A Structure-based Computational Study On Proteins Involved In Interaction Of Rice And Magnaporthe Grisea

This study demonstrated the possibility of a computational structure-based screening method on the whole proteome of Oryza sativa and Magnaporthe grisea. A total of 66710 proteins sequences from TIGR Rice v5 and 11054 protein sequences from Broad M. grisea v6 were used. We performed analysis of these protein sequences by using a variety of computational methods. This study not only provided information on the pathogen-related gene products and associated pathways in rice and its pathogen M. grisea, but also a relatively knowledge less-dependent method to screen potential targets in plant-pathogen recognition.Objective of the study was to propose a method to pick out those gene products that may included in the cell surface recognition without any preliminary professional knowledge about the two species and the detailed invading and resisting process. Preferred method was to filter the gene encoding products from the two species by their function category in Gene Ontology; To predict the three-dimensional structure of screened proteins originated from the two species and perform double-blind docking test by one from one organism and the other from another. Some universal signaling pathways are convinced to be participated during appressorium formation, such as the cAMP and MAPK signaling pathways. Verification of the screened out gene products in the two species that may participate in the host-guest recognition was played by searching other researchers' reports in this field, which might be sequenced genes or segments, microarray assays, yeast two-hybrid systems.Only 18 rice proteins vs. 66710 in the proteome and 24 M. grisea proteins of 11054 in the proteome finally passed the strict cellular location screening, protein three dimensional structure prediction, double-blind protein docking test, manual review of the results steps. 30 pairs of interactions between them were screened. These proteins were thought to be associated with several important cell signaling or metabolic pathways during the cell surface recognition and extracellular glycocalyx destruction. Detailed results were as follows:GO term assignment of protein-coding gene productsWithin the rice proteome, 39587 (P<0.05) gene products of longer than 100 amino acids were assigned at least one GO term id. These coding gene products comprised 59.34% of the whole. For comparison, in the model plant Arabidopsis thaliana, the percentage had increased to 98.53% (27445). In M. grisea, 5545 gene products were assigned at least one GO term id, comprised 49.21% of the whole proteome. In Ustilago maydis, also a phytopathogen, this rate rose up to 59.16%, 3874 proteins were assigned at least one GO item id. And in the widely studied eukaryotic model organism, Saccharomyces cerevisiae, 5879 proteins had their homologues in the GO database, comprised nearly 100% of all the known gene products in the species.Cellular component of rice and M. grisea proteinsThe 'slim' term mapping analysis of the cellular components was done for the filtered 39587 rice proteins and 5545 M. grisea proteins. About 3.82% (1511, no duplicates) of the rice gene products classified in the previous analysis were termed proteins with an association to plasmic membrane and extracellular matrix due to their cellular location (GO:0005576, GO:0005578, GO:0005615, GO:0005618, GO:0005886, GO:0030312, GO:0030313). In M. grisea, that percentage was nearly 8.60% (477). We focused here on cellular component rather than the frequently used category molecular function in order to pick out those proteins may function in the foremost cell surface attachment and recognition between the pathogen and host cells and thus triggering intracellular signal transduction in the two different species' cells. These predictions were based on similarity to sequences of known functions in the GO database.Protein modelling and double-blind protein docking testRather few proteins, which thought to be associated with extracellular matrix and cell membrane and cytoskeleton, had finally passed the protein modelling and double-blind protein docking test steps. For the convenience of description, these proteins were divided into three groups: enzymes located in the extracellular space, upstream signal transduction components located in the plasmic membrane, structure components of cytoskeleton and extracellular matrix skeleton.Several already reported enzymes were found. Many isoforms of XYL-6 were found in the M. grisea proteome (MGG06593, MGG₀7955, MGG₀8401, MGG₀8424, and MGG₁5430). XYL-6 is an endo-β-1,4-D-xylanase (EC 3.2.1.8) and is thought to have the ability to degrade the linear polysaccharideβ-1,4-xylan into xylose. And thus breaking down hemicellulose, which is a major component of the cell wall of plants,Chitin synthesis and chitin degradation play an important role in cellular morphogenesis and influence the cell shape of fungal organisms. Chitinase (EC 3.2.1.14) is an important enzyme for the remodelling of chitin in the cell wall of fungi. This is of extremely importantance in the hyphal growth. Two M. grisea chitinase, MGG₀3599 and MGG₁0333, were also included in the test-passed proteins list.Several upstream signal transduction components and plasmic membrane transporters were also presented in the screened results. Mg-NCS-1, a member of the calmodulin (CaM) family, had its homologue of MGG₀1550 in the M. grisea proteome. The latter was also presented in the list of proteins that passed the strict protein modelling and docking tests. MGG₀1550 was also annotated with the ability to participate in the cAMP signaling.The best described function of the clathrin adaptor complex-1 (AP-1) is to participate in the budding of clathrin-coated vesicles, the classical endocytic pathway of internalization of molecules from the cell surface into intracellular membrane compartments. The internalization occurs in both a constitutive and signal-regulated manner. An M. grisea loci, MGG₁1227, annotated with AP-1 complex subunit sigma-1, shown 58% of amino acid percent identity to a known protein SPAP27G11.06c in Schizosaccharomyces pom.be, with the e-value 3.67e-49.A group of microfilament and microtubule cytoskeleton and related proteins were screened both in M. grisea and rice proteomes (MGG₀3164, MGG₀6650, MGG₀6884, MGG₁1167, LOC_Os02g14000, LOC_Os03g63690, and LOC_Os10g43040). In filamentous fungi, the actin cytoskeleton is required for polarity establishment and maintenance at hyphal tips and for formation of a contractile ring at sites of septation.Pathway searchTwo loci coded products had met the build-in criteria in ptools. The rice loci LOC_Os07g 10600 coded product with a putativecycloartenol-C-24-methyltransferase activity was found in Gramene RiceCyc 2.0.1. This protein was thought to be able to catalyze 4 reactions in 3 different pathways, sterol biosynthesis, cyclopropane and cyclopropene fatty acid biosynthesis, cyclopropane fatty acid (CFA) biosynthesis. Another rice loci, LOC_Os11g37890, coding a putative GDP-mannose 3,5-epimerase (1.1.1.145). This enzyme was thought to be involved in the brassinosteroid biosynthesis II pathway.Two Gallus gallus calmodulin P62149 homologues were found in M.grisea proteome (MGG₀6884, MGG₁1167).To understand the internal mechanism that may play key roles in the initial step of plant-pathogen interaction, we have studied on the relationship between rice and the rice blast pathogen M. grisea by using a structure based screening method to pick out those proteins that may function in the cell surface recognition, then conducting a literature search of the metabolic pathways or cell signaling that may associate with these screened proteins. The structure based method would have an obvious advantage: detailed information about the function of the object protein is not necessary.Template update in protein modelling is the major problem in use PROSPECT II and many other non-commercial predicting programs. Pathway predicting using unannotated genomic/proteomic data is the trend in systems biology, whatever as a study field or a paradigm. Lack of reliable experimental data and structured describe vocabulary limit the use of pathway predicting to the massive automated data analysis pipeline.