Studies On Molecular Mechanisms Of Host Cuticle Infection By Phytopathogenic And Entomopathogenic Fungus

Rice blast fungus Magnaporthe oryzae poses great threat to cereal productionworldwide. Due to its socioeconomic importance and genetic tractability, M. oryzae isconsidered a model species for investigating fungal pathogenic mechanisms.Metarhizium anisopliae is a model fungus for studying the fungal entomopathogenesis,and has been widely used as biological agent for controling agricultural pests. Agrowing body of evidence shows that there are many similarities regarding the hostcuticle infection processes between the two different types of pathogenic fungi.Therefore our understanding and comparsion of the molecular pathogenic mechanismsbetween M. oryzae and M. anisopliae, is of great value in the plant protection field.In the course of plant cuticle infection by M. oryzae, macro-autophage playsessential functions in the infection structure development by recycling the conidialnutrients and organelles to support the appressorium maturation and turgor generation.It is thus necessary to examine the genetic control of macro-autophagy in M. oryzae forimproving our understanding of molecular fungal pathogenic mechanisms, and also fordeveloping novel biocontrol strategy for this important cereal killer. M. acridumdevelops the same infection structure called appressorium as M. oryzae to breach thehost insect cuticle. Therefore, large scale analysis of genes expressed by M. acridumduring appressorium formation and differentiation, coupled with the comparison withthat expressed by phytopathogenic fungi during plant cuticle infection, is important forelucidating the molecular pathogenic mechanisms of entomopathogenic fungi and isalso important for identifying novel pathogenicity-associated genes.In this paper, we performed gene functional analysis of MoVAC8, MoTSC13,MoTOR2and MoATG1in the rice blast fungus M. oryzae. In addition, we conductedlarge scale gene expression analysis of M. acridum during appressorium formation anddifferentiation on its susceptible host locust cuticle, and also compared these genes withthat expressed by phytopathogenic fungi. The main results that were obtained from ourwork are listed as follows:1）MoVAC8protein of M. oryzae is composed of a N terminal SH4domain andnine armadillo repeats, and its SH4domain contains4fatty acyl modification sites.MoTSC13gene encodes an enzyme important for biosynthesizing long-chain fatty acidby catalyzing carbon chain extension reaction. Epifluorescence microscopic analysis showed that MoVAC8:GFP fusion proteins were localized on the vacuolar membranesurface, while MoTSC13:GFP fusion proteins distributed on the peri-nuclear membraneand the endoplasmic reticulum membrane. When ecotopically expressed in a yeastVAC8gene deletion mutant, cDNA of M. oryzae MoVAC8could restore the ability ofyeast mutants to grow in the presence of caffeine, but could not restore the morphologyand inheritance of vacuole in the mutant. When ecotopically expressed in a yeast TSC13temperature senstive mutant, cDNA of M. oryzae MoTSC13gene restored the ability ofmutant to grow under restricted temperature conditions, suggesting the functionalconservation of TSC13between M. oryzae and yeast.2）To investigate whether MoVAC8protein contains a functional SH4domain, weused the first21amino acids of MoVAC8to generate a SH4domain:GFP fusion protein.The SH4domain:GFP fusion proteins in conidia, appressoria, invasive hyphae andvegetative hyphae were associated with plasma membrane and vacuolar membrane, asshown by epifluorescence microscopy. Site-directed mutagenesis of the myristoylationand palmitoylation modification sites within MoVAC8SH4domain showed that thesefatty acyl modification sites are both necessary for the proper localization and functionsof MoVAC8protein in M. oryzae.3）M. oryzae gene deletion mutants⊿Movac8and⊿Motsc13were generated byspilit marker targeted gene replacement method. The⊿Movac8mutant showedincreased sensitivity to caffeine. Compared to wild type strain Guy11, the⊿Motsc13mutant was reduced in mycelial growth and sporulation, and the mutant showedincreased sensitivity to osmotic stress and cell wall disturbing agents. Conidia of the⊿Motsc13mutant was still able to elaborate normal appressorium and accumulate turgorpressure within appressorium, but the mutant was severely impaired in penetration pegformation, therefore leading to a reduction of virulence in the mutant.4）Red fluorescent protein RFP was used to label nuclear protein histone1for invivo observation of the nuclei in M. oryzae by epifluorescence microscope. Live cellimaging showed that nuclei in the⊿Moatg1and⊿Moatg4were misshapen and failedto degenerate even after24h post inocualtion during appressorium development,suggesting the essential function of macro-autophagy for infection-associated nucleardegenetration. By contrast, nuclei showed the same behaviour between⊿Movac8,⊿Motsc13and Guy11, suggesting that nuclear degeneration occurs independently of aPMN pathway in M. oryzae because nuclear degeneration was unaffected in eithermutant. 5）MoTOR2full length gene bioinformatically predicted in M. oryzae genomedatabase was cloned by PCR. Sequencing of the genomic fragment confirmed the closerelationship of MoTOR2amino acid sequences with those of other TOR proteins. Thededuced MoTOR2protein is composed of2,469-amino acid residues with a predictedmolecular mass of278kDa. The MoTOR2protein showed a conserved domainarchitecture common to all eukaryotic TOR proteins, and different domains betweenTOR proteins also exhibited high degree of similarity. The N-terminal region of theMoTOR protein contains tandem HEAT repeats and a FAT domain spanning residues1,257-1,870. Located at the C terminal region is the FRB domain, catalytic kinasedomain and FATC domain, which corresponds to residues1,915-2,004,2,043-2,321,and2,437-2,469, respectively. GFP promoter fusion assay showed that MoTOR2isabudantly and constitutively expressed in mycelia, conidia and appressorium in M.oryzae. While qRT-PCR analysis showed that in the course of appressoriumdevelopment, the expression of MoTOR2tended to increase after inoculation, andapproximate two fold increases were detected at14h post inoculation compared to thatin conidia.6）Rapamycin, a well known inhibitor of TOR kinase, exerts pleiotropic effects ondevelopment of M. oryzae, including vegetative growth, appressoriumdevelopment-associated nutrients mobilization and cell cycle progression, appressriumformation and infection-associated macro-autophagy. MoFKBP12gene deletion mutant⊿Mofkbp12lost sensitivity to rapamycin sensitivity, and yeast two-hybrid assayrevealed that MoFKBP12interacts with FRB domain of MoTOR2protein in arapamycin-dependent manner, suggesting that rapamycin affects development of M.oryzae by forming a FKBP12-rapamycin-TOR kinase complex.7）To genetically characterize fuctions of MoTOR2, we employed RNAi approachto down-regulate the transcriptional levels of MoTOR2and examined the effects ofaltering MoTOR2expression. FRB and FAT domains within MoTOR2were chosen astargets for the RNAi mediated silencing. The expression levels of MoTOR2wereconfirmed to be down-regulated in both mycelia, conidia and appressorium of RNAitransformants by qRT-PCR. RNA silencing of MoTOR2induced by NaAc prevented thefungal conidial autophagic cell death and thus reduced fungal pathogenicity.Appressorium-specific down-regulation of MoTOR2by RNAi did not affectappressorium formation and maturation, but significantly reduces the pathogenicity ofM. oryzae against host rice. However, apppressorium-specific over-expression of MoTOR2together with its positive regulator MoLST8did not have any effects on eitherinfection structure development or burst of macro-autophagy within appressorium,indicating the involvement of other potential regulatory components functioning in theburst of the appressorium-specific macro-autophagy in M. oryzae.8） Epifluorescence microscopic analysis showed that GFP:MoATG1fusionproteins distributed in the cytoplasm, and accmulated as punctate dots in certain areas.In addition, these punctate dots could partcially co-localized with the mRFP markedautophagosomes, and the changes in number of these GFP:MoATG1punctate dotscorrelated with that of autophagosomes. Site-directed mutagenesis analyses of MoATG1indicated that its kinase activity is essential for both the induction of macro-autophagyand the autophagy-dependent conidial cell death. Collectively, these data demonstratedthe essential roles of MoATG1kinase activity in regulating the infection-associatedconidial cell death in M. oryzae.9）We employed RNA fragmentation buffer to degrade and remove host nuclei inthe locust hind wings, which was later used for inoculating with M. acridum spores. Byusing the the DSN normalization and SMART cDNA synthesis method, we constructeda cDNA library of M. acridum germinating and differentiating on locust hind wings. Atotal of8215library clone inserts were sequenced, resulting in7302high qualityexpressed sequence tag (EST) that were assembled into4739unique ESTs composed of1089contigs and3650singlets.10）Blast analysis of the4739ESTs against the M. acridum genome databaseshowed that4711ESTs were derived from M. acridum, and that many functionallycharacterized virulence genes were present in the EST dataset. By annotating andfunctionally categorizing the4739ESTs, a number of fungal adapatation strageties werefound to be employed by M. acridum for infecting host cuticle, such as degradation ofinsect cuticle by proteases, nutritional acqusition by transporter proteins, carbon andnitrogen metabolisms reprograming, signal transduction pathways for controllinginfection structure development, cell wall synthesis and remodelling, endocytosis andexocytosis-mediated communication with extracelluar environments, cell cycledependent morphogenesis, synthesis of metabolites for defending against environmentaland host derived stresses. By searching for the similarity of the4739ESTs in thepathogen-host interaction (PHI) gene database, we found that the M. acridum geneswith roles in different aspects of fungal adaptation strategies could encode importantpathogenicity/virulence determinants. 11）A total of44genes were randomly selected from the M. acridum EST datasetfor gene expression analysis using semi-quantitative RT-PCR to examine expressionpatterns between fungal growth in minimal medium, on locust hind wings and in hosthemolymph. Each of the44genes was detected as being expressed on host cuticle.Among them,26genes were identified as being expressed in all the conditionsexamined The remaining18genes were expressed in minimal medium and locust hindwings, but not in hemolymph, indicating the association of these genes undernutrient-limiting conditions. The gene expression pattern of these18genes indicatedthat as a response to starvation, environmental stress or physical cues from locustcuticle, M. acridum expresses subsets of genes essential for mediating insect infection.12）PLS1encodes tetraspanin in M. oryzae and is a critical virulence gene thatdetermines the penetration peg formation. The full-length cDNA sequences of PLS1homolog gene encoding M. acridum tetraspanin was identified from the EST dataset.Using colony in situ hybridization, the full length DNA sequences of MaPLS1wasidentified from the genomic DNA library of M. acridum. Protein structure andphylogenetic relationship analysis indicated that MaPLS1belongs to fungal tetraspaninfamily. Southern-blot analysis revealed that MaPLS1exists as a single copy gene in M.acridum.13） M. acridum is phylogenetically closely related to phytopathogens, thefunctional categorization of the4,739unique ESTs of M. acridum and the comparisonwith those expressed by fungal phytopathogens enabled the identification various fungaladaptation strategies and conserved virulence genes that are likely to be employed by M.acridum in adapting to locust cuticle infection.