fusion constructs(Appendix A).The ‘A-H’ and ‘H-B’ fusions were transformed into the protoplasts of Guy11 using the PEG-mediated transformation method as described previously (Kershaw and Talbot 2009).Targeted gene deletion mutants were identified by screening putative hygromycin-resistant transformants using PCR-based genotyping with the following primer pairsELP3tF&ELP3tF,andELP3UAF&H853(Appendix B).The positive transformants were further confirmed through Southern blotting.

For complementation of ΔPoelp3strain,a 5.1-kb genomic DNA fragment containing the native promoter,ORF region,and 3´-untranslated region (UTR) ofPoELP3was amplified using primersELP3comslF2andELP3comslR2(Appendix B) and cloned into pKNTG plasmid carrying the neomycin resistance cassette.The resulting construct was re-introduced into the protoplasts of the ΔPoelp3strain through PEG-mediated transformation.The neomycinresistant transformants were evaluated through PCR-based genotyping and quantitative real-time PCR (qRT-PCR).The primers used for qRT-PCR are listed in Appendix B.

2.3.Southern hybridization assay

This study digested 10 µg of genomic DNA isolated from the individual strains withPstI.The digested products were electrophoresed in 1% (w/v) agarose gel and then transferred onto positively charged Nylon membranes(Roche,Germany).The specific hybridization probe was amplified using primersELP3AFandELP3AR.According to the instruction manual,probe labeling,hybridization,and detection were performed using DIG High Prime DNA Labeling and Detection Starter Kit I (Roche).

2.4.Western hybridization assay

The Western blotting assays were performed as described previously (Zhenget al.2018).The mycelial pellets were ground into fine powder in liquid nitrogen and resuspended in 1 mL extraction buffer (10 mmol L-1Tris-HCl pH 7.5,150 mmol L-1NaCl,0.5 mmol L-1EDTA,0.5% NP-40,2 mmol L-1PMSF).The protein extracts were resolved in 10% SDS-PAGE gels and then transferred onto a PVDF membrane (Merck Millipore Ltd.,USA) using a Bio-Rad electroblotting apparatus.After the incubation with the antibodies indicated as following,the Western Bright ECL HRP substrate (Advansta,Menlo Park,USA) was used to detect the chemiluminescent signals.

The determination of H3K14 acetylation was performed as described previously (Leeet al.2014).Whole-cell extract from mycelia grown in liquid CM for two days was separated by 10% SDS-PAGE and detected with an anti-H3K14 antibody (07-353;Upstate Biotechnology,Inc.,Lake Placid,NY,USA) and an anti-mouse secondary antibody(Abmart,China).

Determination of the degradation of GFP-MoATG8 fusion was performed as described previously with minor revision(Zhanget al.2017;Yinet al.2019).For determination of starvation-induced autophagy,2-day-old mycelia grown in liquid CM were transferred into liquid MM medium containing 2 mmol L-1PMSF for the induction of autophagy.The treated mycelia were collected at 2 and 5 h after induction of autophagy,and whole-cell extract detection assays were carried out with an anti-GFP antibody (Abmart,China) and an anti-mouse secondary antibody (Abmart,China).

2.5.Physiological assays

After the removal of aerial mycelia,culture blocks (2 mm×2 mm) were cut from the 5-day-old rice bran agar cultures,laid on their sides on the slides and placed into a moist plastic box.Upon the induction of conidiogenesis for 24 h in a chamber with constant-light conditions,the conidial development process was observed using the Olympus BX51 microscope (Liuet al.2010).

To analyze the conidial morphology,conidia of each strain collected from the 10-day-old rice bran agar cultures were counted using a hemocytometer,and were imaged using the Olympus BX51 microscope.To examine the conidial germination and appressorial formation,10 µL of conidial suspension (5×104conidia mL-1) of each strain was dropped on the hydrophobic coverslip (Fisher Scientific,USA),and then incubated in a moist chamber at 28°C.The percentages of conidial germination and appressorial formation were determined by microscopic examination at 4 and 8 h post-inoculation (hpi),respectively.

The standard cytorrhysis assay was performed for the appressorium turgor measurement as described (de Jonget al.1997).In brief,the conidia were incubated on the surface of hydrophobic coverslips (Fisher Scientific) and treated with different concentrations of glycerol (1,2,and 3 mol L-1) at 24 h post-germination,then examined using an Olympus BX51 microscope (Olympus,Japan) 5 min after glycerol treatment.

2.6.Plant infection assay

Conidial suspensions (1×105conidia mL-1in 0.025%Tween-20 solution) of each strain were used for inoculation assay on the 14-day-old rice seedlings of the susceptible line CO39.Rice seedlings were cultivated under the conditions described previously (Giraldoet al.2013).Plant inoculation and subsequent incubation were performed as previously described (Valentet al.1991).Lesion formation was examined at 5-7 dpi,and the lesion number was calculated from 5 cm-long diseased rice blade sections.The pathogenicity assay on intact or wounded barley was performed using mycelial pellets,which were cultured in CM liquid medium for 48 hpi.Lesion formation on leaves was observed at 5 dpi.

2.7.Live-cell imaging assay

For investigation of the subcellular localization of PoElp3,wild-type strain (Guy11) co-expressingGFP-PoElp3andH1-RFPwas used.Conidia inoculated on hydrophobic coverslip (Fisher Scientific) for 0,3 and 8 h,respectively,were imaged using the Nikon A1R laser scanning confocal microscope system (Nikon,Japan).

For imaging of invasive hyphae,the leaf sheath inoculation was performed using the susceptible rice line CO39.For rapamycin treatment assay,conidial suspension with or without 1 µg mL-1rapamycin was dropped onto the leaf sheath of CO39.The invasive hyphae were checked by microscopic examination at 24 hpi.An Olympus BX51 microscope (Olympus,Japan) or Nikon A1R laser scanning confocal microscope system (Nikon,Japan) was used for microscopic examination and live cell imaging was performed as previously reported (Zhenget al.2017).

2.8.Quantitative real-time PCR (qRT-PCR)

For qRT-PCR analysis,first strand cDNA was generated from 1 µg of total RNA using the Prime Script RT Reagent Kit with gDNA Eraser (TaKaRa,Dalian,China) from 1 µg of total RNA.Real-time PCR was performed by using an SYBR Green PCR Master Mix (TaKaRa,Dalian,China).The primers used for qRT-PCR are listed in Appendix B.

2.9.Statistical analysis

The statistical analysis was performed using thet-test in GraphPad Prism 7.0 Software with the setting for 2-samples assuming unequal variances (GraphPad Software,San Diego,California,USA,www.graphpad.com).

3.Results

3.1.ldentification of the Elp3 ortholog in P.oryzae

To identify the Elp3 ortholog fromP.oryzae,the amino acid sequence of Elp3p protein from theS.cerevisiaegenome database (http://www.yeastgenome.org/) was used for a BLAST search in the National Center for Biotechnology Information (https://blast.ncbi.nlm.nih.gov/Blast.cgi).The search identified MGG_05481.7,termed PoElp3,as the ortholog ofS.cerevisiaeElp3p.The PoElp3 protein shares high amino acid sequence identity with its orthologs in other eukaryotic organisms,such as NcElp3 inNeurospora crassa(XP_961595.1,91% identity),FgElp3 inF.graminearum(FGSG_02040.3,89% identity),BcElp3 inBotrytis cinerea(XP_001555701.1,86% identity),Elp3p inS.cerevisiae(EGA84564.1,77% identity),OsElp3 inOryza sativa(XP_015635801.1,73% identity) and hElp3 inHomo sapiens(NP_060561.3,73% identity).Multiple sequence alignment demonstrated that all these proteins contain two conserved domains:the radicalS-adenosylmethionine(SAM) domain and the histone acetyltransferase (HAT)domain characterized by the conserved A-,B-and D-motifs(Appendix C).

3.2.Defects in fungal growth and conidiation caused by deletion of PoELP3

To determine the function of PoElp3 inP.oryzae,this study generated gene targeted mutants using the split marker approach (Appendix B) (Kershaw and Talbot 2009).The ΔPoelp3candidates were evaluated through PCRbased genotyping,and three mutant strains were further characterized by Southern blotting analysis (Appendix C).For complementation assay,the entire ORF with its native promoter and the downstream sequence was introduced into two strains,ΔPoelp3-120and ΔPoelp3-167.The transformants were further confirmed by PCR-based genotyping (Appendix B).Since the two independentPoELP3defective strains displayed similar phenotypic characteristics,the ΔPoelp3-167was used for subsequent studies (herein referred to as ΔPoelp3).

To explore the role of PoElp3 in vegetative growth,this study cultured the wild-type (Guy11),ΔPoelp3,and complemented strain (ΔPoelp3c) on CM and MM media.The radial growth displayed by the ΔPoelp3strain reduced significantly on CM and MM media (Fig.1-A and B).The ΔPoelp3strain also exhibited a significant reduction in conidiation.The average number of conidia produced by the ΔPoelp3strain was less than half of that of the wildtype and complemented strain (Fig.1-C).Consistently,it was also observed that each conidiophore of the ΔPoelp3strain produced fewer conidia than that of the Guy11 and ΔPoelp3cstrains (Fig.1-D).These results suggested that PoElp3 is involved in the vegetative growth and conidiation of the rice blast fungus.

Fig.1 The ΔPoelp3 strain is defective in vegetative development and conidiation.A and B,colony morphology (A) and colony diameters (B) of Guy11,ΔPoelp3 and ΔPoelp3c on complete medium (CM) and minimal medium (MM) for 7 days.C,number of conidia harvested from a 7-cm rice bran plate at 3 days after induction of conidiation.D,conidia development of Guy11,ΔPoelp3 and ΔPoelp3c was observed under microscope at 12 and 24 h after induction of conidiation.Bar=50 µm.Data were shown as the mean±SE (n=3).The asterisks indicate statistically significant differences (＊＊＊,P＜0.001).

3.3.lnvolvement of PoElp3 in response to diverse stressors

Previous studies inS.cerevisiaeand other organisms indicated that Elp3 homologs are associated with stress response (Wittschiebenet al.2000;Hanet al.2007;Nelissenet al.2010;Leeet al.2014).To determine whether PoElp3 is involved in the response to various stress inducers,Guy11,ΔPoelp3and ΔPoelp3cwere incubated on CM media containing 0.5 mol L-1NaCl,0.005% SDS,1 mol L-1sorbitol,10 mmol L-1H2O2and 200 mg mL-1CR or 50 mg mL-1CFW.Results showed that there was a severe inhibition in the growth of the ΔPoelp3strain on culture media supplemented with NaCl,CR,and CFW (Fig.2).These results suggested that PoElp3 is involved in the regulation of salt stresses and the enforcement of cell wall integrity in the rice blast fungus.

Fig.2 PoElp3 is involved in stress response.A,the Guy11,ΔPoelp3 and ΔPoelp3c were inoculated on complete medium (CM)without or with various stress inducers (0.5 mol L-1 NaCl,0.005% SDS,1 mol L-1 sorbitol,10 mmol L-1 H2O2 and 200 mg mL-1 CR or 50 mg mL-1 Calcofluor white) for 7 days at room temperature.B,colony diameters were measured and subjected to statistical analysis.The growth inhibition rate is relative to the growth rate of each untreated control.Inhibition rate (%)=(Diameter of untreated strain-Diameter of treated strain)/(Diameter of untreated strain)×100.Data were shown as the mean±SE (n=3).The asterisks indicate statistically significant differences (＊,P＜0.05;＊＊＊,P＜0.001).

3.4.Requirement of PoElp3 for appressorial development and turgor pressure

For a successful infection,P.oryzaedevelops appressoria generating tremendous turgor pressure to rupture the cuticle layer of rice leaves.This study investigated whether PoElp3 is involved in the appressorial development and the accumulation of appressoria turgor.These assays observed attenuation in the onset of appressorium development in the ΔPoelp3strain (Fig.3-A and B).The appressorial turgor pressure was further assessed in Guy11,ΔPoelp3,and ΔPoelp3cthrough standard cytorrhysis assay.Results showed that,after the treatment of 1,2,or 3 mol L-1glycerol for 5 min,the appressorial collapse rates of the ΔPoelp3strain were higher (60.66,83.51,and 92.95%,respectively)than that of the wild-type (41.62,67.40,and 89.62%,respectively) and complemented strains (51.74,72.21,and 89.11%,respectively) (Fig.3-C).These results suggested that PoElp3 contributed positively to the pathogenic development ofP.oryzae.

Fig.3 The ΔPoelp3 strain is defective in appressorial formation and turgor pressure.A and B,morphology and number of appresorium formed by Guy11,ΔPoelp3 and ΔPoelp3c incubated on hydrophobic coverslips for 4,8 and 12 h,respectively.Bar=10 µm.C,the percent of appressorial collapse after inoculation in different concentrations of glycerol.Data were shown as the mean±SE (n=3).The asterisks indicate statistically significant differences (＊＊,P＜0.01;＊＊＊,P＜0.001).

3.5.Requirement of PoElp3 for full virulence of P.oryzae

To determine the possible involvement of PoElp3 in the pathogenesis ofP.oryzae,mycelial blocks of Guy11,ΔPoelp3and ΔPoelp3cwere inoculated on detached barley leaves.On both intact and wounded barley leaves,the lesions caused by the ΔPoelp3strain were much smaller than those caused by the wild-type and complemented strains (Fig.4-A).To further confirm this result,the conidial suspension of each strain was inoculated onto the 14-dayold seedlings of the susceptible rice cultivar CO39.At 7 dpi,the ΔPoelp3strain produced fewer and smaller lesions than the wild-type and complemented strains(Fig.4-B and C).The ΔPoelp3strain generated mostly type 1 and type 2 lesions,only a few type 3 lesions,and no type 4 and type 5 lesions.In contrast,the wild-type and complemented strains generated much more type 3 to type 5 lesions (Fig.4-D).

Fig.4 PoElp3 is involved in pathogenicity of Pyricularia oryzae.A,intact (left) and wounded (right) barley leaves were inoculated with culture blocks of Guy11,ΔPoelp3 and ΔPoelp3c.B,leaves of rice cultivar CO39 were sprayed with conidial suspensions (5×104 conidia mL-1) of Guy11,ΔPoelp3 and ΔPoelp3c.C,quantification of lesion number in 5 cm-long rice leaves.D,quantification of different lesion types.Data were shown as the mean±SE (n=3).The asterisks indicate statistically significant differences (＊,P＜0.05).

This study also performed leaf sheath inoculation to investigate appressorial penetration rate and invasive growth of Guy11,ΔPoelp3and ΔPoelp3c.At 24 hpi,the penetration rate of the ΔPoelp3strain is lower (73.40%) than that of the wild-type (92.67%) and complemented strains(86.09%).Detailed analysis showed that the ΔPoelp3strain generated mostly type 1 (26.60%) and type 2 (55.30%)invasive hyphae (IHs),a few type 3 IHs (18.10%) and no type 4 IHs.In contrast,the wild-type and complemented strain produced more type 3 (65.21 and 51.23%,respectively) and type 4 (11.96 and 3.33%,respectively) IHs(Appendix D).

Taken together,these results indicated that PoElp3 was involved in the appressorial penetration and invasive growth,and subsequently the pathogenicity ofP.oryzae.

3.6.Subcellular localization of PoElp3

To assess the subcellular distribution of PoElp3 inP.oryzae,GFP-PoELP3fusion construct was transformed into the protoplasts of the ΔPoelp3strain (ΔPoelp3-167) to create the complementation strains.Phenotyping results showed that the expression ofGFP-PoELP3could rescue the defects caused by deletion ofPoELP3(Fig.5-A;Appendix E),which suggested that GFP-PoElp3 is functional.To further visualize whether PoElp3 localizes to the nuclei,H1-RFGandGFP-PoELP3were co-transformed into the protoplasts of the wild-type strain.Microscopy results showed that the green fluorescence was distributed all throughout the conidial cells and co-localized with red fluorescence (H1-RFP) in the nuclei (Fig.5-B).These results suggested that PoElp3 was localized to the protoplasm inP.oryzae.

Fig.5 PoElp3 localized in nuclei and cytoplasm.A,pathogenicity assay of Guy11,ΔPoelp3 and mutant strain ectopically expressing the GFP-PoELP3 fusion (ΔPoelp3/GFP-PoELP3).B,co-expression of GFP-PoElp3 and H1-RFP in Guy11.The images were recorded at 0,3 and 8 h postinoculation (hpi),respectively.Bar=10 µm.

3.7.Autophagy homeostasis regulated by PoElp3 during fungal growth

To further determine the role of PoElp3 in autophagy ofP.oryzae,GFP-PoAtg8,a widely used nonspecific autophagy marker (Denget al.2008),was introduced into the wild-type and ΔPoelp3strains.The autophagy level was evaluated through immunoblot and quantitatively measured by calculating the percentage of free GFP in the total of intact GFP-PoAtg8 and free GFP together.This study at first detected the autophagy level of the wild-type and ΔPoelp3strains under nutrient-free conditions.The results showed hyperactivation of autophagy in the mycelia of the ΔPoelp3strains under both nutrient-rich and nutrientlimited conditions compared with that in the wild-type strain(Fig.6-A).

This study further investigated the autophagic flux in the Guy11 and ΔPoelp3strains expressing the GFPPoAtg8 fusion.Before the starvation treatment (in liquid CM medium),autophagosome formation was relatively slow,and most of the GFP signals remained in the fungal cytoplasm in the wild-type strain.In contrast,microscopy examinations revealed a profound accumulation of GFP signals in the autophagosomes of the ΔPoelp3strain.Upon the starvation induction (in liquid MM-N medium for 2-h),most of the GFP signals enriched in autophagosomes in both the wild-type and ΔPoelp3strains (Fig.6-B and C).These results suggested that autophagy level was hyperactivated in the ΔPoelp3strain.

Fig.6 Autophagy was hyperactivated in vegetative hyphae of ΔPoelp3 strain.A,the autophagy in wild-type and mutant mycelia before (CM) and after the deprive of nutrient (MM-N 2 h and MM-N 5 h).A representative result of three biological replications was shown.The autophagy level was evaluated by calculating the ratio of GFP/(GFP+GFP-PoAtg8).B,autophagosomes in vegetative hyphae of Guy11 and ΔPoelp3 expressing the GFP-PoATG8 fusion.Bar=10 µm.C,percentage of vacuoles carrying GFP signal.Data were shown as the mean±SE (n=3).The asterisks indicate statistically significant differences (＊＊＊,P＜0.01).

3.8.Vegetative growth and infective growth regulated synergistically by PoElp3 and TOR signaling pathway

To assess the impact ofPoELP3gene deletion on the TOR signaling pathway,rapamycin inhibition assays were performed.Results from these examinations showed that the ΔPoelp3strain was more sensitive to rapamycin treatment compared with the wild-type and complemented strains (Fig.7-A and B;Appendix F),indicating a role of PoElp3 in the TOR signaling pathway.Furthermore,the expression pattern ofPoELP3was monitored in the wild-type strain treated with rapamycin.qRT-PCR results revealed that the transcription level ofPoELP3was not significantly affected by the treatment of rapamycin (Fig.7-C).This study then tested the transcription level ofRS2andRS3,the two genes encoding ribosomal protein S2-like and 40S ribosomal protein S3,respectively (Fernandezet al.2014),to assess whether PoElp3 acts downstream of TOR signaling.The qRT-PCR results showed that the abundance ofRS2andRS3transcripts was down-regulated to the same level by rapamycin in wild-type and ΔPoelp3,indicating that PoElp3 was not required for the inhibition of TOR signaling.This study also detected rapamycin’s effect on invasive hyphal growth in the wild-type and ΔPoelp3strains and found that IHs in the ΔPoelp3strain were more sensitive to rapamycin than those in the wild-type strain (Fig.8).Taken together,these results suggested that PoElp3 genetically interacted with TOR signaling in the regulation of both vegetative-and invasive-hyphae.

Fig.7 PoElp3 acts synergistically with Target-of-Rapamycin (TOR) signaling to regulate autophagy.A,colony morphology of Guy11,ΔPoelp3 and ΔPoelp3c grown on minimal medium (MM) treated with or without rapamycin (Rap).B,inhibition rates of Guy11,ΔPoelp3 and ΔPoelp3c grown on MM treated with rapamycin.C,relative transcription level of PoELP3 was not affected upon treatment of 1 µg mL-1 Rap.D,relative transcription level of TOR-regulated genes PoRS2 and PoRS3 in of Guy11 and ΔPoelp3 strain treated with or without 1 µg mL-1 Rap.Data were shown as the mean±SE (n=3).The asterisks indicate statistically significant difference (＊＊,P＜0.01;＊＊＊,P＜0.001).

Fig.8 PoElp3 acts redundantly with Target-of-Rapamycin (TOR) signaling to regulate invasive growth of rice blast fungus.A,morphology of invasive hyphae in Guy11,ΔPoelp3 and ΔPoelp3c at 24 h post-inoculation (hpi).B,statistical analysis of different types of invasive hyphae (IHs) at 24 hpi.Data were shown as the mean±SE (n=3).

4.Discussion

Maintenance of autophagy homeostasis is essential for the development and pathogenicity of the rice blast fungus.This study reported the role of the Elp3 homolog in the regulation of autophagy homeostasis inP.oryzae.Disruption ofPoELP3caused defects in asexual development and pathogenicity.Further studies showed that PoElp3 negatively regulate autophagy during fungal growth and its possible interaction with the TOR signaling cascade to regulate both vegetative and invasive-hyphal development.

Previous studies in several other organisms indicated that Elp3 orthologs are associated with the response to diverse environmental stimuli (Wittschiebenet al.2000;Hanet al.2007;Nelissenet al.2010;Leeet al.2014).Results of this study showed that the ΔPoelp3strain was highly sensitive to the cell wall biosynthesis perturbing agents CFW and CR,suggesting a role of PoElp3 in maintaining cell wall integrity.These observations are consistent with the role reported for Elp3 inS.cerevisiae(Wittschiebenet al.2000).However,the sensitivity to H2O2was not significantly reduced in the ΔPoelp3strain compared with wild-type,which is different from the situation of its homolog inF.graminearum(Leeet al.2014).

The maintenance of autophagy homeostasis is vital for vegetative and infective development.It involves diverse cellular processes,including the inactivation of the TOR signaling pathway and the activation of autophagy processes that facilitate the degradation of cytosolic components of the conidial cells for the biosynthesis of essential osmolytes (e.g.,glycerol) required to generate sustained turgor pressure during appressorial development and host penetration (Kershaw and Talbot 2009;Wilson and Talbot 2009;Marroquin-Guzman and Wilson 2015).Studies have shown that the activation of TOR signaling suppresses autophagy and promotes the progression of mitosis in the hyphal cells during the development of vegetative and invasive hyphae (Heet al.2018;Sunet al.2018).The ΔPoelp3strain showed compromised vegetative growth and pathogenicity.The pathogenicity defects associated with the ΔPoelp3strain could be attributed to penetration and invasive growth defects.Detailed analysis of the infection-related development processes showed that although conidia produced by ΔPoelp3germinated and formed appressorium,there was a delay in the onset of conidia germination.This study also revealed attenuation in the appressorial turgor pressure generated by the ΔPoelp3strain compared to the wild-type and complemented strains,which may partially account for pathogenicity defects associated with the ΔPoelp3strain.These results suggested that the reductions recorded in the vegetative and invasive growth of the ΔPoelp3strain were likely a consequence of hyperactivation of autophagy in the hyphae.Furthermore,this study demonstrated that PoElp3 genetically interacted with the TOR signaling,but was immune from the inhibition effects of rapamycin on the TOR signaling pathway.InS.cerevisiae,the Elongator mutants are also more sensitive to rapamycin than the wild-type,which could be caused by the decreased activity of TORC1 and the failure in the activation of TORC2 (Candiracciet al.2019).However,there is only one TORC gene inP.oryzae,and whether the elongator-dependent tRNA modification and TOR signaling are reciprocally regulated inP.oryzaeas their counterparts do inS.cerevisiaeremains to be explored.Taken together,these results suggested that PoElp3 was involved in both the fungal growth and pathogenicity by regulating the autophagy pathway.

Previous reports suggested that Elp3 plays diverse functions in eukaryotes and was identified as a noncanonical acetyltransferase that plays a conserved role in modifying tRNA in yeast,plants,and mammals.Meanwhile,the function of Elp3 as HAT remains a controversial subject(Glatt and Muller 2013;Candiracciet al.2019;Daudenet al.2019;Xuet al.2019).In yeast,the Elp3 homolog is localized only in the cytoplasm,and only tRNA but not the lysinecontaining histone could induced its acetyl-CoA hydrolysis activity (Linet al.2019).In contrast,the Elp3 homologs in plant and animal cells are partially nuclear-localized(Hawkeset al.2002;Creppeet al.2009;Qiet al.2019).This study showed that PoElp3 contained all the three conserved HAT domains and was also partially localized in nuclei.Therefore,PoElp3 is likely to possess the acetylation activity (Leeet al.2014;Yinet al.2019).However,unlike its counterpart inF.graminearum,the acetylation level of H3K14 in the ΔPoelp3strain was not significantly reduced(Appendix G).But the possibility still exists that the absence of PoElp3 may cause changes in the profiling of H3K14ac or the overall acetylation level of histone H3 or H4 at other lysine residual(s).Further study is needed to reveal the mechanism connecting PoElp3 to the maintenance of autophagy homeostasis in the rice blast fungus.

5.Conclusion

In summary,results of this study demonstrated that PoElp3 was involved in asexual development,stress response and pathogenicity of the blast fungus by regulating the autophagy signaling.However,how PoElp3 participates in the autophagy pathway remains to be explored.

Acknowledgements

We would like to thank Prof.Zhang Zhengguang at Nanjing Agricultural University,China and Dr.Liu Xiaohong at Zhejiang University,China for kindly providing theGFP-PoATG8plasmid,and Dr.Zheng Wenhui at Fujian Agriculture and Forestry University (FAFU),China for kindly providing theH1-RFPplasmid.We gratefully thank Dr.Norvienyeku Justice at FAFU and Dr.Chen Xiaofeng at Minjiang University,China for editing the manuscript.This work was supported by grants from the National Natural Science Foundation of China (31770156 and 2016YFD0300700) and the Innovative Research Funding of FAFU,China (CXZX2018051 and CXZX2018052).

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